WO2020157798A1 - Lane keeping assist device and lane keeping assist method - Google Patents

Lane keeping assist device and lane keeping assist method Download PDF

Info

Publication number
WO2020157798A1
WO2020157798A1 PCT/JP2019/002737 JP2019002737W WO2020157798A1 WO 2020157798 A1 WO2020157798 A1 WO 2020157798A1 JP 2019002737 W JP2019002737 W JP 2019002737W WO 2020157798 A1 WO2020157798 A1 WO 2020157798A1
Authority
WO
WIPO (PCT)
Prior art keywords
characteristic line
vehicle
road
lane keeping
support
Prior art date
Application number
PCT/JP2019/002737
Other languages
French (fr)
Japanese (ja)
Inventor
宇津井　良彦
Original Assignee
三菱電機株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 三菱電機株式会社 filed Critical 三菱電機株式会社
Priority to PCT/JP2019/002737 priority Critical patent/WO2020157798A1/en
Publication of WO2020157798A1 publication Critical patent/WO2020157798A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/09—Arrangements for giving variable traffic instructions
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems

Definitions

  • the present invention relates to a lane keeping support device and a lane keeping support method.
  • the server device described in Patent Document 1 obtains usage information of a lane keeping function while traveling from a plurality of vehicles having a lane keeping function, and an area where the usage rate of the lane keeping function is low based on the obtained usage information.
  • the area where the usage rate of the lane keeping function is low is an area in which the lane keeping function is intermittently operated because the lane is not recognized temporarily due to fading of the lane. The intermittent operation of the lane keeping function makes the driver uncomfortable.
  • the server device described in Patent Document 1 issues an intermittent operation of the lane keeping function by giving an instruction to stop the lane keeping function to a vehicle traveling in an area where the utilization rate of the lane keeping function is low. Prevent. As a result, the driver can prepare for the stop of the lane keeping function in the area where he/she is going to travel, and when the lane keeping function is stopped, the driver can shift to the manual driving without feeling uncomfortable.
  • Patent Document 1 there is no guarantee that there is another vehicle that travels in advance on the road that the own vehicle normally uses, and the use information of the lane keeping function is not always accumulated for the road that the own vehicle normally uses. There wasn't. Therefore, the invention according to Patent Document 1 has a problem that roads that can prevent intermittent operation of the lane keeping function are limited.
  • the present invention has been made to solve the above problems, and it is an object of the present invention to prevent intermittent operation of the lane keeping function, especially on a road that is regularly used by the host vehicle.
  • the lane keeping assist device assists the own vehicle so as to maintain the running in the lane by using the characteristic line which is detected from the captured image in front of the own vehicle and which defines the lane in which the own vehicle travels.
  • the lane keeping support unit, the feature line accuracy learning unit that learns the accuracy of the feature line detected on the road on which the vehicle is traveling, the accuracy of the feature line that was previously learned by the feature line accuracy learning unit, and the A support availability determination unit that determines availability of support by the lane keeping assistance unit on the road on which the vehicle is traveling is provided by using the number of times of learning performed on the road on which the vehicle is traveling.
  • the accuracy of the characteristic line detected on the road on which the vehicle is traveling is learned, and whether the lane keeping support on the road on which the vehicle is traveling is performed using the accuracy of the learned characteristic line and the number of times of learning. Therefore, it is possible to prevent the intermittent operation of the lane keeping function, especially on the road that the own vehicle normally uses.
  • FIG. 1 is a diagram showing a hardware configuration example of a lane keeping support system according to a first embodiment.
  • FIG. 3 is a diagram showing an example of functional blocks of an ECU (lane keeping assistance device) according to the first embodiment. It is a conceptual diagram regarding the vehicle speed at which the conventional lane keeping assistance device starts lane keeping assistance. It is a flow chart which shows the example of operation of the conventional lane maintenance assistance device. It is a figure explaining the concept of lane detection in the case where the conventional lane maintenance support device carries out lane maintenance assistance.
  • 3 is a flowchart showing an operation example of the ECU according to the first embodiment.
  • FIG. 3 is a diagram showing a road shape on which the host vehicle is traveling in the first embodiment.
  • FIG. 6 is a diagram illustrating a case of a defect in detecting a characteristic line in the image processing apparatus according to the first embodiment.
  • FIG. 3 is a diagram schematically showing a part of the information stored in the map DB according to the first embodiment.
  • 5 is a flowchart showing an operation example by a characteristic line accuracy learning unit of the first embodiment.
  • FIG. 5 is a diagram showing an example of a calculation formula of characteristic line accuracy in the first embodiment.
  • 5 is a diagram showing an example of functional blocks of an ECU according to the second embodiment.
  • FIG. FIG. 9 is a diagram showing an example of a search route searched by the navigation device according to the second embodiment.
  • FIG. 10 is a diagram showing an example of a calculation formula of a route characteristic line accuracy in the second embodiment.
  • FIG. 16 is a diagram showing another example of a calculation formula of route characteristic line accuracy in the second embodiment.
  • 7 is a flowchart showing an operation example of the ECU according to the second embodiment.
  • FIG. 6 is a diagram showing an example of functional blocks of an ECU according to a third embodiment.
  • FIG. 16 is a diagram showing an example of a road shape on which the host vehicle is traveling in the third embodiment. It is a figure explaining the route prediction method by the traveling route prediction part of Embodiment 3, and shows an example of typical driving operation.
  • FIG. 16 is a diagram for explaining a route prediction method by the travel route prediction unit of the third embodiment, showing an example of coefficients for each representative driving operation.
  • FIG. 16 is a diagram showing an example of a calculation formula of a route characteristic line accuracy in the third embodiment.
  • FIG. 7 is a flowchart showing an operation example of the ECU according to the third embodiment.
  • FIG. 8 is a diagram showing an example of functional blocks of an ECU according to the fourth embodiment. It is the figure which represented typically some information memorize
  • FIG. 16 is a diagram showing an example of a calculation formula of route characteristic line accuracy in the fourth embodiment. It is a figure explaining the illuminance estimated by the environment determination part of Embodiment 4.
  • FIG. 16 is a diagram showing an example of a calculation formula of route characteristic line accuracy in the fourth embodiment.
  • FIG. 1A is a diagram illustrating a hardware configuration example of a lane keeping support system 1 according to the first embodiment.
  • the lane keeping assist system 1 is installed in a vehicle and assists the vehicle (hereinafter, referred to as “own vehicle”) to travel in the lane.
  • the electronic control unit 10 (hereinafter referred to as “ECU 10”) is a lane keeping assist device that assists the host vehicle in driving in the lane.
  • FIG. 1B is a diagram showing an example of functional blocks of the ECU 10 according to the first embodiment.
  • the ECU 10 includes a support availability determination unit 11, a lane keeping support unit 12, and a characteristic line accuracy learning unit 13. Details of the ECU 10 will be described later.
  • the camera 20 images the road environment in front of the vehicle.
  • the image processing device 21 processes the captured image input from the camera 20 and detects a characteristic line such as a white line that divides the lane in which the vehicle is traveling.
  • the antenna 31 receives a radio wave from a GPS (Global Positioning Satellite System) satellite.
  • the GPS receiver 30 measures the traveling position of the host vehicle using the radio waves received by the antenna 31.
  • the navigation device 32 includes a map database 33 (hereinafter, referred to as “map DB 33”) that stores various road information.
  • the various types of road information include road information, a characteristic line certainty associated with the road information, and the number of learnings of the characteristic line certainty, which will be described later.
  • the navigation device 32 detects the road on which the vehicle is traveling, using the traveling position measured by the GPS receiver 30 and the information in the map DB 33. In addition, the navigation device 32 uses the information on the travel position measured by the GPS receiver 30 and the map DB 33 to search for the route on which the vehicle is scheduled to travel.
  • the map DB 33 is arranged in the navigation device 32 in FIG. 1A, the map DB 33 may be arranged in the server device outside the vehicle. In this case, for example, the navigation device 32 may access the map DB 33 in
  • the vehicle network ECU 40 communicates with a vehicle network (not shown) that communicates various types of information regarding the vehicle.
  • the various kinds of information regarding the own vehicle include the traveling speed of the own vehicle, the driving operation of the accelerator, the brake, the blinker, and the like, the illuminance around the own vehicle, and the like.
  • the HMI (Human Machine Interface) 50 includes an output device such as a display or a speaker and an input device such as a touch panel, a button or a voice recognition function.
  • the HMI 50 displays or voice outputs the state of the lane keeping support by the ECU 10. Further, the HMI 50 receives inputs of various settings to the ECU 10.
  • the steering wheel 60 is operated by the driver of the vehicle.
  • the steering angle sensor 61 detects the operation angle of the steering wheel 60.
  • the torque sensor 62 detects the operating force of the steering wheel 60.
  • the electric circuit 63 controls the motor 64.
  • the motor 64 applies a steering force to the steering wheel 60.
  • the gear box 65 transmits the steering angle and the steering force generated by the motor 64 to the wheels 66.
  • the wheels 66 face the direction in which the host vehicle travels by receiving the steering force.
  • FIG. 2 is a conceptual diagram regarding the vehicle speed at which the conventional lane keeping support device starts lane keeping support.
  • the horizontal axis of the graph shown in FIG. 2 represents the vehicle speed of the host vehicle, and the vertical axis represents the distribution of the power required for steering and the traveling speed.
  • the power required for steering is the largest when the host vehicle is stopped due to the friction of the front wheels with the road surface, and is a downward-sloping curve that decreases as the host vehicle starts running and the vehicle speed increases. Therefore, the power of the motor 64 or the like required for the assistance by the lane keeping assist device decreases as the vehicle speed increases, and the efficiency in vehicle control improves.
  • the lane keeping assist device to provide assistance at a predetermined vehicle speed or higher in view of the balance between the power required for steering and the traveling speed distribution.
  • the prescribed vehicle speed is often limited to the minimum speed or higher on a motorway or expressway, etc., where the road environment is maintained, it is easy to detect the driving lane, and there is little concern that the assisting operation will become unstable. ..
  • FIG. 3 is a flowchart showing an operation example of a conventional lane keeping assist device.
  • the conventional lane keeping assist device determines whether or not the traveling speed of the host vehicle is higher than a predetermined assist start speed (for example, 60 km/h).
  • the conventional lane keeping assist device performs the operation of step ST2 if the traveling speed is higher than the assistance start speed (step ST1 “YES”), and continues the speed comparison determination if it is low (step ST1 “NO”).
  • step ST2 the conventional lane keeping assist device starts lane keeping assistance because the condition for starting assistance is satisfied in step ST1.
  • FIG. 4 is a diagram for explaining the concept of lane detection when a conventional lane keeping assist device supports lane keeping.
  • FIG. 4 shows a captured image in front of the host vehicle 100. In this captured image, a characteristic structure is shown in addition to a part of the vehicle 100. Characteristic structures are a guard rail 101, a median strip 102, roadway outside lines 103 and 104, a boundary line 105, and the like. The road outside lines 103 and 104 and the boundary line 105 divide the lane.
  • the lane keeping support system 1 according to the first embodiment shown in FIG. 1 is referred to in order to explain a conventional lane keeping support device.
  • the image processing device 21 discriminates the characteristic structure from the image captured by the camera 20 as shown in FIG. 4, and the roadside line 103 existing in front of the host vehicle 100 among the discriminated characteristic structures. And the characteristic line 110 is detected from the boundary line 105.
  • the position information of the characteristic line 110 detected by the image processing device 21 is provided to the ECU 10, which is a lane keeping assist device.
  • the ECU 10 gives a command to the electric circuit 63 while referring to the detection results of the steering angle sensor 61 and the torque sensor 62, drives the wheels 66 so that the characteristic line 110 is located in front of the host vehicle 100, and supports the lane keeping. I do.
  • the conventional lane keeping assist device performs the above-described operation, if the detection of the characteristic line 110 in FIG. 4 becomes unstable, the assist operation itself becomes impossible, and the assist is stopped to allow the driver to manually drive the vehicle. I have to entrust it.
  • To detect the characteristic line 110 it is necessary that the characteristic structure is captured by the camera 20 and recognized by the image processing device 21. This is one of the reasons why the support start speed is set so that the roads for exclusive use of vehicles, expressways, etc., which are maintained road environments, are targeted for support.
  • the conventional lane keeping support device is configured to support only on roads where the lane detection environment is presumed to be good.
  • roads that are capable of traveling at a speed exceeding the support start speed even in general roads, and that have a lane detection environment sufficient for support.
  • the lane keeping assist device relies only on the actual road environment, when the vehicle 100 travels on a road in which an environment sufficient for assistance and an environment not for assistance coexist, the assistance becomes intermittent and uncomfortable for the driver. And the operation becomes unreliable.
  • the lane keeping support system 1 stores the actual operation state of the lane keeping support in the map DB 33 in association with the traveling history of the vehicle 100, so that the lane keeping assist system 1 can be used even in a non-maintained environment.
  • the number of operation opportunities for lane keeping support is increased, the discomfort of the driver is suppressed, and the convenience is improved.
  • FIG. 5 is a flowchart showing an operation example of the ECU 10 according to the first embodiment.
  • the ECU 10 starts the operation shown in the flowchart of FIG. 5 when the ignition switch of the own vehicle 100 is turned on, and ends the operation when the ignition switch is turned off.
  • the support availability determination unit 11 acquires the traveling speed of the host vehicle 100 from the vehicle network ECU 40 or the vehicle network ECU 40 via the navigation device 32. Then, the supportability determination unit 11 determines whether the traveling speed is higher than a predetermined support start speed.
  • the support start speed may be set to a speed that considers the relationship with the power required for steering shown in FIG. 2, or a speed that assumes a maintained traveling environment (for example, 60 km/h), or a lower speed. It may be a fixed value or a variable value according to the traveling road or traveling area.
  • the traveling road is an ordinary road, a motorway, a highway, or the like.
  • the running area is a suburb where disturbances such as pedestrians and bicycles obstruct traveling are small, and an urban area where there is much disturbance.
  • the information on the traveling road and the traveling area is stored in the map DB 33.
  • the support availability determination unit 11 performs the operation of step ST12.
  • the support availability determination unit 11 determines that the support is not suitable for the lane keeping support, and performs the operation of step ST18.
  • step ST12 the navigation device 32 uses the traveling position measured by the GPS receiver 30 and the information in the map DB 33 to detect the road on which the vehicle 100 is traveling (hereinafter, referred to as "traveling road”). To do. Then, the navigation device 32 acquires the information indicating the detected running road (for example, the link data shown in FIG. 8A described later) from the map DB 33 and outputs the information to the supportability determination unit 11. The support availability determination unit 11 acquires information indicating the road on which the vehicle is traveling from the navigation device 32.
  • step ST13 the support availability determination unit 11 uses the information indicating the traveling road acquired from the navigation device 32 to determine the characteristic line accuracy and the learning count stored in association with the information indicating the traveling road on the map. Obtain from DB33.
  • step ST14 the support availability determination unit 11 uses the learning count previously stored in the map DB 33 for the traveling road of the host vehicle 100 to determine whether the traveling road is a road that is regularly used by the driver of the own vehicle 100. Determine whether or not. Specifically, when the number of times of learning acquired from the map DB 33 is equal to or less than a predetermined threshold value (step ST14 “NO”), the support availability determination unit 11 determines that the running road is not the road that the driver regularly uses. After making a determination, the operation of step ST18 is performed. On the other hand, when the learning count is larger than the threshold value (step ST14 “YES”), the supportability determination part 11 determines that the traveling road is a road that is regularly used by the driver, and performs the operation of step ST15.
  • step ST15 the support availability determination unit 11 determines the availability of lane keeping support for the road on which the vehicle 100 is running using the characteristic line accuracy stored in the map DB 33 in the past. Specifically, when the characteristic line accuracy acquired from the map DB 33 is less than or equal to a predetermined threshold value (step ST15 “NO”), the support availability determination unit 11 determines that the running road is not suitable for lane keeping support. Then, the operation of step ST18 is performed. On the other hand, when the characteristic line accuracy is greater than the threshold value (step ST15 “YES”), the support availability determination unit 11 determines that the traveling road is a road suitable for lane keeping support, and performs the operation of step ST16.
  • step ST16 the support availability determination unit 11 instructs the lane keeping support unit 12 to start lane keeping support.
  • the lane keeping assisting unit 12 acquires the characteristic line 110 detected by the image processing device 21 from the image processing device 21. Then, the lane keeping assisting unit 12 commands the electric circuit 63 to refer to the detection results of the steering angle sensor 61 and the torque sensor 62 so that the host vehicle 100 travels in the lane divided by the acquired characteristic line 110.
  • a well-known technique may be used for the lane keeping assisting method by the lane keeping assisting unit 12, and detailed description thereof will be omitted.
  • step ST17 the support availability determination unit 11 instructs the characteristic line accuracy learning unit 13 to learn the characteristic line accuracy.
  • the characteristic line accuracy learning unit 13 learns the accuracy of the characteristic line 110 detected in the lane in which the vehicle 100 has already traveled, and stores the learning result in the map DB 33. The detailed operation in step ST17 will be described later.
  • step ST18 the support availability determination unit 11 determines that the running road is not suitable for lane keeping support, and therefore instructs the lane keeping support unit 12 to stop the lane keeping support.
  • the lane keeping support unit 12 receiving this instruction stops the lane keeping support.
  • the lane keeping support unit 12 may instruct the HMI 50 to notify the driver of the start of the lane keeping support before starting the lane keeping support in step ST16. Further, the lane keeping support unit 12 may instruct the HMI 50 to notify the driver of the stop of the lane keeping support before stopping the lane keeping support in step ST18.
  • the HMI 50 displays, for example, a linear image along a road link on a map screen, an icon indicating an operation state of lane keeping support, or an alarm display on a meter panel or a head-up display.
  • FIG. 6 is a diagram showing a road shape on which the host vehicle 100 is traveling in the first embodiment.
  • road links L1_1, L1_2, L1_3, L1_4, L2_1, L3_1, L4_1, L5_1, L5_2 are road shapes approximated by line segments.
  • the nodes N1_1, N1_2, N1_3, N1_4, N1_5, N2_1, N3_1, N4_1, N5_1, N5_2 are bending points or intersections connecting road links.
  • the host vehicle 100 is traveling on one of the road links.
  • the navigation device 32 considers that the own vehicle 100 exists on any road link, considers that the direction change of the own vehicle 100 at a turning point or an intersection is performed on any node, and determines that Detects the running position and the running road. Therefore, the navigation device 32 outputs the link data of the road link corresponding to the traveling road to the support availability determination unit 11 (see step ST12 in FIG. 5). In the example of FIG. 6, since the vehicle 100 is traveling on the road link L1_2, the navigation device 32 outputs the link data of the road link L1_2 between the node N1_2 and the node N1_3 to the support availability determination unit 11. ..
  • FIG. 7 is a diagram illustrating an example of a defect in feature line detection in the image processing device 21 according to the first embodiment.
  • the image processing device 21 detects the characteristic lines 110 such as white lines existing on the left and right of the host vehicle 100 as described with reference to FIG.
  • the image processing device 21 cannot detect the characteristic lines 110 such as white lines existing on the left and right of the host vehicle 100.
  • the information for the lane keeping support by the lane keeping support unit 12 becomes uncertain, and the lane keeping support stops intermittently.
  • the ECU 10 when the host vehicle 100 travels on a road as shown in FIG. 7, stores the information indicating the running road and the characteristic line accuracy in the map DB 33 in association with each other. deep. Specifically, the ECU 10 stores the link data of the road link corresponding to the road on which the vehicle is running, the characteristic line certainty, and the learning number of the characteristic line certainty in association with each other in the map DB 33.
  • the ECU 10 lanes on the lanes where the lane keeping assistance is intermittently stopped based on the detection status of the characteristic line 110 stored in the map DB 33.
  • Lane maintenance support will be provided on roads where lane maintenance support does not stop intermittently without maintenance support. This prevents discomfort and a decrease in reliability due to intermittent lane keeping support.
  • FIG. 8A is a diagram schematically showing a part of the information stored in the map DB 33 according to the first embodiment.
  • One piece of link data includes at least the ID of the road link itself (L1_2 in this example), the information of the node indicating the start point and the end point of the road link, and may include other various attributes. Examples of attributes include speed limits and road types (general roads, toll roads, expressways, etc.). Furthermore, in the first embodiment, the number of times the characteristic line certainty of this road link is learned and the characteristic line certainty as the latest learning result are added as the attributes of the road link.
  • FIG. 8B is a flowchart showing an operation example by the characteristic line accuracy learning unit 13 of the first embodiment.
  • the characteristic line accuracy learning unit 13 performs the operations of steps ST17-1 to ST17-7 of FIG. 8B in step ST17 of FIG.
  • step ST17-1 the navigation device 32 detects the road on which the vehicle 100 is traveling, using the traveling position measured by the GPS receiver 30 and the information in the map DB 33. Then, the characteristic line accuracy learning unit 13 acquires from the map DB 33 the link data of the road link (for example, the link data shown in FIG. 8A) stored in association with the traveling road detected by the navigation device 32.
  • the link data of the road link for example, the link data shown in FIG. 8A
  • step ST17-2 the characteristic line accuracy learning unit 13 acquires the characteristic line 110 detected by the image processing device 21 from the image processing device 21. Then, the characteristic line accuracy learning unit 13 determines whether or not the characteristic line 110 has been detected by the image processing device 21. When the characteristic line 110 can be detected by the image processing device 21 (step ST17-2 “YES”), the characteristic line accuracy learning unit 13 can detect the characteristic line 110 in the running road link in step ST17-3. The detection length, which is the length of the section, is updated. On the other hand, when the characteristic line 110 cannot be detected by the image processing device 21 (step ST17-2 “NO”), the characteristic line accuracy learning unit 13 skips the operation of step ST17-3.
  • step ST17-4 the navigation device 32 detects the road on which the vehicle 100 is traveling, using the traveling position measured by the GPS receiver 30 and the information in the map DB 33. Then, the characteristic line accuracy learning unit 13 acquires information indicating the traveling road from the navigation device 32, and determines whether the own vehicle 100 has exited from the road link indicated by the link data acquired in step ST17-1. To do.
  • the characteristic line accuracy learning unit 13 performs the operation of step ST17-5.
  • step ST17-4 "NO" the characteristic line accuracy learning unit 13 returns to step ST17-2 and continues to detect the characteristic line 110.
  • step ST17-5 the characteristic line accuracy learning unit 13 updates the learning count by adding “1” to the learning count included in the link data of the exiting road link stored in the map DB 33.
  • step ST17-6 the characteristic line certainty learning unit 13 calculates the characteristic line certainty for the exiting road link. Then, the characteristic line certainty degree learning unit 13 updates the characteristic line certainty degree by overwriting the calculated characteristic line certainty degree on the characteristic line certainty degree of the corresponding link data in the map DB 33.
  • FIG. 8C is a diagram showing an example of a formula for calculating the characteristic line accuracy in the first embodiment.
  • the characteristic line accuracy learning unit 13 calculates the characteristic line accuracy by dividing the detection length accumulated in a certain road link by the cumulative traveling distance in the road link.
  • the characteristic line accuracy learning unit 13 may calculate the characteristic line accuracy by dividing the detection length accumulated in a certain road link by a value obtained by multiplying the length of the road link and the number of times of learning.
  • the characteristic line accuracy relates to the stability of the lane keeping support operation within a certain road link, and it can be said that the higher the characteristic line accuracy, the more stable the lane keeping support conditions are. That is, it can be said that the higher the characteristic line accuracy, the more continuously the lane keeping support operates, and the lower the characteristic line accuracy, the more intermittently the lane keeping support operates.
  • step ST17-7 the support availability determination part 11 determines whether or not the current traveling speed is higher than the support start speed, as in step ST11 of FIG.
  • the support availability determination unit 11 instructs the characteristic line accuracy learning unit 13 to perform the operation of step ST17-1.
  • the support availability determination unit 11 instructs the characteristic line accuracy learning unit 13 to stop the learning of the characteristic line accuracy. Then, the process proceeds to step ST18 of FIG.
  • the ECU 10 includes the lane keeping support unit 12, the characteristic line accuracy learning unit 13, and the support availability determination unit 11.
  • the lane keeping assisting unit 12 uses the characteristic line 110, which is detected from the captured image in front of the own vehicle 100 and divides the lane in which the own vehicle 100 travels, so as to keep the own vehicle 100 running within the lane. Assist.
  • the characteristic line accuracy learning unit 13 learns the accuracy of the characteristic line 110 detected on the road on which the vehicle 100 is traveling.
  • the support availability determination unit 11 uses the accuracy of the characteristic line 110 learned by the characteristic line accuracy learning unit 13 in the past and the number of times of learning performed on the road on which the own vehicle 100 has traveled in the past, to determine the own vehicle 100.
  • the ECU 10 learns by associating the traveling road of the host vehicle 100 with the characteristic line accuracy, so that the lane keeping support is not performed on the road where the lane keeping support has been intermittently performed in the past. Can be prevented. Further, the ECU 10 can determine whether or not the road on which the vehicle 100 is about to travel is a road that the vehicle 100 normally uses, by using the number of times of learning the characteristic line accuracy. Therefore, the ECU 10 can prevent the intermittent operation of the lane keeping function, especially on the road that the vehicle 100 normally uses. Further, the ECU 10 can provide stable and reliable lane keeping support.
  • Embodiment 2 The ECU 10 according to the first embodiment is configured to determine whether or not lane keeping support is possible for each road link. However, roads that are dataized so that the road link length is shortened, roads in which intersections and the like exist continuously, or roads in which high and low characteristic line accuracy is alternately and continuously present. Then, whether or not the lane keeping support is intermittently switched, and as a result, the driver feels uncomfortable. Therefore, the ECU 10 according to the second embodiment is configured to determine whether or not lane keeping assistance is possible for the entire search route searched by the navigation device 32.
  • FIG. 9 is a diagram showing an example of functional blocks of the ECU 10 according to the second embodiment.
  • the ECU 10 according to the second embodiment has a configuration including a support availability determination unit 11a instead of the support availability determination unit 11 in the ECU 10 of the first embodiment shown in FIG. 1B.
  • parts that are the same as or correspond to those in FIGS. 1A and 1B are assigned the same reference numerals and explanations thereof are omitted.
  • the support availability determination unit 11a uses the accuracy of the characteristic line 110 learned by the characteristic line accuracy learning unit 13 in the past to determine whether the lane keeping assistance unit 12 can assist the search route searched by the navigation device 32. ..
  • FIG. 10A is a diagram showing an example of a search route searched by the navigation device 32 according to the second embodiment.
  • the search route shown in FIG. 10A is represented by road link groups L1_1, L1_2, L1_3, and L1_4 that connect the node N1_1 that is the departure point O and the node N1_5 that is the destination D.
  • FIG. 10B is a diagram showing an example of a calculation formula of route characteristic line accuracy in the second embodiment.
  • the support availability determination unit 11a calculates the characteristic line accuracy (hereinafter, referred to as "route characteristic line accuracy") in the entire search route connecting the node N1_1 which is the starting point O or the traveling position and the node N1_5 which is the destination D. To do.
  • the route characteristic line certainty is obtained by integrating the characteristic line certainty stored in the map DB 33 in association with the road link (Lx_y) included in the searched route.
  • FIG. 10C is a diagram showing another example of the calculation formula of the route characteristic line accuracy in the second embodiment.
  • the support availability determination unit 11a associates the function ⁇ (n) whose value gradually decreases with the distance n from the departure point O with the road link (Lx_y) included in the searched route, and associates it with the map DB 33.
  • the product-sum operation is performed with the feature line accuracy stored in.
  • the distance n from the departure place O may be a physical distance or a number representing the number of the road link in the search route.
  • the supportability determination unit 11a preferably sequentially recalculates the route characteristic line accuracy while the host vehicle 100 is traveling.
  • FIG. 11 is a flowchart showing an operation example of the ECU 10 according to the second embodiment.
  • the operation in steps ST11 to ST18 in FIG. 11 is the same as the operation in steps ST11 to ST18 in FIG.
  • the support availability determination unit 11a determines whether or not the searched route exists in the navigation device 32 in step ST21.
  • the searched route exists step ST21 “YES”
  • the support availability determination part 11a acquires the link data of the road link group scheduled to travel along the searched route from the map DB 33.
  • the supportability determination part 11a skips the operations of steps ST22 to ST24.
  • step ST23 the support availability determination unit 11a calculates the formula shown in FIG. 10B or 10C using the link data acquired from the map DB 33, and obtains the route feature line accuracy.
  • step ST24 the support availability determination unit 11a compares the route feature line accuracy calculated in step ST23 with a predetermined threshold value to determine whether or not lane keeping support is available for the entire searched route.
  • the supportability determination unit 11a determines that the searched route is not suitable for lane keeping support, and performs the operation of step ST12. .
  • the support availability determination unit 11a determines the availability of lane keeping support for each road link on which the vehicle 100 is traveling.
  • step ST24 “YES” when the route feature line accuracy is larger than the threshold value (step ST24 “YES”), the supportability determination unit 11a determines that the searched route is a searched route suitable for lane keeping support, and performs the operation of step ST16. .. In step ST16, the supportability determination part 11a instructs the lane keeping support part 12 to start lane keeping support.
  • the supportability determination unit 11a uses the accuracy of the characteristic line 110 learned by the characteristic line accuracy learning unit 13 in the past to maintain the lane on the searched route searched by the navigation device 32. Whether or not the support by the support unit 12 is possible is determined. With this configuration, the ECU 10 can improve continuity of preventing intermittent operation of the lane keeping function.
  • Embodiment 3 As described in the second embodiment, since the ECU 10 according to the first embodiment is configured to determine whether or not lane keeping support is possible for each road link, the lane keeping support may or may not be intermittently switched. .. Therefore, the ECU 10 according to the third embodiment is configured to predict a route along which the host vehicle 100 travels and determine whether or not lane keeping assistance is possible on the predicted route.
  • FIG. 12 is a diagram showing an example of functional blocks of the ECU 10 according to the third embodiment.
  • the ECU 10 according to the third embodiment has a configuration including a support availability determination unit 11b instead of the support availability determination unit 11 in the ECU 10 according to the first embodiment shown in FIG. 1B.
  • the ECU 10 according to the third embodiment has a configuration in which the travel route prediction unit 14 is added to the ECU 10 according to the first embodiment shown in FIG. 1B.
  • parts that are the same as or correspond to those in FIGS. 1A and 1B are assigned the same reference numerals and explanations thereof are omitted.
  • the travel route prediction unit 14 predicts a route along which the host vehicle 100 travels by using the information in the map DB 33 and the information in the vehicle network ECU 40, and the predicted route (hereinafter, referred to as “predicted route”) is determined as the support availability determination unit 11b. Output to.
  • the support availability determination unit 11b uses the accuracy of the characteristic line 110 learned by the feature line accuracy learning unit 13 in the past to determine whether the lane keeping support unit 12 can assist the route predicted by the travel route prediction unit 14. To do.
  • FIG. 13A is a diagram showing an example of a road shape on which the host vehicle 100 is traveling in the third embodiment.
  • FIG. 13B is a diagram for explaining the route prediction method by the traveling route prediction unit 14 according to the third embodiment, and shows an example of typical driving operation.
  • FIG. 13B shows cases #1 to # as typical combinations of driving operations that the driver of the vehicle 100 traveling on the road link L1_1 may perform before the node N1_2, which is the next intersection or the like. 4 etc. are shown.
  • the cases #1 to #4 and the like shown in FIG. 13B are assumed to be given to the travel route prediction unit 14 in advance.
  • FIG. 13C is a diagram for explaining the route prediction method by the travel route prediction unit 14 according to the third embodiment, and shows an example of the coefficient for each representative driving operation.
  • FIG. 13C shows a type of situation at the node N1_2 when a driving operation such as cases #1 to #4 is performed, and a road link in which the own vehicle 100 is predicted to travel because it is connected to the node N1_2.
  • the coefficients ⁇ (n, L) of L1_2, L2_1, and L3_1 are shown.
  • the coefficient ⁇ (n,L) is a coefficient of the road link L connected to a certain node n, and the larger the value of the coefficient, the higher the possibility that the vehicle 100 travels on the road link L.
  • the own vehicle 100 is most likely to go straight at this node N1_2.
  • the coefficient ⁇ when going straight from N1_2 to the road link L1_2 is “0.8”, which is the highest. Note that the correspondence relationship between the cases #1 to #4 and the like shown in FIGS. 13B and 13C and the coefficient ⁇ (n,L) is assumed to be given to the travel route prediction unit 14 in advance.
  • the travel route prediction unit 14 acquires, from the vehicle network ECU 40, information indicating a driving operation such as an accelerator and a brake performed by the driver of the vehicle 100. Then, the travel route prediction unit 14 determines a case corresponding to the acquired information indicating the driving operation, and extracts the coefficient ⁇ (n,L) of each road link predicted to pass in that case. The travel route prediction unit 14 outputs the extracted coefficient ⁇ (n, L) of each road link to the support availability determination unit 11b.
  • FIG. 13D is a diagram showing an example of a calculation formula of route characteristic line accuracy in the third embodiment.
  • the route characteristic line accuracy in the third embodiment is the accuracy of the characteristic line in the predicted route, while the route characteristic line accuracy in the second embodiment is the accuracy of the characteristic line in the searched route.
  • the support availability determination unit 11b acquires the coefficient ⁇ (n,L) of each road link in which the vehicle 100 may travel from the travel route prediction unit 14. Then, the supportability determination unit 11b performs a product-sum operation of the coefficient ⁇ (n, L) of each road link and the characteristic line accuracy stored in the map DB 33 in association with each road link according to the calculation formula of FIG. 13D. , Obtain the route characteristic line accuracy of the route predicted to travel.
  • FIG. 14 is a flowchart showing an operation example of the ECU 10 according to the third embodiment.
  • the operation in steps ST11 to ST18 in FIG. 14 is the same as the operation in steps ST11 to ST18 in FIG.
  • the support availability determination unit 11b instructs the travel route prediction unit 14 to predict the travel route.
  • the traveling route prediction unit 14 predicts the traveling route in step ST31.
  • the traveling route prediction unit 14 acquires information indicating the driving operation from the vehicle network ECU 40, determines the case corresponding to the driving operation, and determines the coefficient of each road link predicted to pass in the determined case. Extract. Then, the travel route prediction unit 14 outputs the extracted coefficient of each road link to the support availability determination unit 11b.
  • the traveling route prediction unit 14 causes the own vehicle 100 to go straight through the node N1_2 and follow the road link L1_2. It is predicted that the vehicle will travel, and the coefficient “1.0” of the road link L1_2 is output to the support availability determination unit 11b.
  • step ST32 the support availability determination unit 11b acquires the coefficient of each road link from the travel route prediction unit 14.
  • the support availability determination unit 11b also acquires from the map DB 33 the link data stored in the map DB 33 in association with each road link.
  • step ST33 the support availability determination unit 11b uses the characteristic line accuracy included in the link data acquired from the map DB 33 and the coefficient of each road link acquired from the travel route prediction unit 14 to perform the calculation shown in FIG. 13D. An equation is calculated to obtain the route feature line accuracy.
  • step ST34 the support availability determination unit 11b compares the route feature line accuracy calculated in step ST33 with a predetermined threshold value to determine whether or not lane keeping assistance is available on the predicted route.
  • the supportability determination unit 11b determines that the predicted route is not suitable for lane keeping support, and performs the operation of step ST12.
  • the support availability determination unit 11b determines the availability of lane keeping assistance for each road link in which the vehicle 100 is traveling.
  • step ST34 “YES” the support availability determination unit 11b determines that the predicted route is a route suitable for lane keeping support, and performs the operation of step ST16.
  • the supportability determination part 11b instructs the lane keeping support part 12 to start lane keeping support.
  • the ECU 10 includes the travel route prediction unit 14 that predicts the route along which the vehicle 100 travels.
  • the support availability determination unit 11b uses the accuracy of the characteristic line 110 learned by the feature line accuracy learning unit 13 in the past to determine whether the lane keeping support unit 12 can assist the route predicted by the travel route prediction unit 14. To do.
  • the ECU 10 can predict the travel route even on the road that is the daily travel road where the driver of the vehicle 100 does not use the navigation device 32. Therefore, the ECU 10 can improve continuity of preventing intermittent operation of the lane keeping function.
  • the performances of the camera 20 and the image processing device 21 used for detecting the characteristic line are affected by the illuminance of the traveling environment at the time of traveling. Therefore, the ECU 10 according to the fourth embodiment learns the characteristic line accuracy for each illuminance and uses the learned characteristic line accuracy corresponding to the illuminance at the time of determining whether or not lane keeping support is possible. And
  • FIG. 15 is a diagram showing an example of functional blocks of the ECU 10 according to the fourth embodiment.
  • the ECU 10 according to the fourth embodiment includes a support availability determination unit 11c and a feature line accuracy learning unit 13c in place of the support availability determination unit 11 and the feature line accuracy learning unit 13 in the ECU 10 of the first embodiment shown in FIG. 1B. It is a configuration provided with.
  • the ECU 10 according to the fourth embodiment has a configuration in which an environment determination unit 15 is added to the ECU 10 according to the first embodiment shown in FIG. 1B.
  • the navigation device 32 according to the fourth embodiment is configured to include a map DB 33c instead of the map DB 33 of the first embodiment shown in FIG. 1B.
  • parts that are the same as or correspond to those in FIGS. 1A and 1B are assigned the same reference numerals and explanations thereof are omitted.
  • the environment determination unit 15 uses the information from the vehicle network ECU 40 to determine the environment in which the vehicle 100 travels, and outputs the determined environment to the support availability determination unit 11c.
  • the environment determined by the environment determination unit 15 is, for example, an environment of illuminance.
  • the characteristic line accuracy learning unit 13c learns the accuracy of the characteristic line 110 detected on the road on which the vehicle 100 is traveling for each environment determined by the environment determination unit 15, and stores the learning result in the map DB 33c.
  • the support availability determination unit 11c uses the accuracy of the feature line 110 for each environment that has been learned by the feature line accuracy learning unit 13c in the past, and determines whether or not the lane keeping support unit 12 can support the road on which the vehicle 100 is traveling and the environment. To judge.
  • FIG. 16 is a diagram schematically showing a part of the information stored in the map DB 33c of the fourth embodiment.
  • One piece of link data includes at least the ID of the road link itself (L1_2 in this example), the information of the node indicating the start point and the end point of the road link, and may include other various attributes. Examples of attributes include speed limits and road types (general roads, toll roads, expressways, etc.). Further, in the fourth embodiment, the number of times the characteristic line accuracy of the road link is learned and the characteristic line accuracy for each illuminance range at the time of learning are added as the attributes of the road link.
  • the characteristic line accuracy is “0 or more and less than a” (hereinafter, referred to as “0 to a”) where the illuminance is the lowest, and “a or more and less than b” where the illuminance is medium (hereinafter, “a to b”). ”) and the highest illuminance is "more than b and less than c" (hereinafter referred to as "b to c").
  • FIG. 17A is a diagram showing an example of a search route searched by the navigation device 32 of the fourth embodiment.
  • the search route shown in FIG. 17A is represented by road link groups L1_1, L1_2, L1_3, L1_4 that connect the node N1_1 that is the departure point O and the node N1_5 that is the destination D.
  • FIG. 17B is a diagram showing a part of the link data of the road link groups L1_1, L1_2, L1_3, L1_4 along the search route shown in FIG. 17A.
  • FIG. 17C is a diagram showing an example of a calculation formula of route characteristic line accuracy in the fourth embodiment.
  • the route feature line accuracy in the fourth embodiment may be the accuracy of the feature line in the searched route or the accuracy of the feature line in the predicted route.
  • the environment determination unit 15 acquires the illuminance t of the current traveling environment from the vehicle network ECU 40 and outputs it to the supportability determination unit 11c.
  • the support availability determination unit 11c acquires, from the map DB 33c, the characteristic line accuracy (Lx_y, t) of each road link included in the searched route that corresponds to the illuminance t.
  • the support availability determination unit 11c acquires the characteristic line accuracy stored as “v2” in the link data. Then, the supportability determination unit 11c calculates the sum of the characteristic line accuracies (Lx_y, t) of the road links included in the searched route, which correspond to the illuminance t, according to the calculation formula of FIG. obtain.
  • the environment determination unit 15 estimates the time required for the vehicle 100 to pass each road link on the search route, and based on the current time, the estimated illuminance T at the estimated time when the vehicle 100 passes each road link on the search route. You may expect.
  • the supportability determination unit 11c adds the future illuminance change by calculating the route characteristic line accuracy using the characteristic line accuracy of each road link included in the searched route that corresponds to the expected illuminance T. It is possible to determine whether or not lane keeping support is available.
  • FIG. 18A is a diagram illustrating the illuminance predicted by the environment determination unit 15 according to the fourth embodiment.
  • the horizontal axis of the graph shown in FIG. 18A is the estimated time when the vehicle 100 passes through the road link, and the vertical axis is the expected illuminance T.
  • the host vehicle 100 travels on the search route shown in FIG. 17A.
  • the current time is the dusk time zone.
  • the environment determination unit 15 acquires the illuminance t corresponding to the illuminance range “b to c” from the vehicle network ECU 40 at present when the vehicle 100 is traveling on the road link L1_1.
  • the supportability determination unit 11c acquires “v3” as the characteristic line accuracy of the road link L1_1 from the map DB 33c.
  • the environment determination unit 15 determines that the vehicle 100 passes through the road links L1_2, L1_3, L1_4 based on the current illuminance t corresponding to the illuminance range “b to c” acquired from the vehicle network ECU 40 and the current time. Predict each expected illuminance T at the estimated time. For example, the environment determination unit 15 estimates the illuminance by using the current traveling position (latitude/longitude) and date and time detected by the navigation device 32. In FIG. 18A, since the current time is in the dusk time zone, the expected illuminance T becomes lower as the time advances.
  • the expected illuminance T at the estimated time of passing the road link L1_2 corresponds to the illuminance range “b to c”
  • the expected illuminance T at the estimated time of passing the road link L1_3 falls to the illuminance range “a to b”.
  • the expected illuminance T at the estimated time of passing through the road link L1_4 corresponds to the illuminance range “0 to a”.
  • the supportability determination unit 11c sets "v3" as the characteristic line accuracy of the road link L1_2, "v2" as the characteristic line accuracy of the road link L1_3, and "v1" as the characteristic line accuracy of the road link L1_4. Obtain from DB33c.
  • FIG. 18B is a diagram showing an example of a calculation formula of route characteristic line accuracy in the fourth embodiment.
  • the support availability determination unit 11c obtains the route characteristic line accuracy by calculating the calculation formula of FIG. 18B using the characteristic line accuracy of the road links L1_1, L1_2, L1_3, L1_4 acquired from the map DB 33c as described above.
  • the environment determination unit 15 may re-estimate the illuminance when the actual illuminance t and the expected illuminance T of the traveling environment deviate due to the lighting of the headlamp of the vehicle 100, the lighting of the streetlight of the road on which the vehicle is traveling, or the like. Good.
  • the support availability determination unit 11c may recalculate the route feature line accuracy based on the predicted illuminance T re-estimated by the environment determination unit 15, and determine the support availability again by the threshold value determination.
  • the ECU 10 includes the environment determination unit 15 that determines the environment in which the host vehicle 100 travels.
  • the characteristic line accuracy learning unit 13c learns the accuracy of the characteristic line 110 detected on the road on which the own vehicle 100 is traveling for each environment determined by the environment determination unit 15.
  • the support availability determination unit 11c uses the accuracy of the feature line 110 for each environment that has been learned by the feature line accuracy learning unit 13c in the past, and determines whether or not the lane keeping support unit 12 can support the road on which the vehicle 100 is traveling and the environment. To judge.
  • the ECU 10 can suppress the variation in the support availability determination due to the difference in the characteristic line detection performance due to the difference in the traveling environment.
  • the illuminance environment is shown as an example of the environment, but the environment is not limited to the illuminance environment, and any environment that affects the feature line detection performance may be used. Further, the configuration of the fourth embodiment can be combined with any of the configurations of the first to third embodiments.
  • the threshold value for the number of learnings and the threshold value for the characteristic line accuracy described in each embodiment do not need to be fixed values, and are variable values that can be adjusted and set according to the driver's preference. May be.
  • the threshold value is a variable value
  • the ECU 10 displays the threshold value setting screen by using, for example, the HMI 50, and receives the threshold value information set by the driver.
  • the lane keeping assist device determines whether or not lane keeping assist is possible, it is suitable for use as a driving assistance device or the like that assists driving such as lane keeping.
  • 1 lane maintenance support system 10 ECU (lane maintenance support device), 11, 11a, 11b, 11c support availability determination unit, 12 lane maintenance support unit, 13 and 13c characteristic line accuracy learning unit, 14 travel route prediction unit, 15 environment Judgment unit, 20 camera, 21 image processing device, 30 GPS receiver, 31 antenna, 32 navigation device, 33, 33c map DB, 40 vehicle network ECU, 50 HMI, 60 steering wheel, 61 steering angle sensor, 62 torque sensor, 63 electric circuit, 64 motor, 65 gearbox, 66 wheels, 100 own vehicle, 101 guardrail, 102 median strip, 103, 104 road outside line, 105 boundary line, 110 characteristic line, 121 faint, 122 break, L1_1, L1_2 , L1_3, L1_4, L2_1, L3_1, L4_1, L5_1, L5_2 Road links, N1_1, N1_2, N1_3, N1_4, N1_5, N2_1, N3_1, N4_1, N5_1, N5_2 nodes.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)
  • Navigation (AREA)

Abstract

This lane keeping assist unit (12) uses a characteristic line (110) that divides the lane on which a host vehicle (100) is traveling, detected from an image imaged in front of the host vehicle (100), to assist the host vehicle (100) to keep traveling within the lane. A characteristic line accuracy learning unit (13) learns the accuracy of the characteristic line (110) detected on the road on which the host vehicle (100) is traveling. An assist advisability determination unit (11) uses the accuracy of the characteristic line (110) learned by the characteristic line accuracy learning unit (13) in the past and the number of times learning was performed on the road that the host vehicle (100) traveled in the past to determine the advisability of assisting using the lane keeping assist unit (12) on the road on which the host vehicle (100) is traveling.

Description

車線維持支援装置及び車線維持支援方法Lane maintenance support device and lane maintenance support method
　この発明は、車線維持支援装置及び車線維持支援方法に関するものである。 The present invention relates to a lane keeping support device and a lane keeping support method.
　従来、車両前方の道路をカメラで撮像して走行車線を認識し、車線内の走行を維持するようにハンドル角の制御を行う車線維持機能を備えた車両がある。例えば特許文献１に記載されたサーバ装置は、車線維持機能を有する複数の車両から走行中の車線維持機能の利用情報を取得し、取得した利用情報に基づいて車線維持機能の利用率が低い領域を抽出する。車線維持機能の利用率が低い領域とは、車線のかすれ等により一時的に車線が認識されず、車線維持機能が断続的に動作する領域である。車線維持機能の断続的な動作は、運転者に不快感を与える。そこで、特許文献１に記載されたサーバ装置は、車線維持機能の利用率が低い領域の手前を走行する車両に対して車線維持機能の停止を指示することにより、車線維持機能の断続的な動作を防止する。これにより、運転者は、これから走行する領域における車線維持機能の停止に対する準備ができ、車線維持機能が停止した際に違和感なく手動運転に移行することができる。 Conventionally, there are vehicles equipped with a lane keeping function that images the road in front of the vehicle with a camera to recognize the driving lane and controls the steering wheel angle so as to keep driving in the lane. For example, the server device described in Patent Document 1 obtains usage information of a lane keeping function while traveling from a plurality of vehicles having a lane keeping function, and an area where the usage rate of the lane keeping function is low based on the obtained usage information. To extract. The area where the usage rate of the lane keeping function is low is an area in which the lane keeping function is intermittently operated because the lane is not recognized temporarily due to fading of the lane. The intermittent operation of the lane keeping function makes the driver uncomfortable. Therefore, the server device described in Patent Document 1 issues an intermittent operation of the lane keeping function by giving an instruction to stop the lane keeping function to a vehicle traveling in an area where the utilization rate of the lane keeping function is low. Prevent. As a result, the driver can prepare for the stop of the lane keeping function in the area where he/she is going to travel, and when the lane keeping function is stopped, the driver can shift to the manual driving without feeling uncomfortable.
特開２０１７－１０２５５６号公報Japanese Patent Laid-Open No. 2017-102556
　特許文献１に係る発明においては、自車両が常用する道路を事前に走行する他車両が必ず存在する保障が無く、自車両が常用する道路について車線維持機能の利用情報が蓄積されるとは限らなかった。そのため、特許文献１に係る発明には、車線維持機能の断続的な動作を防止可能な道路が限られるという課題があった。 In the invention according to Patent Document 1, there is no guarantee that there is another vehicle that travels in advance on the road that the own vehicle normally uses, and the use information of the lane keeping function is not always accumulated for the road that the own vehicle normally uses. There wasn't. Therefore, the invention according to Patent Document 1 has a problem that roads that can prevent intermittent operation of the lane keeping function are limited.
　この発明は、上記のような課題を解決するためになされたもので、特に自車両が常用する道路において、車線維持機能の断続的な動作を防止することを目的とする。 The present invention has been made to solve the above problems, and it is an object of the present invention to prevent intermittent operation of the lane keeping function, especially on a road that is regularly used by the host vehicle.
　この発明に係る車線維持支援装置は、自車両前方の撮像画像から検出された、自車両が走行する車線を区画する特徴線を用いて、車線内の走行を維持するように自車両を支援する車線維持支援部と、自車両が走行中の道路で検出された特徴線の確度を学習する特徴線確度学習部と、過去に特徴線確度学習部により学習された特徴線の確度と過去に自車両が走行した道路に対して行われた学習回数とを用いて、自車両が走行する道路における車線維持支援部による支援の可否を判定する支援可否判定部とを備えるものである。 The lane keeping assist device according to the present invention assists the own vehicle so as to maintain the running in the lane by using the characteristic line which is detected from the captured image in front of the own vehicle and which defines the lane in which the own vehicle travels. The lane keeping support unit, the feature line accuracy learning unit that learns the accuracy of the feature line detected on the road on which the vehicle is traveling, the accuracy of the feature line that was previously learned by the feature line accuracy learning unit, and the A support availability determination unit that determines availability of support by the lane keeping assistance unit on the road on which the vehicle is traveling is provided by using the number of times of learning performed on the road on which the vehicle is traveling.
　この発明によれば、自車両が走行中の道路で検出された特徴線の確度を学習し、学習した特徴線の確度と学習回数とを用いて自車両が走行する道路における車線維持支援の可否を判定するようにしたので、特に自車両が常用する道路において、車線維持機能の断続的な動作を防止することができる。 According to the present invention, the accuracy of the characteristic line detected on the road on which the vehicle is traveling is learned, and whether the lane keeping support on the road on which the vehicle is traveling is performed using the accuracy of the learned characteristic line and the number of times of learning. Therefore, it is possible to prevent the intermittent operation of the lane keeping function, especially on the road that the own vehicle normally uses.
実施の形態１に係る車線維持支援システムのハードウェア構成例を示す図である。FIG. 1 is a diagram showing a hardware configuration example of a lane keeping support system according to a first embodiment. 実施の形態１に係るＥＣＵ（車線維持支援装置）の機能ブロック例を示す図である。FIG. 3 is a diagram showing an example of functional blocks of an ECU (lane keeping assistance device) according to the first embodiment. 従来の車線維持支援装置が車線維持支援を開始する車両速度に関する概念図である。It is a conceptual diagram regarding the vehicle speed at which the conventional lane keeping assistance device starts lane keeping assistance. 従来の車線維持支援装置の動作例を示すフローチャートである。It is a flow chart which shows the example of operation of the conventional lane maintenance assistance device. 従来の車線維持支援装置が車線維持支援する場合の、車線検出の概念を説明する図である。It is a figure explaining the concept of lane detection in the case where the conventional lane maintenance support device carries out lane maintenance assistance. 実施の形態１に係るＥＣＵの動作例を示すフローチャートである。3 is a flowchart showing an operation example of the ECU according to the first embodiment. 実施の形態１において自車両が走行中の道路形状を示す図である。FIG. 3 is a diagram showing a road shape on which the host vehicle is traveling in the first embodiment. 実施の形態１の画像処理装置における特徴線検出の不具合の事例を説明する図である。FIG. 6 is a diagram illustrating a case of a defect in detecting a characteristic line in the image processing apparatus according to the first embodiment. 実施の形態１の地図ＤＢに記憶されている情報の一部を模式的に表した図である。FIG. 3 is a diagram schematically showing a part of the information stored in the map DB according to the first embodiment. 実施の形態１の特徴線確度学習部による動作例を示すフローチャートである。5 is a flowchart showing an operation example by a characteristic line accuracy learning unit of the first embodiment. 実施の形態１における特徴線確度の計算式の一例を示す図である。FIG. 5 is a diagram showing an example of a calculation formula of characteristic line accuracy in the first embodiment. 実施の形態２に係るＥＣＵの機能ブロック例を示す図である。5 is a diagram showing an example of functional blocks of an ECU according to the second embodiment. FIG. 実施の形態２のナビゲーション装置が探索した探索経路の一例を示す図である。FIG. 9 is a diagram showing an example of a search route searched by the navigation device according to the second embodiment. 実施の形態２における経路特徴線確度の計算式の一例を示す図である。FIG. 10 is a diagram showing an example of a calculation formula of a route characteristic line accuracy in the second embodiment. 実施の形態２における経路特徴線確度の計算式の別の例を示す図である。FIG. 16 is a diagram showing another example of a calculation formula of route characteristic line accuracy in the second embodiment. 実施の形態２に係るＥＣＵの動作例を示すフローチャートである。7 is a flowchart showing an operation example of the ECU according to the second embodiment. 実施の形態３に係るＥＣＵの機能ブロック例を示す図である。FIG. 6 is a diagram showing an example of functional blocks of an ECU according to a third embodiment. 実施の形態３において自車両が走行中の道路形状の一例を示す図である。FIG. 16 is a diagram showing an example of a road shape on which the host vehicle is traveling in the third embodiment. 実施の形態３の走行経路予測部による経路予測方法を説明する図であり、代表的な運転操作の例を示す。It is a figure explaining the route prediction method by the traveling route prediction part of Embodiment 3, and shows an example of typical driving operation. 実施の形態３の走行経路予測部による経路予測方法を説明する図であり、代表的な運転操作ごとの係数の例を示す。FIG. 16 is a diagram for explaining a route prediction method by the travel route prediction unit of the third embodiment, showing an example of coefficients for each representative driving operation. 実施の形態３における経路特徴線確度の計算式の一例を示す図である。FIG. 16 is a diagram showing an example of a calculation formula of a route characteristic line accuracy in the third embodiment. 実施の形態３に係るＥＣＵの動作例を示すフローチャートである。7 is a flowchart showing an operation example of the ECU according to the third embodiment. 実施の形態４に係るＥＣＵの機能ブロック例を示す図である。FIG. 8 is a diagram showing an example of functional blocks of an ECU according to the fourth embodiment. 実施の形態４の地図ＤＢに記憶されている情報の一部を模式的に表した図である。It is the figure which represented typically some information memorize|stored in map DB of Embodiment 4. 実施の形態４のナビゲーション装置が探索した探索経路の一例を示す図である。It is a figure which shows an example of the search route which the navigation apparatus of Embodiment 4 searched. 図１７Ａに示される探索経路に沿った道路リンク群のリンクデータの一部を示す図である。It is a figure which shows a part of link data of the road link group along the search route shown by FIG. 17A. 実施の形態４における経路特徴線確度の計算式の一例を示す図である。FIG. 16 is a diagram showing an example of a calculation formula of route characteristic line accuracy in the fourth embodiment. 実施の形態４の環境判定部により予想された照度を説明する図である。It is a figure explaining the illuminance estimated by the environment determination part of Embodiment 4. 実施の形態４における経路特徴線確度の計算式の一例を示す図である。FIG. 16 is a diagram showing an example of a calculation formula of route characteristic line accuracy in the fourth embodiment.
　以下、この発明をより詳細に説明するために、この発明を実施するための形態について、添付の図面に従って説明する。
実施の形態１．
　図１Ａは、実施の形態１に係る車線維持支援システム１のハードウェア構成例を示す図である。車線維持支援システム１は、車両に搭載され、この車両（以下、「自車両」と称する）が車線内を走行するように支援するものである。 Hereinafter, in order to explain the present invention in more detail, modes for carrying out the present invention will be described with reference to the accompanying drawings.
Embodiment 1.
FIG. 1A is a diagram illustrating a hardware configuration example of a lane keeping support system 1 according to the first embodiment. The lane keeping assist system 1 is installed in a vehicle and assists the vehicle (hereinafter, referred to as “own vehicle”) to travel in the lane.
　電子制御ユニット１０（以下、「ＥＣＵ１０」と称する）は、自車両が車線内を走行するように支援する車線維持支援装置である。図１Ｂは、実施の形態１に係るＥＣＵ１０の機能ブロック例を示す図である。このＥＣＵ１０は、支援可否判定部１１、車線維持支援部１２、及び特徴線確度学習部１３を備える。ＥＣＵ１０の詳細は後述する。 The electronic control unit 10 (hereinafter referred to as “ECU 10”) is a lane keeping assist device that assists the host vehicle in driving in the lane. FIG. 1B is a diagram showing an example of functional blocks of the ECU 10 according to the first embodiment. The ECU 10 includes a support availability determination unit 11, a lane keeping support unit 12, and a characteristic line accuracy learning unit 13. Details of the ECU 10 will be described later.
　カメラ２０は、自車両前方の道路環境を撮像する。画像処理装置２１は、カメラ２０から入力された撮像画像を処理し、自車両が走行する車線を区画する白線等の特徴線を検出する。 The camera 20 images the road environment in front of the vehicle. The image processing device 21 processes the captured image input from the camera 20 and detects a characteristic line such as a white line that divides the lane in which the vehicle is traveling.
　アンテナ３１は、ＧＰＳ（Ｇｌｏｂａｌ　Ｐｏｓｉｔｉｏｎｉｎｇ　Ｓａｔｅｌｌｉｔｅ　Ｓｙｓｔｅｍ）衛星からの電波を受信する。ＧＰＳ受信機３０は、アンテナ３１が受信した電波を用いて、自車両の走行位置を計測する。ナビゲーション装置３２は、道路の各種情報を記憶している地図データベース３３（以下、「地図ＤＢ３３」と称する）を備える。道路の各種情報には、後述する、道路情報と、道路情報に関連付けられた特徴線確度、特徴線確度の学習回数とが含まれる。このナビゲーション装置３２は、ＧＰＳ受信機３０により計測された走行位置及び地図ＤＢ３３の情報を用いて、自車両が走行中の道路を検出する。また、ナビゲーション装置３２は、ＧＰＳ受信機３０により計測された走行位置及び地図ＤＢ３３の情報を用いて、自車両が走行する予定の経路を探索する。
　なお、図１Ａでは、地図ＤＢ３３がナビゲーション装置３２内に配置されているが、地図ＤＢ３３が車外のサーバ装置内に配置されてもよい。この場合、例えば、ナビゲーション装置３２が、無線通信を介してサーバ装置内の地図ＤＢ３３にアクセスすればよい。 The antenna 31 receives a radio wave from a GPS (Global Positioning Satellite System) satellite. The GPS receiver 30 measures the traveling position of the host vehicle using the radio waves received by the antenna 31. The navigation device 32 includes a map database 33 (hereinafter, referred to as “map DB 33”) that stores various road information. The various types of road information include road information, a characteristic line certainty associated with the road information, and the number of learnings of the characteristic line certainty, which will be described later. The navigation device 32 detects the road on which the vehicle is traveling, using the traveling position measured by the GPS receiver 30 and the information in the map DB 33. In addition, the navigation device 32 uses the information on the travel position measured by the GPS receiver 30 and the map DB 33 to search for the route on which the vehicle is scheduled to travel.
Although the map DB 33 is arranged in the navigation device 32 in FIG. 1A, the map DB 33 may be arranged in the server device outside the vehicle. In this case, for example, the navigation device 32 may access the map DB 33 in the server device via wireless communication.
　車両ネットワークＥＣＵ４０は、自車両に関する種々の情報を通信している車両ネットワーク（図示せず）との通信を行う。自車両に関する種々の情報には、自車両の走行速度、アクセル、ブレーキ及びウインカ等の運転操作、並びに自車両の周囲の照度等が含まれる。 The vehicle network ECU 40 communicates with a vehicle network (not shown) that communicates various types of information regarding the vehicle. The various kinds of information regarding the own vehicle include the traveling speed of the own vehicle, the driving operation of the accelerator, the brake, the blinker, and the like, the illuminance around the own vehicle, and the like.
　ＨＭＩ（Ｈｕｍａｎ　Ｍａｃｈｉｎｅ　Ｉｎｔｅｒｆａｃｅ）５０は、ディスプレイ又はスピーカ等の出力デバイスと、タッチパネル、ボタン又は音声認識機能等の入力デバイスとを含む。このＨＭＩ５０は、ＥＣＵ１０による車線維持支援の状態を表示又は音声出力する。また、ＨＭＩ５０は、ＥＣＵ１０に対する各種設定の入力を受け付ける。 The HMI (Human Machine Interface) 50 includes an output device such as a display or a speaker and an input device such as a touch panel, a button or a voice recognition function. The HMI 50 displays or voice outputs the state of the lane keeping support by the ECU 10. Further, the HMI 50 receives inputs of various settings to the ECU 10.
　ステアリングホイール６０は、自車両の運転者が操作する。舵角センサ６１は、ステアリングホイール６０の操作角度を検出する。トルクセンサ６２は、ステアリングホイール６０の操作力を検出する。電動回路６３は、モータ６４を制御する。モータ６４は、ステアリングホイール６０に操舵力を加える。ギヤボックス６５は、モータ６４によって発生する操舵角及び操舵力を、車輪６６に伝える。車輪６６は、操舵力を受けて自車両が走行する方向に向く。 The steering wheel 60 is operated by the driver of the vehicle. The steering angle sensor 61 detects the operation angle of the steering wheel 60. The torque sensor 62 detects the operating force of the steering wheel 60. The electric circuit 63 controls the motor 64. The motor 64 applies a steering force to the steering wheel 60. The gear box 65 transmits the steering angle and the steering force generated by the motor 64 to the wheels 66. The wheels 66 face the direction in which the host vehicle travels by receiving the steering force.
　図２は、従来の車線維持支援装置が車線維持支援を開始する車両速度に関する概念図である。図２に示されるグラフの横軸は自車両の車速を表したものであり、縦軸は操舵に要するパワー、及び走行速度の分布状況を表したものである。操舵に要するパワーは、前輪の路面との摩擦により、自車両が停止した状態で最も大きく、自車両が走行を開始し車速が上昇するにつれて低下する、右下がりの曲線となる。したがって、車線維持支援装置による支援に要するモータ６４等のパワーは、車速が上昇するにつれて減少し、車両制御面での能率が向上する。一方、実際の走行速度分布によれば、中低速での走行が多く、自動車専用道及び高速道路等における高速走行の機会が少ない。したがって、車線維持支援装置は、操舵に要するパワーと走行速度分布とのバランスから、所定の車速以上での支援が現実的である。また、所定の車速は、道路環境が整備され、走行車線の検出が容易で支援動作が不安定となる懸念が少ない、自動車専用道及び高速道路等の最低速度以上に限定されている場合が多い。 FIG. 2 is a conceptual diagram regarding the vehicle speed at which the conventional lane keeping support device starts lane keeping support. The horizontal axis of the graph shown in FIG. 2 represents the vehicle speed of the host vehicle, and the vertical axis represents the distribution of the power required for steering and the traveling speed. The power required for steering is the largest when the host vehicle is stopped due to the friction of the front wheels with the road surface, and is a downward-sloping curve that decreases as the host vehicle starts running and the vehicle speed increases. Therefore, the power of the motor 64 or the like required for the assistance by the lane keeping assist device decreases as the vehicle speed increases, and the efficiency in vehicle control improves. On the other hand, according to the actual traveling speed distribution, there are many medium-low speed travelings, and there are few opportunities for high-speed traveling on a motorway, an expressway or the like. Therefore, it is practical for the lane keeping assist device to provide assistance at a predetermined vehicle speed or higher in view of the balance between the power required for steering and the traveling speed distribution. In addition, the prescribed vehicle speed is often limited to the minimum speed or higher on a motorway or expressway, etc., where the road environment is maintained, it is easy to detect the driving lane, and there is little concern that the assisting operation will become unstable. ..
　図３は、従来の車線維持支援装置の動作例を示すフローチャートである。ステップＳＴ１において、従来の車線維持支援装置は、自車両の走行速度が予め定められた支援開始速度（例えば、６０ｋｍ／ｈ）より高いか否かを判定する。従来の車線維持支援装置は、走行速度が支援開始速度より高ければ（ステップＳＴ１“ＹＥＳ”）、ステップＳＴ２の動作を行い、低ければ（ステップＳＴ１“ＮＯ”）、そのまま速度の比較判定を続ける。ステップＳＴ２において、従来の車線維持支援装置は、ステップＳＴ１にて支援開始の条件が成立したため、車線維持支援を開始する。 FIG. 3 is a flowchart showing an operation example of a conventional lane keeping assist device. In step ST1, the conventional lane keeping assist device determines whether or not the traveling speed of the host vehicle is higher than a predetermined assist start speed (for example, 60 km/h). The conventional lane keeping assist device performs the operation of step ST2 if the traveling speed is higher than the assistance start speed (step ST1 “YES”), and continues the speed comparison determination if it is low (step ST1 “NO”). In step ST2, the conventional lane keeping assist device starts lane keeping assistance because the condition for starting assistance is satisfied in step ST1.
　図４は、従来の車線維持支援装置が車線維持支援する場合の、車線検出の概念を説明する図である。図４には、自車両１００の前方の撮像画像が示されている。この撮像画像には、自車両１００の一部に加え、特徴的構造物が写っている。特徴的構造物は、ガードレール１０１、中央分離帯１０２、車道外側線１０３，１０４、及び境界線１０５等である。車道外側線１０３，１０４及び境界線１０５は、車線を区画するものである。 FIG. 4 is a diagram for explaining the concept of lane detection when a conventional lane keeping assist device supports lane keeping. FIG. 4 shows a captured image in front of the host vehicle 100. In this captured image, a characteristic structure is shown in addition to a part of the vehicle 100. Characteristic structures are a guard rail 101, a median strip 102, roadway outside lines 103 and 104, a boundary line 105, and the like. The road outside lines 103 and 104 and the boundary line 105 divide the lane.
　ここで、従来の車線維持支援装置を説明するために、図１に示される実施の形態１に係る車線維持支援システム１を援用する。画像処理装置２１は、図４に示されるようなカメラ２０の撮像画像から上記特徴的構造物を弁別し、弁別した上記特徴的構造物のうちの自車両１００の前方に存在する車道外側線１０３と境界線１０５とから特徴線１１０を検出する。画像処理装置２１が検出した特徴線１１０の位置情報は、車線維持支援装置であるＥＣＵ１０に提供される。ＥＣＵ１０は、舵角センサ６１及びトルクセンサ６２の検出結果を参照しつつ電動回路６３に指令を与えて、特徴線１１０が自車両１００の前方に位置するように車輪６６を駆動し、車線維持支援を行う。 Here, the lane keeping support system 1 according to the first embodiment shown in FIG. 1 is referred to in order to explain a conventional lane keeping support device. The image processing device 21 discriminates the characteristic structure from the image captured by the camera 20 as shown in FIG. 4, and the roadside line 103 existing in front of the host vehicle 100 among the discriminated characteristic structures. And the characteristic line 110 is detected from the boundary line 105. The position information of the characteristic line 110 detected by the image processing device 21 is provided to the ECU 10, which is a lane keeping assist device. The ECU 10 gives a command to the electric circuit 63 while referring to the detection results of the steering angle sensor 61 and the torque sensor 62, drives the wheels 66 so that the characteristic line 110 is located in front of the host vehicle 100, and supports the lane keeping. I do.
　従来の車線維持支援装置は、以上のような動作を行うため、図４における特徴線１１０の検出が不安定になると、支援動作そのものが不可能となり、支援を停止して運転者の手動運転に委ねざるを得ない。特徴線１１０の検出には、特徴的構造物がカメラ２０で撮像されて画像処理装置２１で認識されることが必要である。このことは、整備された道路環境である自動車専用道及び高速道路等が支援の対象となるように支援開始速度が設定されることの理由の一つである。 Since the conventional lane keeping assist device performs the above-described operation, if the detection of the characteristic line 110 in FIG. 4 becomes unstable, the assist operation itself becomes impossible, and the assist is stopped to allow the driver to manually drive the vehicle. I have to entrust it. To detect the characteristic line 110, it is necessary that the characteristic structure is captured by the camera 20 and recognized by the image processing device 21. This is one of the reasons why the support start speed is set so that the roads for exclusive use of vehicles, expressways, etc., which are maintained road environments, are targeted for support.
　このように、従来の車線維持支援装置は、車線検出環境が良いと推定される道路だけで支援することを想定して構成されている。しかしながら、実際の環境では、一般道路であっても、支援開始速度を超える速度での走行が可能であり、かつ、支援に十分な車線検出環境が整えられている道路が存在する。一方、車線維持支援装置が実際の道路環境だけに頼った場合、支援に十分な環境とそうでない環境とが混在する道路を自車両１００が走行すると、支援が断続的になり、運転者にとって不快かつ信頼性の低い動作となる。 In this way, the conventional lane keeping support device is configured to support only on roads where the lane detection environment is presumed to be good. However, in an actual environment, there are roads that are capable of traveling at a speed exceeding the support start speed even in general roads, and that have a lane detection environment sufficient for support. On the other hand, when the lane keeping assist device relies only on the actual road environment, when the vehicle 100 travels on a road in which an environment sufficient for assistance and an environment not for assistance coexist, the assistance becomes intermittent and uncomfortable for the driver. And the operation becomes unreliable.
　そこで、実施の形態１に係る車線維持支援システム１は、自車両１００の走行履歴に関連付けて、実際の車線維持支援の動作状況を地図ＤＢ３３に記憶しておくことで、整備された環境以外でも車線維持支援の動作機会を増やし、運転者の不快感を抑制すると共に利便性を向上させる。 Therefore, the lane keeping support system 1 according to the first embodiment stores the actual operation state of the lane keeping support in the map DB 33 in association with the traveling history of the vehicle 100, so that the lane keeping assist system 1 can be used even in a non-maintained environment. The number of operation opportunities for lane keeping support is increased, the discomfort of the driver is suppressed, and the convenience is improved.
　次に、ＥＣＵ１０の詳細を説明する。
　図５は、実施の形態１に係るＥＣＵ１０の動作例を示すフローチャートである。ＥＣＵ１０は、例えば、自車両１００のイグニッションスイッチがオンになると図５のフローチャートに示される動作を開始し、イグニッションスイッチがオフになると動作を終了する。 Next, details of the ECU 10 will be described.
FIG. 5 is a flowchart showing an operation example of the ECU 10 according to the first embodiment. For example, the ECU 10 starts the operation shown in the flowchart of FIG. 5 when the ignition switch of the own vehicle 100 is turned on, and ends the operation when the ignition switch is turned off.
　ステップＳＴ１１において、支援可否判定部１１は、車両ネットワークＥＣＵ４０から、又はナビゲーション装置３２を介して車両ネットワークＥＣＵ４０から、自車両１００の走行速度を取得する。そして、支援可否判定部１１は、走行速度が予め定められた支援開始速度より高いか否かを判定する。支援開始速度は、図２に示された操舵に要するパワーとの関係を考慮した速度、又は整備された走行環境を想定した速度（例えば、６０ｋｍ／ｈ）に設定されてもよいし、より低速な固定値でもよいし、走行道路又は走行地域に応じた可変値でもよい。走行道路は、一般道路、自動車専用道、及び高速道路等である。走行地域とは、歩行者及び自転車等の走行の障害となる外乱が少ない郊外、及び外乱が多い都会の市街地等である。走行道路及び走行地域の情報は、地図ＤＢ３３に記憶されている。走行速度が支援開始速度より高い場合（ステップＳＴ１１“ＹＥＳ”）、支援可否判定部１１は、ステップＳＴ１２の動作を行う。一方、走行速度が支援開始速度以下である場合（ステップＳＴ１１“ＮＯ”）、支援可否判定部１１は、車線維持支援に適していないと判断して、ステップＳＴ１８の動作を行う。 In step ST11, the support availability determination unit 11 acquires the traveling speed of the host vehicle 100 from the vehicle network ECU 40 or the vehicle network ECU 40 via the navigation device 32. Then, the supportability determination unit 11 determines whether the traveling speed is higher than a predetermined support start speed. The support start speed may be set to a speed that considers the relationship with the power required for steering shown in FIG. 2, or a speed that assumes a maintained traveling environment (for example, 60 km/h), or a lower speed. It may be a fixed value or a variable value according to the traveling road or traveling area. The traveling road is an ordinary road, a motorway, a highway, or the like. The running area is a suburb where disturbances such as pedestrians and bicycles obstruct traveling are small, and an urban area where there is much disturbance. The information on the traveling road and the traveling area is stored in the map DB 33. When the traveling speed is higher than the support start speed (step ST11 “YES”), the support availability determination unit 11 performs the operation of step ST12. On the other hand, when the traveling speed is equal to or lower than the support start speed (step ST11 “NO”), the support availability determination unit 11 determines that the support is not suitable for the lane keeping support, and performs the operation of step ST18.
　ステップＳＴ１２において、ナビゲーション装置３２は、ＧＰＳ受信機３０により計測された走行位置、及び地図ＤＢ３３の情報を用いて、自車両１００が走行中の道路（以下、「走行中道路」と称する）を検出する。そして、ナビゲーション装置３２は、検出した走行中道路を示す情報（例えば、後述する図８Ａに示されるリンクデータ）を地図ＤＢ３３から取得して支援可否判定部１１へ出力する。支援可否判定部１１は、ナビゲーション装置３２から走行中道路を示す情報を取得する。 In step ST12, the navigation device 32 uses the traveling position measured by the GPS receiver 30 and the information in the map DB 33 to detect the road on which the vehicle 100 is traveling (hereinafter, referred to as "traveling road"). To do. Then, the navigation device 32 acquires the information indicating the detected running road (for example, the link data shown in FIG. 8A described later) from the map DB 33 and outputs the information to the supportability determination unit 11. The support availability determination unit 11 acquires information indicating the road on which the vehicle is traveling from the navigation device 32.
　ステップＳＴ１３において、支援可否判定部１１は、ナビゲーション装置３２から取得した走行中道路を示す情報を用いて、当該走行中道路を示す情報に関連付けて記憶されている特徴線確度及び学習回数を、地図ＤＢ３３から取得する。 In step ST13, the support availability determination unit 11 uses the information indicating the traveling road acquired from the navigation device 32 to determine the characteristic line accuracy and the learning count stored in association with the information indicating the traveling road on the map. Obtain from DB33.
　ステップＳＴ１４において、支援可否判定部１１は、自車両１００の走行中道路に対して過去に地図ＤＢ３３に記憶された学習回数を用いて、走行中道路が自車両１００の運転者が常用する道路か否かを判定する。具体的には、支援可否判定部１１は、地図ＤＢ３３から取得した学習回数が予め定められた閾値以下である場合（ステップＳＴ１４“ＮＯ”）、走行中道路が運転者の常用する道路ではないと判断して、ステップＳＴ１８の動作を行う。一方、学習回数が閾値より大きい場合（ステップＳＴ１４“ＹＥＳ”）、支援可否判定部１１は、走行中道路が運転者の常用する道路であると判断して、ステップＳＴ１５の動作を行う。 In step ST14, the support availability determination unit 11 uses the learning count previously stored in the map DB 33 for the traveling road of the host vehicle 100 to determine whether the traveling road is a road that is regularly used by the driver of the own vehicle 100. Determine whether or not. Specifically, when the number of times of learning acquired from the map DB 33 is equal to or less than a predetermined threshold value (step ST14 “NO”), the support availability determination unit 11 determines that the running road is not the road that the driver regularly uses. After making a determination, the operation of step ST18 is performed. On the other hand, when the learning count is larger than the threshold value (step ST14 “YES”), the supportability determination part 11 determines that the traveling road is a road that is regularly used by the driver, and performs the operation of step ST15.
　ステップＳＴ１５において、支援可否判定部１１は、自車両１００の走行中道路に対して過去に地図ＤＢ３３に記憶された特徴線確度を用いて、車線維持支援の可否を判定する。具体的には、支援可否判定部１１は、地図ＤＢ３３から取得した特徴線確度が予め定められた閾値以下である場合（ステップＳＴ１５“ＮＯ”）、走行中道路が車線維持支援に適していない道路であると判断して、ステップＳＴ１８の動作を行う。一方、特徴線確度が閾値より大きい場合（ステップＳＴ１５“ＹＥＳ”）、支援可否判定部１１は、走行中道路が車線維持支援に適した道路であると判断して、ステップＳＴ１６の動作を行う。 In step ST15, the support availability determination unit 11 determines the availability of lane keeping support for the road on which the vehicle 100 is running using the characteristic line accuracy stored in the map DB 33 in the past. Specifically, when the characteristic line accuracy acquired from the map DB 33 is less than or equal to a predetermined threshold value (step ST15 “NO”), the support availability determination unit 11 determines that the running road is not suitable for lane keeping support. Then, the operation of step ST18 is performed. On the other hand, when the characteristic line accuracy is greater than the threshold value (step ST15 “YES”), the support availability determination unit 11 determines that the traveling road is a road suitable for lane keeping support, and performs the operation of step ST16.
　ステップＳＴ１６において、支援可否判定部１１は、車線維持支援部１２に対して車線維持支援を開始するように指示する。この指示を受けた車線維持支援部１２は、画像処理装置２１が検出する特徴線１１０を、画像処理装置２１から取得する。そして、車線維持支援部１２は、取得した特徴線１１０により区画される車線内を自車両１００が走行するように、舵角センサ６１及びトルクセンサ６２の検出結果を参照しつつ電動回路６３に指令を与える。車線維持支援部１２による車線維持支援方法は、周知の技術を用いればよいため、詳細な説明を省略する。 In step ST16, the support availability determination unit 11 instructs the lane keeping support unit 12 to start lane keeping support. Upon receiving this instruction, the lane keeping assisting unit 12 acquires the characteristic line 110 detected by the image processing device 21 from the image processing device 21. Then, the lane keeping assisting unit 12 commands the electric circuit 63 to refer to the detection results of the steering angle sensor 61 and the torque sensor 62 so that the host vehicle 100 travels in the lane divided by the acquired characteristic line 110. give. A well-known technique may be used for the lane keeping assisting method by the lane keeping assisting unit 12, and detailed description thereof will be omitted.
　ステップＳＴ１７において、支援可否判定部１１は、特徴線確度学習部１３に対して特徴線確度を学習するように指示する。この指示を受けた特徴線確度学習部１３は、自車両１００が走行済みの車線で検出された特徴線１１０の確度を学習し、学習結果を地図ＤＢ３３に記憶させる。ステップＳＴ１７における詳細な動作は、後述する。 In step ST17, the support availability determination unit 11 instructs the characteristic line accuracy learning unit 13 to learn the characteristic line accuracy. Upon receiving this instruction, the characteristic line accuracy learning unit 13 learns the accuracy of the characteristic line 110 detected in the lane in which the vehicle 100 has already traveled, and stores the learning result in the map DB 33. The detailed operation in step ST17 will be described later.
　ステップＳＴ１８において、支援可否判定部１１は、走行中道路が車線維持支援に適していない道路であると判断したため、車線維持支援部１２に対して車線維持支援を停止するように指示する。この指示を受けた車線維持支援部１２は、車線維持支援を停止する。 In step ST18, the support availability determination unit 11 determines that the running road is not suitable for lane keeping support, and therefore instructs the lane keeping support unit 12 to stop the lane keeping support. The lane keeping support unit 12 receiving this instruction stops the lane keeping support.
　なお、車線維持支援部１２は、ステップＳＴ１６において車線維持支援を開始する前に、ＨＭＩ５０に対して車線維持支援の開始を運転者に通知するように指示してもよい。また、車線維持支援部１２は、ステップＳＴ１８において車線維持支援を停止する前に、ＨＭＩ５０に対して車線維持支援の停止を運転者に通知するように指示してもよい。ＨＭＩ５０は、例えば、地図画面での道路リンクに沿った線状画像の表示、車線維持支援の動作状態を表すアイコンの表示、又は、メータパネル若しくはヘッドアップディスプレイ等における警報表示を行う。 Note that the lane keeping support unit 12 may instruct the HMI 50 to notify the driver of the start of the lane keeping support before starting the lane keeping support in step ST16. Further, the lane keeping support unit 12 may instruct the HMI 50 to notify the driver of the stop of the lane keeping support before stopping the lane keeping support in step ST18. The HMI 50 displays, for example, a linear image along a road link on a map screen, an icon indicating an operation state of lane keeping support, or an alarm display on a meter panel or a head-up display.
　次に、特徴線確度学習部１３による特徴線確度の学習例を説明する。
　図６は、実施の形態１において自車両１００が走行中の道路形状を示す図である。図６において、道路リンクＬ１＿１，Ｌ１＿２，Ｌ１＿３，Ｌ１＿４，Ｌ２＿１，Ｌ３＿１、Ｌ４＿１，Ｌ５＿１，Ｌ５＿２は、道路形状を線分で近似したものである。ノードＮ１＿１，Ｎ１＿２，Ｎ１＿３，Ｎ１＿４，Ｎ１＿５，Ｎ２＿１，Ｎ３＿１，Ｎ４＿１，Ｎ５＿１，Ｎ５＿２は、道路リンク同士を接続する屈曲点又は交差点等である。自車両１００は、道路リンクのいずれかを走行中である。ナビゲーション装置３２は、自車両１００がいずれかの道路リンク上に存在するとみなし、屈曲点又は交差点等における自車両１００の方向転換をいずれかのノード上で行われるものとみなして、自車両１００の走行位置及び走行中道路を検出する。そのため、ナビゲーション装置３２は、走行中道路に対応する道路リンクのリンクデータを、支援可否判定部１１へ出力することになる（図５のステップＳＴ１２参照）。図６の例では、自車両１００が道路リンクＬ１＿２を走行中であるため、ナビゲーション装置３２は、ノードＮ１＿２とノードＮ１＿３との間の道路リンクＬ１＿２のリンクデータを、支援可否判定部１１へ出力する。 Next, an example of learning the characteristic line certainty by the characteristic line certainty learning unit 13 will be described.
FIG. 6 is a diagram showing a road shape on which the host vehicle 100 is traveling in the first embodiment. In FIG. 6, road links L1_1, L1_2, L1_3, L1_4, L2_1, L3_1, L4_1, L5_1, L5_2 are road shapes approximated by line segments. The nodes N1_1, N1_2, N1_3, N1_4, N1_5, N2_1, N3_1, N4_1, N5_1, N5_2 are bending points or intersections connecting road links. The host vehicle 100 is traveling on one of the road links. The navigation device 32 considers that the own vehicle 100 exists on any road link, considers that the direction change of the own vehicle 100 at a turning point or an intersection is performed on any node, and determines that Detects the running position and the running road. Therefore, the navigation device 32 outputs the link data of the road link corresponding to the traveling road to the support availability determination unit 11 (see step ST12 in FIG. 5). In the example of FIG. 6, since the vehicle 100 is traveling on the road link L1_2, the navigation device 32 outputs the link data of the road link L1_2 between the node N1_2 and the node N1_3 to the support availability determination unit 11. ..
　図７は、実施の形態１の画像処理装置２１における特徴線検出の不具合の事例を説明する図である。図７に示される環境を自車両１００が走行中である場合、画像処理装置２１は、図４で説明したように自車両１００の左右に存在する白線等の特徴線１１０を検出する。しかしながら、多数のはみ出し車両の走行による白線のかすれ１２１、及び駐車場入り口等があることによる白線の途切れ１２２等が、一般道路では存在することがある。このような状況では、画像処理装置２１は、自車両１００の左右に存在する白線等の特徴線１１０を検出できない。すると、車線維持支援部１２による車線維持支援のための情報が不確定となり、車線維持支援が断続的に停止する。そのため、従来の車線維持支援装置では、このような車線維持支援の断続的な停止が、この道路を走行するたびに発生するため、運転者にとって不快かつ信頼性の低い動作となる。これに対し、実施の形態１に係るＥＣＵ１０は、図７のような道路を自車両１００が走行する際に、走行中道路を示す情報と特徴線確度等とを関連付けて地図ＤＢ３３に記憶させておく。具体的には、ＥＣＵ１０は、走行中道路に相当する道路リンクのリンクデータと、特徴線確度と、特徴線確度の学習回数とを関連付けて地図ＤＢ３３に記憶させておく。そして、ＥＣＵ１０は、その後に、自車両１００が該当の道路リンクの走行を開始すると、地図ＤＢ３３に記憶済みの特徴線１１０の検出状況に基づいて、車線維持支援が断続的に停止する道路では車線維持支援を行わず、車線維持支援が断続的に停止しない道路では車線維持支援を行う。これにより、車線維持支援の断続的な停止による不快感及び信頼性低下を防止する。 FIG. 7 is a diagram illustrating an example of a defect in feature line detection in the image processing device 21 according to the first embodiment. When the host vehicle 100 is traveling in the environment shown in FIG. 7, the image processing device 21 detects the characteristic lines 110 such as white lines existing on the left and right of the host vehicle 100 as described with reference to FIG. However, there may be white line blurs 121 due to the traveling of a large number of protruding vehicles, and white line breaks 122 due to the entrance of a parking lot or the like. In such a situation, the image processing device 21 cannot detect the characteristic lines 110 such as white lines existing on the left and right of the host vehicle 100. Then, the information for the lane keeping support by the lane keeping support unit 12 becomes uncertain, and the lane keeping support stops intermittently. Therefore, in the conventional lane keeping assist device, such an intermittent stop of the lane keeping assist occurs each time the vehicle travels on this road, which is an uncomfortable and unreliable operation for the driver. On the other hand, when the host vehicle 100 travels on a road as shown in FIG. 7, the ECU 10 according to the first embodiment stores the information indicating the running road and the characteristic line accuracy in the map DB 33 in association with each other. deep. Specifically, the ECU 10 stores the link data of the road link corresponding to the road on which the vehicle is running, the characteristic line certainty, and the learning number of the characteristic line certainty in association with each other in the map DB 33. Then, when the host vehicle 100 subsequently starts traveling on the relevant road link, the ECU 10 lanes on the lanes where the lane keeping assistance is intermittently stopped based on the detection status of the characteristic line 110 stored in the map DB 33. Lane maintenance support will be provided on roads where lane maintenance support does not stop intermittently without maintenance support. This prevents discomfort and a decrease in reliability due to intermittent lane keeping support.
　図８Ａは、実施の形態１の地図ＤＢ３３に記憶されている情報の一部を模式的に表した図である。一つのリンクデータは、その道路リンク自体のＩＤ（この例の場合はＬ１＿２）、及びその道路リンクの始点と終点を表すノードの情報を最低限含み、それ以外に種々の属性を含み得る。属性の例としては、制限速度、及び道路種別（一般道、有料道路、及び高速道路等）等がある。さらに、実施の形態１では、道路リンクの属性として、この道路リンクの特徴線確度を学習した回数と、最新の学習結果としての特徴線確度とが追加されている。 FIG. 8A is a diagram schematically showing a part of the information stored in the map DB 33 according to the first embodiment. One piece of link data includes at least the ID of the road link itself (L1_2 in this example), the information of the node indicating the start point and the end point of the road link, and may include other various attributes. Examples of attributes include speed limits and road types (general roads, toll roads, expressways, etc.). Furthermore, in the first embodiment, the number of times the characteristic line certainty of this road link is learned and the characteristic line certainty as the latest learning result are added as the attributes of the road link.
　図８Ｂは、実施の形態１の特徴線確度学習部１３による動作例を示すフローチャートである。特徴線確度学習部１３は、図５のステップＳＴ１７において、図８ＢのステップＳＴ１７－１～ＳＴ１７－７の動作を行う。 FIG. 8B is a flowchart showing an operation example by the characteristic line accuracy learning unit 13 of the first embodiment. The characteristic line accuracy learning unit 13 performs the operations of steps ST17-1 to ST17-7 of FIG. 8B in step ST17 of FIG.
　ステップＳＴ１７－１において、ナビゲーション装置３２は、ＧＰＳ受信機３０により計測された走行位置及び地図ＤＢ３３の情報を用いて、自車両１００の走行中道路を検出する。そして、特徴線確度学習部１３は、ナビゲーション装置３２が検出した走行中道路に関連付けて記憶されている道路リンクのリンクデータ（例えば、図８Ａに示されるリンクデータ）を、地図ＤＢ３３から取得する。 In step ST17-1, the navigation device 32 detects the road on which the vehicle 100 is traveling, using the traveling position measured by the GPS receiver 30 and the information in the map DB 33. Then, the characteristic line accuracy learning unit 13 acquires from the map DB 33 the link data of the road link (for example, the link data shown in FIG. 8A) stored in association with the traveling road detected by the navigation device 32.
　ステップＳＴ１７－２において、特徴線確度学習部１３は、画像処理装置２１が検出する特徴線１１０を、画像処理装置２１から取得する。そして、特徴線確度学習部１３は、画像処理装置２１において特徴線１１０の検出ができたか否かを判定する。画像処理装置２１により特徴線１１０が検出できた場合（ステップＳＴ１７－２“ＹＥＳ”）、ステップＳＴ１７－３において、特徴線確度学習部１３は、走行中の道路リンクにおいて特徴線１１０が検出できた区間の長さである検出長を更新する。一方、画像処理装置２１により特徴線１１０が検出できなかった場合（ステップＳＴ１７－２“ＮＯ”）、特徴線確度学習部１３は、ステップＳＴ１７－３の動作をスキップする。 In step ST17-2, the characteristic line accuracy learning unit 13 acquires the characteristic line 110 detected by the image processing device 21 from the image processing device 21. Then, the characteristic line accuracy learning unit 13 determines whether or not the characteristic line 110 has been detected by the image processing device 21. When the characteristic line 110 can be detected by the image processing device 21 (step ST17-2 “YES”), the characteristic line accuracy learning unit 13 can detect the characteristic line 110 in the running road link in step ST17-3. The detection length, which is the length of the section, is updated. On the other hand, when the characteristic line 110 cannot be detected by the image processing device 21 (step ST17-2 “NO”), the characteristic line accuracy learning unit 13 skips the operation of step ST17-3.
　ステップＳＴ１７－４において、ナビゲーション装置３２は、ＧＰＳ受信機３０により計測された走行位置及び地図ＤＢ３３の情報を用いて、自車両１００の走行中道路を検出する。そして、特徴線確度学習部１３は、ナビゲーション装置３２から走行中道路を示す情報を取得し、自車両１００が、ステップＳＴ１７－１で取得したリンクデータが示す道路リンクから退出したか否かを判定する。自車両１００が道路リンクから退出した場合（ステップＳＴ１７－４“ＹＥＳ”）、特徴線確度学習部１３は、ステップＳＴ１７－５の動作を行う。一方、自車両１００が道路リンクから退出していなければ（ステップＳＴ１７－４“ＮＯ”）、特徴線確度学習部１３は、ステップＳＴ１７－２へ戻って特徴線１１０の検出を続ける。 In step ST17-4, the navigation device 32 detects the road on which the vehicle 100 is traveling, using the traveling position measured by the GPS receiver 30 and the information in the map DB 33. Then, the characteristic line accuracy learning unit 13 acquires information indicating the traveling road from the navigation device 32, and determines whether the own vehicle 100 has exited from the road link indicated by the link data acquired in step ST17-1. To do. When the host vehicle 100 exits from the road link (step ST17-4 “YES”), the characteristic line accuracy learning unit 13 performs the operation of step ST17-5. On the other hand, if the host vehicle 100 has not left the road link (step ST17-4 "NO"), the characteristic line accuracy learning unit 13 returns to step ST17-2 and continues to detect the characteristic line 110.
　ステップＳＴ１７－５において、特徴線確度学習部１３は、地図ＤＢ３３に記憶されている、退出した道路リンクのリンクデータに含まれる学習回数に「１」を加算することによって、学習回数を更新する。 In step ST17-5, the characteristic line accuracy learning unit 13 updates the learning count by adding “1” to the learning count included in the link data of the exiting road link stored in the map DB 33.
　ステップＳＴ１７－６において、特徴線確度学習部１３は、退出した道路リンクについて特徴線確度を計算する。そして、特徴線確度学習部１３は、計算した特徴線確度を地図ＤＢ３３の該当するリンクデータの特徴線確度に上書きすることによって特徴線確度を更新する。 In step ST17-6, the characteristic line certainty learning unit 13 calculates the characteristic line certainty for the exiting road link. Then, the characteristic line certainty degree learning unit 13 updates the characteristic line certainty degree by overwriting the calculated characteristic line certainty degree on the characteristic line certainty degree of the corresponding link data in the map DB 33.
　図８Ｃは、実施の形態１における特徴線確度の計算式の一例を示す図である。特徴線確度学習部１３は、ある道路リンク内で累積した検出長を、その道路リンク内の累積走行距離で除して、特徴線確度を計算する。または、特徴線確度学習部１３は、ある道路リンク内で累積した検出長を、その道路リンクの長さと学習回数とを乗じた値で除して、特徴線確度を計算してもよい。 FIG. 8C is a diagram showing an example of a formula for calculating the characteristic line accuracy in the first embodiment. The characteristic line accuracy learning unit 13 calculates the characteristic line accuracy by dividing the detection length accumulated in a certain road link by the cumulative traveling distance in the road link. Alternatively, the characteristic line accuracy learning unit 13 may calculate the characteristic line accuracy by dividing the detection length accumulated in a certain road link by a value obtained by multiplying the length of the road link and the number of times of learning.
　特徴線確度は、ある道路リンク内での車線維持支援動作の安定性に関係し、特徴線確度が高いほど安定した車線維持支援のための条件がそろっていると言える。つまり、特徴線確度が高いほど、車線維持支援が連続的に動作し、特徴線確度が低いほど、車線維持支援が断続的に動作すると言える。 The characteristic line accuracy relates to the stability of the lane keeping support operation within a certain road link, and it can be said that the higher the characteristic line accuracy, the more stable the lane keeping support conditions are. That is, it can be said that the higher the characteristic line accuracy, the more continuously the lane keeping support operates, and the lower the characteristic line accuracy, the more intermittently the lane keeping support operates.
　ステップＳＴ１７－７において、支援可否判定部１１は、図５のステップＳＴ１１と同様に、現在の走行速度が支援開始速度より高いか否かを判定する。走行速度が支援開始速度より高い場合（ステップＳＴ１７－７“ＹＥＳ”）、支援可否判定部１１は、ステップＳＴ１７－１の動作を行うように特徴線確度学習部１３に指示する。一方、走行速度が支援開始速度以下である場合（ステップＳＴ１７－７“ＮＯ”）、支援可否判定部１１は、特徴線確度の学習を停止するように特徴線確度学習部１３に指示する。そして、処理は、図５のステップＳＴ１８へ進む。 In step ST17-7, the support availability determination part 11 determines whether or not the current traveling speed is higher than the support start speed, as in step ST11 of FIG. When the traveling speed is higher than the support start speed (step ST17-7 “YES”), the support availability determination unit 11 instructs the characteristic line accuracy learning unit 13 to perform the operation of step ST17-1. On the other hand, when the traveling speed is equal to or lower than the support start speed (step ST17-7 “NO”), the support availability determination unit 11 instructs the characteristic line accuracy learning unit 13 to stop the learning of the characteristic line accuracy. Then, the process proceeds to step ST18 of FIG.
　以上のように、実施の形態１に係るＥＣＵ１０は、車線維持支援部１２、特徴線確度学習部１３、及び支援可否判定部１１を備える。車線維持支援部１２は、自車両１００の前方の撮像画像から検出された、自車両１００が走行する車線を区画する特徴線１１０を用いて、車線内の走行を維持するように自車両１００を支援する。特徴線確度学習部１３は、自車両１００が走行中の道路で検出された特徴線１１０の確度を学習する。支援可否判定部１１は、過去に特徴線確度学習部１３により学習された特徴線１１０の確度と過去に自車両１００が走行した道路に対して行われた学習回数とを用いて、自車両１００が走行する道路における車線維持支援部１２による支援の可否を判定する。このように、ＥＣＵ１０は、自車両１００の走行道路と特徴線確度とを関連付けて学習することで、過去に車線維持支援の断続動作があった道路では、車線維持支援を行わず、断続動作を防止することができる。また、ＥＣＵ１０は、特徴線確度の学習回数を用いることによって、自車両１００がこれから走行する道路が、自車両１００の常用する道路か否かを判定することができる。したがって、ＥＣＵ１０は、特に自車両１００が常用する道路において、車線維持機能の断続的な動作を防止することができる。また、ＥＣＵ１０は、安定し信頼に足る車線維持支援を提供することができる。 As described above, the ECU 10 according to the first embodiment includes the lane keeping support unit 12, the characteristic line accuracy learning unit 13, and the support availability determination unit 11. The lane keeping assisting unit 12 uses the characteristic line 110, which is detected from the captured image in front of the own vehicle 100 and divides the lane in which the own vehicle 100 travels, so as to keep the own vehicle 100 running within the lane. Assist. The characteristic line accuracy learning unit 13 learns the accuracy of the characteristic line 110 detected on the road on which the vehicle 100 is traveling. The support availability determination unit 11 uses the accuracy of the characteristic line 110 learned by the characteristic line accuracy learning unit 13 in the past and the number of times of learning performed on the road on which the own vehicle 100 has traveled in the past, to determine the own vehicle 100. It is determined whether or not the lane keeping support unit 12 can support the road on which the vehicle runs. In this way, the ECU 10 learns by associating the traveling road of the host vehicle 100 with the characteristic line accuracy, so that the lane keeping support is not performed on the road where the lane keeping support has been intermittently performed in the past. Can be prevented. Further, the ECU 10 can determine whether or not the road on which the vehicle 100 is about to travel is a road that the vehicle 100 normally uses, by using the number of times of learning the characteristic line accuracy. Therefore, the ECU 10 can prevent the intermittent operation of the lane keeping function, especially on the road that the vehicle 100 normally uses. Further, the ECU 10 can provide stable and reliable lane keeping support.
実施の形態２．
　実施の形態１に係るＥＣＵ１０は、道路リンクごとに車線維持支援の可否を判定する構成である。しかしながら、道路リンク長が短くなるようにデータ化された道路、交差点等が連続して存在するような道路、又は特徴線確度が高い道路と低い道路とが交互に連続して存在するような道路では、車線維持支援の可否が断続的に切り替わり、結果として運転者に不快感を与える。そこで、実施の形態２に係るＥＣＵ１０は、ナビゲーション装置３２により探索された探索経路全体における車線維持支援の可否を判定する構成とする。 Embodiment 2.
The ECU 10 according to the first embodiment is configured to determine whether or not lane keeping support is possible for each road link. However, roads that are dataized so that the road link length is shortened, roads in which intersections and the like exist continuously, or roads in which high and low characteristic line accuracy is alternately and continuously present. Then, whether or not the lane keeping support is intermittently switched, and as a result, the driver feels uncomfortable. Therefore, the ECU 10 according to the second embodiment is configured to determine whether or not lane keeping assistance is possible for the entire search route searched by the navigation device 32.
　図９は、実施の形態２に係るＥＣＵ１０の機能ブロック例を示す図である。実施の形態２に係るＥＣＵ１０は、図１Ｂに示された実施の形態１のＥＣＵ１０における支援可否判定部１１に代えて支援可否判定部１１ａを備える構成である。図９において図１Ａ及び図１Ｂと同一又は相当する部分は、同一の符号を付し説明を省略する。 FIG. 9 is a diagram showing an example of functional blocks of the ECU 10 according to the second embodiment. The ECU 10 according to the second embodiment has a configuration including a support availability determination unit 11a instead of the support availability determination unit 11 in the ECU 10 of the first embodiment shown in FIG. 1B. In FIG. 9, parts that are the same as or correspond to those in FIGS. 1A and 1B are assigned the same reference numerals and explanations thereof are omitted.
　支援可否判定部１１ａは、過去に特徴線確度学習部１３により学習された特徴線１１０の確度を用いて、ナビゲーション装置３２により探索された探索経路における車線維持支援部１２による支援の可否を判定する。 The support availability determination unit 11a uses the accuracy of the characteristic line 110 learned by the characteristic line accuracy learning unit 13 in the past to determine whether the lane keeping assistance unit 12 can assist the search route searched by the navigation device 32. ..
　図１０Ａは、実施の形態２のナビゲーション装置３２が探索した探索経路の一例を示す図である。図１０Ａに示される探索経路は、出発地ＯであるノードＮ１＿１と目的地ＤであるノードＮ１＿５とを結ぶ道路リンク群Ｌ１＿１，Ｌ１＿２，Ｌ１＿３，Ｌ１＿４で表される。 FIG. 10A is a diagram showing an example of a search route searched by the navigation device 32 according to the second embodiment. The search route shown in FIG. 10A is represented by road link groups L1_1, L1_2, L1_3, and L1_4 that connect the node N1_1 that is the departure point O and the node N1_5 that is the destination D.
　図１０Ｂは、実施の形態２における経路特徴線確度の計算式の一例を示す図である。支援可否判定部１１ａは、出発地Ｏ又は走行位置であるノードＮ１＿１と目的地ＤであるノードＮ１＿５とを結ぶ探索経路全体での特徴線確度（以下、「経路特徴線確度」と称する）を計算する。図１０Ｂの計算式では、探索経路に含まれる道路リンク（Ｌｘ＿ｙ）に関連付けて地図ＤＢ３３に記憶された特徴線確度を積算することにより、経路特徴線確度が得られる。 FIG. 10B is a diagram showing an example of a calculation formula of route characteristic line accuracy in the second embodiment. The support availability determination unit 11a calculates the characteristic line accuracy (hereinafter, referred to as "route characteristic line accuracy") in the entire search route connecting the node N1_1 which is the starting point O or the traveling position and the node N1_5 which is the destination D. To do. In the calculation formula of FIG. 10B, the route characteristic line certainty is obtained by integrating the characteristic line certainty stored in the map DB 33 in association with the road link (Lx_y) included in the searched route.
　図１０Ｃは、実施の形態２における経路特徴線確度の計算式の別の例を示す図である。図１０Ｃの計算式では、支援可否判定部１１ａは、出発地Ｏからの距離ｎに応じて値が漸減する関数α（ｎ）と、探索経路に含まれる道路リンク（Ｌｘ＿ｙ）に関連付けて地図ＤＢ３３に記憶された特徴線確度との積和演算を行う。出発地Ｏからの距離ｎは、物理的距離でもよいし、探索経路において何番目の道路リンクかを表す数でもよい。 FIG. 10C is a diagram showing another example of the calculation formula of the route characteristic line accuracy in the second embodiment. In the calculation formula of FIG. 10C, the support availability determination unit 11a associates the function α(n) whose value gradually decreases with the distance n from the departure point O with the road link (Lx_y) included in the searched route, and associates it with the map DB 33. The product-sum operation is performed with the feature line accuracy stored in. The distance n from the departure place O may be a physical distance or a number representing the number of the road link in the search route.
　図１０Ｂの計算式では実質的に探索経路全体に同一の重みが設定されているのに対し、図１０Ｃの計算式では途中の経路変更等を考慮して、出発地Ｏから遠いほど経路変更等の影響が少なくなるような重みが設定されている。なお、支援可否判定部１１ａは、図１０Ｃの計算式を用いて経路特徴線確度を計算する場合、自車両１００の走行中に順次経路特徴線確度を再計算することが好ましい。 In the calculation formula of FIG. 10B, substantially the same weight is set for the entire search route, whereas in the calculation formula of FIG. The weight is set so that the effect of is lessened. When calculating the route characteristic line accuracy using the calculation formula of FIG. 10C, the supportability determination unit 11a preferably sequentially recalculates the route characteristic line accuracy while the host vehicle 100 is traveling.
　図１１は、実施の形態２に係るＥＣＵ１０の動作例を示すフローチャートである。図１１のステップＳＴ１１～ＳＴ１８における動作は、図５のステップＳＴ１１～ＳＴ１８における動作と同じである。 FIG. 11 is a flowchart showing an operation example of the ECU 10 according to the second embodiment. The operation in steps ST11 to ST18 in FIG. 11 is the same as the operation in steps ST11 to ST18 in FIG.
　支援可否判定部１１ａは、走行速度が支援開始速度より高い場合（ステップＳＴ１１“ＹＥＳ”）、ステップＳＴ２１において、ナビゲーション装置３２に探索経路が存在するか否かを判定する。探索経路が存在する場合（ステップＳＴ２１“ＹＥＳ”）、ステップＳＴ２２において、支援可否判定部１１ａは、探索経路に沿った走行予定の道路リンク群のリンクデータを地図ＤＢ３３から取得する。一方、探索経路が存在しない場合（ステップＳＴ２１“ＮＯ”）、支援可否判定部１１ａは、ステップＳＴ２２～ＳＴ２４の動作をスキップする。 When the traveling speed is higher than the support start speed (step ST11 “YES”), the support availability determination unit 11a determines whether or not the searched route exists in the navigation device 32 in step ST21. When the searched route exists (step ST21 “YES”), in step ST22, the support availability determination part 11a acquires the link data of the road link group scheduled to travel along the searched route from the map DB 33. On the other hand, when there is no searched route (step ST21 “NO”), the supportability determination part 11a skips the operations of steps ST22 to ST24.
　ステップＳＴ２３において、支援可否判定部１１ａは、地図ＤＢ３３から取得したリンクデータを用いて図１０Ｂ又は図１０Ｃに示される計算式を計算し、経路特徴線確度を得る。 In step ST23, the support availability determination unit 11a calculates the formula shown in FIG. 10B or 10C using the link data acquired from the map DB 33, and obtains the route feature line accuracy.
　ステップＳＴ２４において、支援可否判定部１１ａは、ステップＳＴ２３で計算した経路特徴線確度と予め定められた閾値とを比較して、探索経路全体における車線維持支援の可否を判定する。経路特徴線確度が閾値以下である場合（ステップＳＴ２４“ＮＯ”）、支援可否判定部１１ａは、探索経路が車線維持支援に適していない探索経路であると判断して、ステップＳＴ１２の動作を行う。ステップＳＴ１２以降において、支援可否判定部１１ａは、自車両１００が走行中の道路リンクごとに車線維持支援の可否を判定する。 In step ST24, the support availability determination unit 11a compares the route feature line accuracy calculated in step ST23 with a predetermined threshold value to determine whether or not lane keeping support is available for the entire searched route. When the route characteristic line accuracy is equal to or lower than the threshold value (step ST24 “NO”), the supportability determination unit 11a determines that the searched route is not suitable for lane keeping support, and performs the operation of step ST12. .. After step ST12, the support availability determination unit 11a determines the availability of lane keeping support for each road link on which the vehicle 100 is traveling.
　一方、経路特徴線確度が閾値より大きい場合（ステップＳＴ２４“ＹＥＳ”）、支援可否判定部１１ａは、探索経路が車線維持支援に適した探索経路であると判断して、ステップＳＴ１６の動作を行う。ステップＳＴ１６において、支援可否判定部１１ａは、車線維持支援部１２に対して車線維持支援を開始するように指示する。 On the other hand, when the route feature line accuracy is larger than the threshold value (step ST24 “YES”), the supportability determination unit 11a determines that the searched route is a searched route suitable for lane keeping support, and performs the operation of step ST16. .. In step ST16, the supportability determination part 11a instructs the lane keeping support part 12 to start lane keeping support.
　以上のように、実施の形態２の支援可否判定部１１ａは、過去に特徴線確度学習部１３により学習された特徴線１１０の確度を用いて、ナビゲーション装置３２により探索された探索経路における車線維持支援部１２による支援の可否を判定する。この構成により、ＥＣＵ１０は、車線維持機能の断続的動作防止の連続性を向上させることができる。 As described above, the supportability determination unit 11a according to the second embodiment uses the accuracy of the characteristic line 110 learned by the characteristic line accuracy learning unit 13 in the past to maintain the lane on the searched route searched by the navigation device 32. Whether or not the support by the support unit 12 is possible is determined. With this configuration, the ECU 10 can improve continuity of preventing intermittent operation of the lane keeping function.
実施の形態３．
　実施の形態２で説明したように、実施の形態１に係るＥＣＵ１０は、道路リンクごとに車線維持支援の可否を判定する構成であるため、車線維持支援の可否が断続的に切り替わる場合があった。そこで、実施の形態３に係るＥＣＵ１０は、自車両１００が走行する経路を予測し、予測した経路における車線維持支援の可否を判定する構成とする。 Embodiment 3.
As described in the second embodiment, since the ECU 10 according to the first embodiment is configured to determine whether or not lane keeping support is possible for each road link, the lane keeping support may or may not be intermittently switched. .. Therefore, the ECU 10 according to the third embodiment is configured to predict a route along which the host vehicle 100 travels and determine whether or not lane keeping assistance is possible on the predicted route.
　図１２は、実施の形態３に係るＥＣＵ１０の機能ブロック例を示す図である。実施の形態３に係るＥＣＵ１０は、図１Ｂに示された実施の形態１のＥＣＵ１０における支援可否判定部１１に代えて支援可否判定部１１ｂを備える構成である。また、実施の形態３に係るＥＣＵ１０は、図１Ｂに示された実施の形態１のＥＣＵ１０に対して走行経路予測部１４が追加された構成である。図１２において図１Ａ及び図１Ｂと同一又は相当する部分は、同一の符号を付し説明を省略する。 FIG. 12 is a diagram showing an example of functional blocks of the ECU 10 according to the third embodiment. The ECU 10 according to the third embodiment has a configuration including a support availability determination unit 11b instead of the support availability determination unit 11 in the ECU 10 according to the first embodiment shown in FIG. 1B. Further, the ECU 10 according to the third embodiment has a configuration in which the travel route prediction unit 14 is added to the ECU 10 according to the first embodiment shown in FIG. 1B. In FIG. 12, parts that are the same as or correspond to those in FIGS. 1A and 1B are assigned the same reference numerals and explanations thereof are omitted.
　走行経路予測部１４は、地図ＤＢ３３の情報及び車両ネットワークＥＣＵ４０の情報を用いて自車両１００が走行する経路を予測し、予測した経路（以下、「予測経路」と称する）を支援可否判定部１１ｂへ出力する。 The travel route prediction unit 14 predicts a route along which the host vehicle 100 travels by using the information in the map DB 33 and the information in the vehicle network ECU 40, and the predicted route (hereinafter, referred to as “predicted route”) is determined as the support availability determination unit 11b. Output to.
　支援可否判定部１１ｂは、過去に特徴線確度学習部１３により学習された特徴線１１０の確度を用いて、走行経路予測部１４により予測された経路における車線維持支援部１２による支援の可否を判定する。 The support availability determination unit 11b uses the accuracy of the characteristic line 110 learned by the feature line accuracy learning unit 13 in the past to determine whether the lane keeping support unit 12 can assist the route predicted by the travel route prediction unit 14. To do.
　次に、走行経路の予測方法を説明する。
　図１３Ａは、実施の形態３において自車両１００が走行中の道路形状の一例を示す図である。図１３Ｂは、実施の形態３の走行経路予測部１４による経路予測方法を説明する図であり、代表的な運転操作の例を示す。図１３Ｂには、道路リンクＬ１＿１を走行中の自車両１００における運転者が、次の交差点等であるノードＮ１＿２の手前で行う可能性がある代表的な運転操作の組み合わせとして、ケース＃１～＃４等が示されている。なお、図１３Ｂに示されるケース＃１～＃４等は、走行経路予測部１４に予め与えられているものとする。 Next, a method of predicting a travel route will be described.
FIG. 13A is a diagram showing an example of a road shape on which the host vehicle 100 is traveling in the third embodiment. FIG. 13B is a diagram for explaining the route prediction method by the traveling route prediction unit 14 according to the third embodiment, and shows an example of typical driving operation. FIG. 13B shows cases #1 to # as typical combinations of driving operations that the driver of the vehicle 100 traveling on the road link L1_1 may perform before the node N1_2, which is the next intersection or the like. 4 etc. are shown. The cases #1 to #4 and the like shown in FIG. 13B are assumed to be given to the travel route prediction unit 14 in advance.
　図１３Ｃは、実施の形態３の走行経路予測部１４による経路予測方法を説明する図であり、代表的な運転操作ごとの係数の例を示す。図１３Ｃには、ケース＃１～＃４等の運転操作が行われた場合のノードＮ１＿２での状況の種別と、ノードＮ１＿２に接続しているために自車両１００の走行が予測される道路リンクＬ１＿２，Ｌ２＿１，Ｌ３＿１の係数β（ｎ，Ｌ）とが示されている。係数β（ｎ，Ｌ）は、あるノードｎに接続する道路リンクＬの係数であり、係数の値が大きいほどその道路リンクＬを自車両１００が走行する可能性が高い。例えば、ケース＃１のように、ノードＮ１＿２の手前で何も運転操作が行われなかった場合、自車両１００はこのノードＮ１＿２で直進する可能性が最も高いため、ケース＃１の場合は、ノードＮ１＿２から道路リンクＬ１＿２へ直進する場合の係数βが「０．８」と最も高い。
　なお、図１３Ｂ及び図１３Ｃに示されるケース＃１～＃４等と係数β（ｎ，Ｌ）との対応関係は、走行経路予測部１４に予め与えられているものとする。 FIG. 13C is a diagram for explaining the route prediction method by the travel route prediction unit 14 according to the third embodiment, and shows an example of the coefficient for each representative driving operation. FIG. 13C shows a type of situation at the node N1_2 when a driving operation such as cases #1 to #4 is performed, and a road link in which the own vehicle 100 is predicted to travel because it is connected to the node N1_2. The coefficients β(n, L) of L1_2, L2_1, and L3_1 are shown. The coefficient β(n,L) is a coefficient of the road link L connected to a certain node n, and the larger the value of the coefficient, the higher the possibility that the vehicle 100 travels on the road link L. For example, if no driving operation is performed before the node N1_2 as in case #1, the own vehicle 100 is most likely to go straight at this node N1_2. The coefficient β when going straight from N1_2 to the road link L1_2 is “0.8”, which is the highest.
Note that the correspondence relationship between the cases #1 to #4 and the like shown in FIGS. 13B and 13C and the coefficient β(n,L) is assumed to be given to the travel route prediction unit 14 in advance.
　走行経路予測部１４は、自車両１００の運転者が行ったアクセル及びブレーキ等の運転操作を示す情報を、車両ネットワークＥＣＵ４０から取得する。そして、走行経路予測部１４は、取得した運転操作を示す情報に対応するケースを判定し、そのケースにおいて通過が予測される各道路リンクの係数β（ｎ，Ｌ）を抽出する。走行経路予測部１４は、抽出した各道路リンクの係数β（ｎ，Ｌ）を支援可否判定部１１ｂへ出力する。 The travel route prediction unit 14 acquires, from the vehicle network ECU 40, information indicating a driving operation such as an accelerator and a brake performed by the driver of the vehicle 100. Then, the travel route prediction unit 14 determines a case corresponding to the acquired information indicating the driving operation, and extracts the coefficient β(n,L) of each road link predicted to pass in that case. The travel route prediction unit 14 outputs the extracted coefficient β(n, L) of each road link to the support availability determination unit 11b.
　図１３Ｄは、実施の形態３における経路特徴線確度の計算式の一例を示す図である。なお、実施の形態３の経路特徴線確度は、予測経路における特徴線の確度であり、一方、実施の形態２の経路特徴線確度は、探索経路における特徴線の確度である。支援可否判定部１１ｂは、自車両１００が走行する可能性がある各道路リンクの係数β（ｎ，Ｌ）を、走行経路予測部１４から取得する。そして、支援可否判定部１１ｂは、図１３Ｄの計算式に従って各道路リンクの係数β（ｎ，Ｌ）と、各道路リンクに関連付けて地図ＤＢ３３に記憶された特徴線確度との積和演算を行い、走行が予測される経路の経路特徴線確度を得る。 FIG. 13D is a diagram showing an example of a calculation formula of route characteristic line accuracy in the third embodiment. The route characteristic line accuracy in the third embodiment is the accuracy of the characteristic line in the predicted route, while the route characteristic line accuracy in the second embodiment is the accuracy of the characteristic line in the searched route. The support availability determination unit 11b acquires the coefficient β(n,L) of each road link in which the vehicle 100 may travel from the travel route prediction unit 14. Then, the supportability determination unit 11b performs a product-sum operation of the coefficient β(n, L) of each road link and the characteristic line accuracy stored in the map DB 33 in association with each road link according to the calculation formula of FIG. 13D. , Obtain the route characteristic line accuracy of the route predicted to travel.
　図１４は、実施の形態３に係るＥＣＵ１０の動作例を示すフローチャートである。図１４のステップＳＴ１１～ＳＴ１８における動作は、図５のステップＳＴ１１～ＳＴ１８における動作と同じである。 FIG. 14 is a flowchart showing an operation example of the ECU 10 according to the third embodiment. The operation in steps ST11 to ST18 in FIG. 14 is the same as the operation in steps ST11 to ST18 in FIG.
　支援可否判定部１１ｂは、走行速度が支援開始速度より高い場合（ステップＳＴ１１“ＹＥＳ”）、走行経路を予測するように走行経路予測部１４に指示する。この指示を受けた走行経路予測部１４は、ステップＳＴ３１において、走行経路を予測する。上記のように、走行経路予測部１４は、車両ネットワークＥＣＵ４０から運転操作を示す情報を取得し、運転操作に対応するケースを判定し、判定したケースにおいて通過が予測される各道路リンクの係数を抽出する。そして、走行経路予測部１４は、抽出した各道路リンクの係数を支援可否判定部１１ｂへ出力する。例えば、自車両１００の運転者がノードＮ１＿２の手前でアクセルのみ操作した場合（つまり、ケース＃２の場合）、走行経路予測部１４は、自車両１００がノードＮ１＿２を直進して道路リンクＬ１＿２を走行すると予測し、道路リンクＬ１＿２の係数「１．０」を支援可否判定部１１ｂへ出力する。 When the travel speed is higher than the support start speed (step ST11 “YES”), the support availability determination unit 11b instructs the travel route prediction unit 14 to predict the travel route. Receiving this instruction, the traveling route prediction unit 14 predicts the traveling route in step ST31. As described above, the traveling route prediction unit 14 acquires information indicating the driving operation from the vehicle network ECU 40, determines the case corresponding to the driving operation, and determines the coefficient of each road link predicted to pass in the determined case. Extract. Then, the travel route prediction unit 14 outputs the extracted coefficient of each road link to the support availability determination unit 11b. For example, when the driver of the own vehicle 100 operates only the accelerator in front of the node N1_2 (that is, in case #2), the traveling route prediction unit 14 causes the own vehicle 100 to go straight through the node N1_2 and follow the road link L1_2. It is predicted that the vehicle will travel, and the coefficient “1.0” of the road link L1_2 is output to the support availability determination unit 11b.
　ステップＳＴ３２において、支援可否判定部１１ｂは、各道路リンクの係数を走行経路予測部１４から取得する。また、支援可否判定部１１ｂは、各道路リンクに関連付けて地図ＤＢ３３に記憶されたリンクデータを、地図ＤＢ３３から取得する。 In step ST32, the support availability determination unit 11b acquires the coefficient of each road link from the travel route prediction unit 14. The support availability determination unit 11b also acquires from the map DB 33 the link data stored in the map DB 33 in association with each road link.
　ステップＳＴ３３において、支援可否判定部１１ｂは、地図ＤＢ３３から取得したリンクデータに含まれる特徴線確度と、走行経路予測部１４から取得した各道路リンクの係数とを用いて、図１３Ｄに示される計算式を計算し、経路特徴線確度を得る。 In step ST33, the support availability determination unit 11b uses the characteristic line accuracy included in the link data acquired from the map DB 33 and the coefficient of each road link acquired from the travel route prediction unit 14 to perform the calculation shown in FIG. 13D. An equation is calculated to obtain the route feature line accuracy.
　ステップＳＴ３４において、支援可否判定部１１ｂは、ステップＳＴ３３で計算した経路特徴線確度と予め定められた閾値とを比較して、予測経路における車線維持支援の可否を判定する。経路特徴線確度が閾値以下である場合（ステップＳＴ３４“ＮＯ”）、支援可否判定部１１ｂは、予測経路が車線維持支援に適していない経路であると判断して、ステップＳＴ１２の動作を行う。ステップＳＴ１２以降において、支援可否判定部１１ｂは、自車両１００が走行中の道路リンクごとに車線維持支援の可否を判定する。 In step ST34, the support availability determination unit 11b compares the route feature line accuracy calculated in step ST33 with a predetermined threshold value to determine whether or not lane keeping assistance is available on the predicted route. When the route characteristic line accuracy is equal to or less than the threshold value (step ST34 “NO”), the supportability determination unit 11b determines that the predicted route is not suitable for lane keeping support, and performs the operation of step ST12. After step ST12, the support availability determination unit 11b determines the availability of lane keeping assistance for each road link in which the vehicle 100 is traveling.
　一方、経路特徴線確度が閾値より大きい場合（ステップＳＴ３４“ＹＥＳ”）、支援可否判定部１１ｂは、予測経路が車線維持支援に適した経路であると判断して、ステップＳＴ１６の動作を行う。ステップＳＴ１６において、支援可否判定部１１ｂは、車線維持支援部１２に対して車線維持支援を開始するように指示する。 On the other hand, when the route feature line accuracy is larger than the threshold value (step ST34 “YES”), the support availability determination unit 11b determines that the predicted route is a route suitable for lane keeping support, and performs the operation of step ST16. In step ST16, the supportability determination part 11b instructs the lane keeping support part 12 to start lane keeping support.
　以上のように、実施の形態３に係るＥＣＵ１０は、自車両１００が走行する経路を予測する走行経路予測部１４を備える。支援可否判定部１１ｂは、過去に特徴線確度学習部１３により学習された特徴線１１０の確度を用いて、走行経路予測部１４により予測された経路における車線維持支援部１２による支援の可否を判定する。この構成により、ＥＣＵ１０は、自車両１００の運転者がナビゲーション装置３２を使用しない日常走行の道路であっても、走行経路を予測することができる。したがって、ＥＣＵ１０は、車線維持機能の断続的動作防止の連続性を向上させることができる。 As described above, the ECU 10 according to the third embodiment includes the travel route prediction unit 14 that predicts the route along which the vehicle 100 travels. The support availability determination unit 11b uses the accuracy of the characteristic line 110 learned by the feature line accuracy learning unit 13 in the past to determine whether the lane keeping support unit 12 can assist the route predicted by the travel route prediction unit 14. To do. With this configuration, the ECU 10 can predict the travel route even on the road that is the daily travel road where the driver of the vehicle 100 does not use the navigation device 32. Therefore, the ECU 10 can improve continuity of preventing intermittent operation of the lane keeping function.
実施の形態４．
　実施の形態１～３に係る車線維持支援システム１において、特徴線検出に使用されるカメラ２０と画像処理装置２１の性能は、走行する時点の走行環境の照度によって影響を受ける。そこで、実施の形態４に係るＥＣＵ１０は、特徴線確度を照度ごとに学習しておき、車線維持支援の可否を判定する際に、その際の照度に対応する学習済みの特徴線確度を用いる構成とする。 Fourth Embodiment
In the lane keeping support system 1 according to the first to third embodiments, the performances of the camera 20 and the image processing device 21 used for detecting the characteristic line are affected by the illuminance of the traveling environment at the time of traveling. Therefore, the ECU 10 according to the fourth embodiment learns the characteristic line accuracy for each illuminance and uses the learned characteristic line accuracy corresponding to the illuminance at the time of determining whether or not lane keeping support is possible. And
　図１５は、実施の形態４に係るＥＣＵ１０の機能ブロック例を示す図である。実施の形態４に係るＥＣＵ１０は、図１Ｂに示された実施の形態１のＥＣＵ１０における支援可否判定部１１及び特徴線確度学習部１３に代えて支援可否判定部１１ｃ及び特徴線確度学習部１３ｃを備える構成である。また、実施の形態４に係るＥＣＵ１０は、図１Ｂに示された実施の形態１のＥＣＵ１０に対して環境判定部１５が追加された構成である。さらに、実施の形態４に係るナビゲーション装置３２は、図１Ｂに示された実施の形態１の地図ＤＢ３３に代えて地図ＤＢ３３ｃを備える構成である。図１５において図１Ａ及び図１Ｂと同一又は相当する部分は、同一の符号を付し説明を省略する。 FIG. 15 is a diagram showing an example of functional blocks of the ECU 10 according to the fourth embodiment. The ECU 10 according to the fourth embodiment includes a support availability determination unit 11c and a feature line accuracy learning unit 13c in place of the support availability determination unit 11 and the feature line accuracy learning unit 13 in the ECU 10 of the first embodiment shown in FIG. 1B. It is a configuration provided with. Further, the ECU 10 according to the fourth embodiment has a configuration in which an environment determination unit 15 is added to the ECU 10 according to the first embodiment shown in FIG. 1B. Furthermore, the navigation device 32 according to the fourth embodiment is configured to include a map DB 33c instead of the map DB 33 of the first embodiment shown in FIG. 1B. In FIG. 15, parts that are the same as or correspond to those in FIGS. 1A and 1B are assigned the same reference numerals and explanations thereof are omitted.
　環境判定部１５は、車両ネットワークＥＣＵ４０の情報を用いて、自車両１００が走行する環境を判定し、判定した環境を支援可否判定部１１ｃへ出力する。環境判定部１５が判定する環境とは、例えば照度の環境である。 The environment determination unit 15 uses the information from the vehicle network ECU 40 to determine the environment in which the vehicle 100 travels, and outputs the determined environment to the support availability determination unit 11c. The environment determined by the environment determination unit 15 is, for example, an environment of illuminance.
　特徴線確度学習部１３ｃは、自車両１００が走行中の道路で検出された特徴線１１０の確度を、環境判定部１５により判定された環境ごとに学習し、学習結果を地図ＤＢ３３ｃに記憶させる。 The characteristic line accuracy learning unit 13c learns the accuracy of the characteristic line 110 detected on the road on which the vehicle 100 is traveling for each environment determined by the environment determination unit 15, and stores the learning result in the map DB 33c.
　支援可否判定部１１ｃは、過去に特徴線確度学習部１３ｃにより学習された環境ごとの特徴線１１０の確度を用いて、自車両１００が走行する道路及び環境における車線維持支援部１２による支援の可否を判定する。 The support availability determination unit 11c uses the accuracy of the feature line 110 for each environment that has been learned by the feature line accuracy learning unit 13c in the past, and determines whether or not the lane keeping support unit 12 can support the road on which the vehicle 100 is traveling and the environment. To judge.
　図１６は、実施の形態４の地図ＤＢ３３ｃに記憶されている情報の一部を模式的に表した図である。一つのリンクデータは、その道路リンク自体のＩＤ（この例の場合はＬ１＿２）、及びその道路リンクの始点と終点を表すノードの情報を最低限含み、それ以外に種々の属性を含み得る。属性の例としては、制限速度、及び道路種別（一般道、有料道路、及び高速道路等）等がある。さらに、実施の形態４では、道路リンクの属性として、この道路リンクの特徴線確度を学習した回数と、学習時の照度範囲ごとの特徴線確度とが追加されている。この例では、特徴線確度は、最も照度が低い「０以上ａ未満」（以下、「０～ａ」と称する）と、照度が中程度の「ａ以上ｂ未満」（以下、「ａ～ｂ」と称する）と、最も照度が高い「ｂ以上ｃ未満」（以下、「ｂ～ｃ」と称する）の三段階に分けられている。 FIG. 16 is a diagram schematically showing a part of the information stored in the map DB 33c of the fourth embodiment. One piece of link data includes at least the ID of the road link itself (L1_2 in this example), the information of the node indicating the start point and the end point of the road link, and may include other various attributes. Examples of attributes include speed limits and road types (general roads, toll roads, expressways, etc.). Further, in the fourth embodiment, the number of times the characteristic line accuracy of the road link is learned and the characteristic line accuracy for each illuminance range at the time of learning are added as the attributes of the road link. In this example, the characteristic line accuracy is “0 or more and less than a” (hereinafter, referred to as “0 to a”) where the illuminance is the lowest, and “a or more and less than b” where the illuminance is medium (hereinafter, “a to b”). ") and the highest illuminance is "more than b and less than c" (hereinafter referred to as "b to c").
　図１７Ａは、実施の形態４のナビゲーション装置３２が探索した探索経路の一例を示す図である。図１７Ａに示される探索経路は、出発地ＯであるノードＮ１＿１と目的地ＤであるノードＮ１＿５とを結ぶ道路リンク群Ｌ１＿１，Ｌ１＿２，Ｌ１＿３，Ｌ１＿４で表される。図１７Ｂは、図１７Ａに示される探索経路に沿った道路リンク群Ｌ１＿１，Ｌ１＿２，Ｌ１＿３，Ｌ１＿４のリンクデータの一部を示す図である。 FIG. 17A is a diagram showing an example of a search route searched by the navigation device 32 of the fourth embodiment. The search route shown in FIG. 17A is represented by road link groups L1_1, L1_2, L1_3, L1_4 that connect the node N1_1 that is the departure point O and the node N1_5 that is the destination D. FIG. 17B is a diagram showing a part of the link data of the road link groups L1_1, L1_2, L1_3, L1_4 along the search route shown in FIG. 17A.
　図１７Ｃは、実施の形態４における経路特徴線確度の計算式の一例を示す図である。なお、実施の形態４の経路特徴線確度は、探索経路における特徴線の確度であってもよいし、予測経路における特徴線の確度であってもよい。環境判定部１５は、現在の走行環境の照度ｔを車両ネットワークＥＣＵ４０から取得して支援可否判定部１１ｃへ出力する。支援可否判定部１１ｃは、照度ｔに該当する、探索経路に含まれる各道路リンクの特徴線確度（Ｌｘ＿ｙ，ｔ）を、地図ＤＢ３３ｃから取得する。例えば、現在の走行環境の照度ｔが照度範囲「ａ～ｂ」に該当する場合、支援可否判定部１１ｃは、リンクデータに「ｖ２」として記憶されている特徴線確度を取得する。そして、支援可否判定部１１ｃは、図１７Ｃの計算式に従って、照度ｔに該当する、探索経路に含まれる各道路リンクの特徴線確度（Ｌｘ＿ｙ，ｔ）の総和を計算し、経路特徴線確度を得る。 FIG. 17C is a diagram showing an example of a calculation formula of route characteristic line accuracy in the fourth embodiment. The route feature line accuracy in the fourth embodiment may be the accuracy of the feature line in the searched route or the accuracy of the feature line in the predicted route. The environment determination unit 15 acquires the illuminance t of the current traveling environment from the vehicle network ECU 40 and outputs it to the supportability determination unit 11c. The support availability determination unit 11c acquires, from the map DB 33c, the characteristic line accuracy (Lx_y, t) of each road link included in the searched route that corresponds to the illuminance t. For example, when the illuminance t of the current traveling environment corresponds to the illuminance range “a to b”, the support availability determination unit 11c acquires the characteristic line accuracy stored as “v2” in the link data. Then, the supportability determination unit 11c calculates the sum of the characteristic line accuracies (Lx_y, t) of the road links included in the searched route, which correspond to the illuminance t, according to the calculation formula of FIG. obtain.
　さらに、環境判定部１５は、自車両１００が探索経路の各道路リンク通過に要する所要時間を推定し、現在の時刻を元に、探索経路の各道路リンクを通過する推定時刻における予想照度Ｔを予想してもよい。この場合、支援可否判定部１１ｃは、予想照度Ｔに該当する、探索経路に含まれる各道路リンクの特徴線確度を用いて経路特徴線確度を計算することにより、将来の照度変化を加味して車線維持支援の可否を判定することができる。 Furthermore, the environment determination unit 15 estimates the time required for the vehicle 100 to pass each road link on the search route, and based on the current time, the estimated illuminance T at the estimated time when the vehicle 100 passes each road link on the search route. You may expect. In this case, the supportability determination unit 11c adds the future illuminance change by calculating the route characteristic line accuracy using the characteristic line accuracy of each road link included in the searched route that corresponds to the expected illuminance T. It is possible to determine whether or not lane keeping support is available.
　図１８Ａは、実施の形態４の環境判定部１５により予想された照度を説明する図である。図１８Ａに示されるグラフの横軸は自車両１００が道路リンクを通過する推定時刻、縦軸は予想照度Ｔである。ここでは、自車両１００が図１７Ａに示される探索経路を走行する場合を想定する。また、現在の時刻が夕暮れの時間帯であるものとする。さらに、自車両１００が道路リンクＬ１＿１を走行している現在、環境判定部１５により車両ネットワークＥＣＵ４０から照度範囲「ｂ～ｃ」に該当する照度ｔが取得されたとする。この場合、支援可否判定部１１ｃは、道路リンクＬ１＿１の特徴線確度として「ｖ３」を地図ＤＢ３３ｃから取得する。 FIG. 18A is a diagram illustrating the illuminance predicted by the environment determination unit 15 according to the fourth embodiment. The horizontal axis of the graph shown in FIG. 18A is the estimated time when the vehicle 100 passes through the road link, and the vertical axis is the expected illuminance T. Here, it is assumed that the host vehicle 100 travels on the search route shown in FIG. 17A. In addition, it is assumed that the current time is the dusk time zone. Furthermore, it is assumed that the environment determination unit 15 acquires the illuminance t corresponding to the illuminance range “b to c” from the vehicle network ECU 40 at present when the vehicle 100 is traveling on the road link L1_1. In this case, the supportability determination unit 11c acquires “v3” as the characteristic line accuracy of the road link L1_1 from the map DB 33c.
　また、環境判定部１５は、車両ネットワークＥＣＵ４０から取得した照度範囲「ｂ～ｃ」に該当する現在の照度ｔと現在の時刻等に基づいて、自車両１００が道路リンクＬ１＿２，Ｌ１＿３，Ｌ１＿４を通過する推定時刻における各予想照度Ｔを予想する。例えば、環境判定部１５は、ナビゲーション装置３２で検出された、現在の走行位置（緯度経度）及び日時を利用して、照度を予想する。図１８Ａでは現在の時刻が夕暮れの時間帯であるため、予想照度Ｔは、時間が進むにつれて低くなっていく。具体的には、道路リンクＬ１＿２を通過する推定時刻における予想照度Ｔが照度範囲「ｂ～ｃ」に該当し、道路リンクＬ１＿３を通過する推定時刻における予想照度Ｔが照度範囲「ａ～ｂ」に該当し、道路リンクＬ１＿４を通過する推定時刻における予想照度Ｔが照度範囲「０～ａ」に該当する。この場合、支援可否判定部１１ｃは、道路リンクＬ１＿２の特徴線確度として「ｖ３」を、道路リンクＬ１＿３の特徴線確度として「ｖ２」を、道路リンクＬ１＿４の特徴線確度として「ｖ１」を、地図ＤＢ３３ｃから取得する。 Further, the environment determination unit 15 determines that the vehicle 100 passes through the road links L1_2, L1_3, L1_4 based on the current illuminance t corresponding to the illuminance range “b to c” acquired from the vehicle network ECU 40 and the current time. Predict each expected illuminance T at the estimated time. For example, the environment determination unit 15 estimates the illuminance by using the current traveling position (latitude/longitude) and date and time detected by the navigation device 32. In FIG. 18A, since the current time is in the dusk time zone, the expected illuminance T becomes lower as the time advances. Specifically, the expected illuminance T at the estimated time of passing the road link L1_2 corresponds to the illuminance range “b to c”, and the expected illuminance T at the estimated time of passing the road link L1_3 falls to the illuminance range “a to b”. Correspondingly, the expected illuminance T at the estimated time of passing through the road link L1_4 corresponds to the illuminance range “0 to a”. In this case, the supportability determination unit 11c sets "v3" as the characteristic line accuracy of the road link L1_2, "v2" as the characteristic line accuracy of the road link L1_3, and "v1" as the characteristic line accuracy of the road link L1_4. Obtain from DB33c.
　図１８Ｂは、実施の形態４における経路特徴線確度の計算式の一例を示す図である。支援可否判定部１１ｃは、上記のように地図ＤＢ３３ｃから取得した道路リンクＬ１＿１，Ｌ１＿２，Ｌ１＿３，Ｌ１＿４の特徴線確度を用いて、図１８Ｂの計算式を計算することによって経路特徴線確度を得る。 FIG. 18B is a diagram showing an example of a calculation formula of route characteristic line accuracy in the fourth embodiment. The support availability determination unit 11c obtains the route characteristic line accuracy by calculating the calculation formula of FIG. 18B using the characteristic line accuracy of the road links L1_1, L1_2, L1_3, L1_4 acquired from the map DB 33c as described above.
　なお、環境判定部１５は、自車両１００のヘッドランプ点灯又は走行中道路の街灯点灯等によって走行環境の実際の照度ｔと予想照度Ｔとにずれが生じた場合、照度を予想し直してもよい。その場合、支援可否判定部１１ｃは、環境判定部１５により予想し直された予想照度Ｔに基づいて経路特徴線確度を計算し直し、閾値判定により支援可否を判定し直せばよい。 Note that the environment determination unit 15 may re-estimate the illuminance when the actual illuminance t and the expected illuminance T of the traveling environment deviate due to the lighting of the headlamp of the vehicle 100, the lighting of the streetlight of the road on which the vehicle is traveling, or the like. Good. In that case, the support availability determination unit 11c may recalculate the route feature line accuracy based on the predicted illuminance T re-estimated by the environment determination unit 15, and determine the support availability again by the threshold value determination.
　以上のように、実施の形態４に係るＥＣＵ１０は、自車両１００が走行する環境を判定する環境判定部１５を備える。特徴線確度学習部１３ｃは、自車両１００が走行中の道路で検出された特徴線１１０の確度を、環境判定部１５により判定された環境ごとに学習する。支援可否判定部１１ｃは、過去に特徴線確度学習部１３ｃにより学習された環境ごとの特徴線１１０の確度を用いて、自車両１００が走行する道路及び環境における車線維持支援部１２による支援の可否を判定する。この構成により、ＥＣＵ１０は、走行環境の差異に起因した特徴線検出性能の差異による支援可否判定のばらつきを抑止することができる。 As described above, the ECU 10 according to the fourth embodiment includes the environment determination unit 15 that determines the environment in which the host vehicle 100 travels. The characteristic line accuracy learning unit 13c learns the accuracy of the characteristic line 110 detected on the road on which the own vehicle 100 is traveling for each environment determined by the environment determination unit 15. The support availability determination unit 11c uses the accuracy of the feature line 110 for each environment that has been learned by the feature line accuracy learning unit 13c in the past, and determines whether or not the lane keeping support unit 12 can support the road on which the vehicle 100 is traveling and the environment. To judge. With this configuration, the ECU 10 can suppress the variation in the support availability determination due to the difference in the characteristic line detection performance due to the difference in the traveling environment.
　なお、実施の形態４では、環境の一例として照度の環境を示したが、照度の環境に限定されるものではなく、特徴線検出性能に影響を及ぼす環境であれば何でもよい。また、実施の形態４の構成は、実施の形態１～３のいずれの構成とも組み合わせが可能である。 In the fourth embodiment, the illuminance environment is shown as an example of the environment, but the environment is not limited to the illuminance environment, and any environment that affects the feature line detection performance may be used. Further, the configuration of the fourth embodiment can be combined with any of the configurations of the first to third embodiments.
　また、各実施の形態に記載された、学習回数に対する閾値、及び特徴線確度に対する閾値は、固定値である必要はなく、運転者の好み等に応じて調整及び設定が可能な可変値であってもよい。閾値が可変値である場合、ＥＣＵ１０は、例えばＨＭＩ５０を用いて、閾値設定画面を表示させ、運転者が設定した閾値の情報を受け付ける。 Further, the threshold value for the number of learnings and the threshold value for the characteristic line accuracy described in each embodiment do not need to be fixed values, and are variable values that can be adjusted and set according to the driver's preference. May be. When the threshold value is a variable value, the ECU 10 displays the threshold value setting screen by using, for example, the HMI 50, and receives the threshold value information set by the driver.
　本発明はその発明の範囲内において、各実施の形態の自由な組み合わせ、各実施の形態の任意の構成要素の変形、又は各実施の形態の任意の構成要素の省略が可能である。 Within the scope of the invention, it is possible to freely combine the embodiments, modify any constituent element of each embodiment, or omit any constituent element of each embodiment.
　この発明に係る車線維持支援装置は、車線維持支援の可否を判定するようにしたので、車線維持等の運転を支援する運転支援装置等に用いるのに適している。 Since the lane keeping assist device according to the present invention determines whether or not lane keeping assist is possible, it is suitable for use as a driving assistance device or the like that assists driving such as lane keeping.
　１　車線維持支援システム、１０　ＥＣＵ（車線維持支援装置）、１１，１１ａ，１１ｂ，１１ｃ　支援可否判定部、１２　車線維持支援部、１３，１３ｃ　特徴線確度学習部、１４　走行経路予測部、１５　環境判定部、２０　カメラ、２１　画像処理装置、３０　ＧＰＳ受信機、３１　アンテナ、３２　ナビゲーション装置、３３，３３ｃ　地図ＤＢ、４０　車両ネットワークＥＣＵ、５０　ＨＭＩ、６０　ステアリングホイール、６１　舵角センサ、６２　トルクセンサ、６３　電動回路、６４　モータ、６５　ギヤボックス、６６　車輪、１００　自車両、１０１　ガードレール、１０２　中央分離帯、１０３，１０４　車道外側線、１０５　境界線、１１０　特徴線、１２１　かすれ、１２２　途切れ、Ｌ１＿１，Ｌ１＿２，Ｌ１＿３，Ｌ１＿４，Ｌ２＿１，Ｌ３＿１，Ｌ４＿１，Ｌ５＿１，Ｌ５＿２　道路リンク、Ｎ１＿１，Ｎ１＿２，Ｎ１＿３，Ｎ１＿４，Ｎ１＿５，Ｎ２＿１，Ｎ３＿１，Ｎ４＿１，Ｎ５＿１，Ｎ５＿２　ノード。 1 lane maintenance support system, 10 ECU (lane maintenance support device), 11, 11a, 11b, 11c support availability determination unit, 12 lane maintenance support unit, 13 and 13c characteristic line accuracy learning unit, 14 travel route prediction unit, 15 environment Judgment unit, 20 camera, 21 image processing device, 30 GPS receiver, 31 antenna, 32 navigation device, 33, 33c map DB, 40 vehicle network ECU, 50 HMI, 60 steering wheel, 61 steering angle sensor, 62 torque sensor, 63 electric circuit, 64 motor, 65 gearbox, 66 wheels, 100 own vehicle, 101 guardrail, 102 median strip, 103, 104 road outside line, 105 boundary line, 110 characteristic line, 121 faint, 122 break, L1_1, L1_2 , L1_3, L1_4, L2_1, L3_1, L4_1, L5_1, L5_2 Road links, N1_1, N1_2, N1_3, N1_4, N1_5, N2_1, N3_1, N4_1, N5_1, N5_2 nodes.

Claims (5)

  1. 　自車両前方の撮像画像から検出された、前記自車両が走行する車線を区画する特徴線を用いて、前記車線内の走行を維持するように前記自車両を支援する車線維持支援部と、
    　前記自車両が走行中の道路で検出された特徴線の確度を学習する特徴線確度学習部と、
    　過去に前記特徴線確度学習部により学習された特徴線の確度と過去に前記自車両が走行した道路に対して行われた学習回数とを用いて、前記自車両が走行する道路における前記車線維持支援部による支援の可否を判定する支援可否判定部とを備える車線維持支援装置。     A lane keeping assist unit that assists the own vehicle to maintain traveling in the lane, using a characteristic line that divides the lane in which the own vehicle travels, detected from a captured image in front of the own vehicle,
    A characteristic line accuracy learning unit that learns the accuracy of the characteristic line detected on the road on which the vehicle is traveling,
    Using the accuracy of the characteristic line learned by the characteristic line accuracy learning unit in the past and the number of learnings performed on the road on which the own vehicle has traveled in the past, the lane keeping on the road on which the own vehicle travels A lane keeping assist device, comprising: a support availability determination unit that determines availability of support by a support unit.
  2. 　前記支援可否判定部は、過去に前記特徴線確度学習部により学習された特徴線の確度を用いて、ナビゲーション装置により探索された探索経路における前記車線維持支援部による支援の可否を判定することを特徴とする請求項１記載の車線維持支援装置。 The support availability determination unit determines the availability of support by the lane keeping assistance unit in the search route searched by the navigation device, using the accuracy of the feature line learned by the feature line accuracy learning unit in the past. The lane keeping assist device according to claim 1.
  3. 　前記自車両が走行する経路を予測する走行経路予測部を備え、
    　前記支援可否判定部は、過去に前記特徴線確度学習部により学習された特徴線の確度を用いて、前記走行経路予測部により予測された経路における前記車線維持支援部による支援の可否を判定することを特徴とする請求項１記載の車線維持支援装置。     A travel route prediction unit that predicts a route along which the vehicle travels,
    The support availability determination unit determines the availability of support by the lane maintenance assistance unit on the route predicted by the travel route prediction unit, using the accuracy of the characteristic line learned by the characteristic line accuracy learning unit in the past. The lane keeping assist device according to claim 1, wherein:
  4. 　前記自車両が走行する環境を判定する環境判定部を備え、
    　前記特徴線確度学習部は、前記自車両が走行中の道路で検出された特徴線の確度を、前記環境判定部により判定された環境ごとに学習し、
    　前記支援可否判定部は、過去に前記特徴線確度学習部により学習された環境ごとの特徴線の確度を用いて、前記自車両が走行する道路及び環境における前記車線維持支援部による支援の可否を判定することを特徴とする請求項１記載の車線維持支援装置。     An environment determination unit that determines the environment in which the vehicle is traveling,
    The characteristic line accuracy learning unit learns the accuracy of the characteristic line detected on the road on which the own vehicle is traveling, for each environment determined by the environment determination unit,
    The support availability determination unit uses the accuracy of the feature line for each environment learned by the feature line accuracy learning unit in the past to determine the availability of support by the lane maintenance support unit on the road and the environment on which the vehicle is traveling. The lane keeping assist device according to claim 1, wherein the determination is made.
  5. 　車線維持支援部が、自車両前方の撮像画像から検出された、前記自車両が走行する車線を区画する特徴線を用いて、前記車線内の走行を維持するように前記自車両を支援し、
    　特徴線確度学習部が、前記自車両が走行中の道路で検出された特徴線の確度を学習し、
    　支援可否判定部が、過去に前記特徴線確度学習部により学習された特徴線の確度と過去に前記自車両が走行した道路に対して行われた学習回数とを用いて、前記自車両が走行する道路における前記車線維持支援部による支援の可否を判定する車線維持支援方法。     A lane keeping assist unit is detected from a captured image in front of the own vehicle, using a characteristic line that divides the lane in which the own vehicle travels, and assists the own vehicle to maintain traveling in the lane,
    The characteristic line accuracy learning unit learns the accuracy of the characteristic line detected on the road on which the vehicle is traveling,
    The support availability determination unit uses the accuracy of the characteristic line learned by the characteristic line accuracy learning unit in the past and the number of times of learning performed on the road on which the own vehicle has traveled in the past, and the own vehicle travels. A lane keeping assist method for determining whether or not assistance by the lane keeping assist unit on a road to be performed is possible.
PCT/JP2019/002737 2019-01-28 2019-01-28 Lane keeping assist device and lane keeping assist method WO2020157798A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2019/002737 WO2020157798A1 (en) 2019-01-28 2019-01-28 Lane keeping assist device and lane keeping assist method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2019/002737 WO2020157798A1 (en) 2019-01-28 2019-01-28 Lane keeping assist device and lane keeping assist method

Publications (1)

Publication Number Publication Date
WO2020157798A1 true WO2020157798A1 (en) 2020-08-06

Family

ID=71840942

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2019/002737 WO2020157798A1 (en) 2019-01-28 2019-01-28 Lane keeping assist device and lane keeping assist method

Country Status (1)

Country Link
WO (1) WO2020157798A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210303873A1 (en) * 2018-10-15 2021-09-30 Mitsubishi Electric Corporation Lane link generation device and computer readable medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2013244767A (en) * 2012-05-23 2013-12-09 Isuzu Motors Ltd Lane departure warning device, vehicle mounted with the same, and method for controlling the same
JP2014218098A (en) * 2013-05-01 2014-11-20 トヨタ自動車株式会社 Drive support device and drive support method
JP2017126364A (en) * 2017-03-22 2017-07-20 株式会社デンソー Driving support device

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2013244767A (en) * 2012-05-23 2013-12-09 Isuzu Motors Ltd Lane departure warning device, vehicle mounted with the same, and method for controlling the same
JP2014218098A (en) * 2013-05-01 2014-11-20 トヨタ自動車株式会社 Drive support device and drive support method
JP2017126364A (en) * 2017-03-22 2017-07-20 株式会社デンソー Driving support device

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210303873A1 (en) * 2018-10-15 2021-09-30 Mitsubishi Electric Corporation Lane link generation device and computer readable medium
US12020491B2 (en) * 2018-10-15 2024-06-25 Mitsubishi Electric Corporation Lane link generation device and computer readable medium

Similar Documents

Publication Publication Date Title
JP4367431B2 (en) Vehicle driving support system
US7031829B2 (en) Car navigation system prioritizing automatic travel road
US5612882A (en) Method and apparatus for providing navigation guidance
JP5541103B2 (en) Travel guidance device, travel guidance method, and computer program
US20060215020A1 (en) Visual recognition apparatus, methods, and programs for vehicles
JP2000020891A (en) Navigation device
US7280915B2 (en) Navigation device and method of presenting information corresponding to travel course stage
EP4035960A1 (en) Drive assistance device and computer program
JP7439529B2 (en) Driving support equipment and computer programs
JP4561675B2 (en) Driving support device and computer program
JP4662797B2 (en) Car navigation system
WO2022259733A1 (en) Driving assistance device
JP6822373B2 (en) Automatic driving proposal device and automatic driving proposal method
WO2020157798A1 (en) Lane keeping assist device and lane keeping assist method
JP2019203781A (en) Travel route guiding device and travel route guiding program
JP2006092258A (en) Running-out alert control device and running-out alert control program
JP6080899B2 (en) Vehicle travel control device
EP4170286A1 (en) Drive assistance device and computer program
JP7014263B2 (en) Route setting device
JP2006224904A (en) Vehicle control device
JP2019132654A (en) Travel assisting device, travel assisting system, caution point extraction method, and program
JP7501039B2 (en) Driving assistance device and computer program
JP2010043938A (en) Present position calculating device
JP4679913B2 (en) Car navigation system
US20090105936A1 (en) Route guidance apparatus, route guidance method, route guidance program and computer-readable recording medium

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19913115

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19913115

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP