CN114419874A - Target driving safety risk early warning method based on data fusion of roadside sensing equipment - Google Patents
Target driving safety risk early warning method based on data fusion of roadside sensing equipment Download PDFInfo
- Publication number
- CN114419874A CN114419874A CN202111413903.9A CN202111413903A CN114419874A CN 114419874 A CN114419874 A CN 114419874A CN 202111413903 A CN202111413903 A CN 202111413903A CN 114419874 A CN114419874 A CN 114419874A
- Authority
- CN
- China
- Prior art keywords
- risk
- target vehicle
- target
- early warning
- early
- Prior art date
- Legal status (The legal status 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 status listed.)
- Granted
Links
- 230000004927 fusion Effects 0.000 title claims abstract description 39
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000012216 screening Methods 0.000 claims abstract description 6
- 230000003068 static effect Effects 0.000 claims description 31
- 238000004364 calculation method Methods 0.000 claims description 18
- 238000000926 separation method Methods 0.000 claims description 13
- 230000033001 locomotion Effects 0.000 claims description 12
- 238000012937 correction Methods 0.000 claims description 10
- 230000008447 perception Effects 0.000 claims description 6
- 239000003550 marker Substances 0.000 claims description 3
- 238000001514 detection method Methods 0.000 description 13
- 230000003993 interaction Effects 0.000 description 4
- 230000000007 visual effect Effects 0.000 description 4
- 206010039203 Road traffic accident Diseases 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- NTSBMKIZRSBFTA-AIDOXSFESA-N Digoxigenin bisdigitoxoside Chemical compound C1[C@H](O)[C@H](O)[C@@H](C)O[C@H]1O[C@@H]1[C@@H](C)O[C@@H](O[C@@H]2C[C@@H]3[C@]([C@@H]4[C@H]([C@]5(CC[C@@H]([C@@]5(C)[C@H](O)C4)C=4COC(=O)C=4)O)CC3)(C)CC2)C[C@@H]1O NTSBMKIZRSBFTA-AIDOXSFESA-N 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 231100001261 hazardous Toxicity 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000004575 stone Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 210000001260 vocal cord Anatomy 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
- G08G1/0116—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from roadside infrastructure, e.g. beacons
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/04—Detecting movement of traffic to be counted or controlled using optical or ultrasonic detectors
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to a target driving safety risk early warning method based on roadside sensing equipment data fusion, which is used for carrying out risk early warning on a target vehicle and comprises the following steps: s1: acquiring sensing data of roadside sensing equipment, wherein the roadside sensing equipment is arranged at the roadside of a region to be early-warned, and a target vehicle runs in the region to be early-warned; s2: performing data fusion on the sensing data acquired by the roadside sensing equipment to acquire fused data; s3: performing multi-target tracking on the area to be early-warned according to the fusion data; s4: acquiring a driving safety risk field of a target vehicle in the area to be early warned based on the multi-target tracking result; s5: screening objects with the risk higher than a risk threshold value to the target vehicle in a traffic safety risk field and sending out early warning. Compared with the prior art, the method and the system can accurately and efficiently carry out risk early warning on the target vehicle.
Description
Technical Field
The invention relates to the field of sensor data fusion, in particular to a target driving safety risk early warning method based on roadside sensing equipment data fusion.
Background
In a complex traffic environment, target perception and safety early warning of traffic participants of traffic nodes such as passenger-cargo vehicles, non-motor vehicles and pedestrians which are intersected are one of the most concerned and painful problems for traffic managers. Under the scene, the single-vehicle sensing is often faced with the problems of insufficient sensing range, easy shielding of sensing visual field due to self visual angle, high computing cost and high energy consumption caused by a large amount of computing load, credible judgment of data interaction and the like, and the problems become various bottlenecks restricting the single-vehicle sensing capability and precision. In addition, for the weakest but most flexible traffic group in the environment, the non-motor vehicles and pedestrians cannot be equipped with high-precision sensing equipment such as laser radar and millimeter wave radar, and the sensing capability of the environment is at the level of personal judgment, which also leads to the current situation of great imbalance of the sensing levels of all participants in the traffic environment.
Roadside sensing equipment, such as a roadside video camera and a roadside millimeter wave radar, can be customized and installed at a traffic key collision point and can acquire a full scene sensing view in a mode of 'Godi view angle'. In addition, the roadside sensing equipment can be provided with an edge computing unit with strong computing power, and sensing computing speed and precision are guaranteed. With the development of 5G technology, large data volume communications can be transmitted with lower latency, which makes it possible to personalize perceptual data transmission. For the intelligent networked vehicles with higher computing power, original sensing data and roadside sensing judgment results can be transmitted to the intelligent networked vehicles for decision making, and the credible data interaction judgment capability is improved. For other vehicles with low or no computing capacity, non-motor vehicles and pedestrians, the roadside perception result can be sent to the client side of the vehicle, so that the vehicle also has the capacity of obtaining wide-range perception, and the perception level of each participant in a complex traffic scene is balanced.
In recent decades, with the development of neural networks, the most advanced target detection algorithms (Yolo-V4, EfficientNet, SSD, DetectoRS, etc.) are widely used to detect vehicles on roads. These algorithms may assign a bounding box to each vehicle in the video and output its type and confidence. For the calibrated camera, relative coordinates of the target can be further calculated, and the detection distance can even exceed 1km under the condition of good visual field. In addition, the image data of the target can be acquired to extract features, and even a license plate can be identified. However, the disadvantages of video-based detection are also apparent. The detection precision can be seriously influenced by local feature shielding due to poor illumination conditions. Millimeter wave radar is another mainstream target detection roadside sensor. The structured data output by the radar coordinate system comprises the position and the accurate speed of the target in the radar coordinate system. The millimeter wave radar is hardly influenced by environmental factors, and has good robustness for partially-shielded target detection. But the detection range is limited, and the farthest detection range generally does not exceed 300 m. For larger vehicles, they will be detected as two or more objects approaching the radar. The camera and the radar sensor have respective advantages and can make up for the defects of each other to a certain extent. Therefore, data fusion of the two sensors is considered as an effective means for improving detection accuracy.
In the field of automatic driving safety, driving safety assistance systems have been studied for a long time. Since the 90 s of the 20 th century, many driving safety assistance algorithms have been proposed by automotive companies. For longitudinal security, a safe distance model is mainly adopted. When the following distance is less than the safe distance, the auxiliary system will give an alarm and automatically brake. Many safe distance models determine the safe state of a vehicle by analyzing the safe distance of the relative movement of front and rear vehicles in real time. For lateral safety, the driver safety assistance algorithm is mainly based on the current position of the car (CCP), lane crossing Time (TLC) and variable vocal cords (VRBS). Existing safety models are mostly based on vehicle kinematics and dynamics, and their description of vehicle driving safety is usually based on information about the vehicle state, such as position, speed, acceleration and yaw rate, and information about the relative motion of the vehicle, including relative speed and relative distance. However, these models have the following problems: the influence of all types of traffic factors on driving safety is difficult to reflect; it is difficult to describe the interaction between driver behavior characteristics, vehicle state and road environment; it is difficult to provide accurate judgment basis for vehicle control.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provide a target driving safety risk early warning method based on roadside sensing equipment data fusion.
The purpose of the invention can be realized by the following technical scheme:
a target driving safety risk early warning method based on roadside sensing device data fusion is used for carrying out risk early warning on a target vehicle and comprises the following steps:
s1: acquiring sensing data of roadside sensing equipment, wherein the roadside sensing equipment is arranged at the roadside of a region to be early-warned, and a target vehicle runs in the region to be early-warned;
s2: performing data fusion on the sensing data acquired by the roadside sensing equipment to acquire fused data;
s3: performing multi-target tracking on the area to be early-warned according to the fusion data;
s4: acquiring a driving safety risk field of a target vehicle in the area to be early warned based on the multi-target tracking result;
s5: screening objects with the risk higher than a risk threshold value to the target vehicle in a traffic safety risk field and sending out early warning.
Preferably, the driving safety risk field comprises a static risk field with a risk generation source of a relatively static object in the area to be warned and a mobile risk field with a risk generation source of a relatively moving object in the area to be warned.
Preferably, the objects which are relatively static in the area to be early-warned comprise lane separation lines and hard separation belts.
Preferably, the calculation formula of the still risk of the relatively still object a to the target vehicle j in the area to be early warned is as follows:
wherein E isRFor static risk, LTaRisk factor for the type of lane marker, RaIs the object a is located (x)a,ya) The road condition influencing factor, D is the width of the lane, D is the width of the target vehicle j, rajDistance vector between object a and target vehicle j, k1Magnification factor of distance (x)a,ya) Is the coordinate of object a, (x)j,yj) The coordinates of the target vehicle j.
Preferably, the calculation formula of the movement risk of the relatively moving object b to the target vehicle j in the area to be warned is as follows:
rbj=(xj-xb,yj-yb)
wherein E isVFor moving risks, G is the magnitude of the risk factor between two objects of unit mass per unit distance, RbIs the object b is located (x)b,yb) Road condition influencing factor of (C), TbjCorrection factor r for type between object b and target vehicle jbjIs the distance vector, k, between the object b and the target vehicle j2Is the magnification factor of the distance, k3For the correction of risks of different speeds, vbjIs the relative speed between the object b and the target vehicle j, and theta is rbj、vbjAngle (x) betweenb,yb) Is the coordinate of object b, (x)j,yj) The coordinates of the target vehicle j.
Preferably, the roadside sensing device comprises a video camera and a millimeter wave radar.
Preferably, in the step S2, the sensing data of the video camera and the millimeter wave radar are fused by using a data fusion algorithm, so as to obtain fused data.
Preferably, the data fusion algorithm specifically includes:
wherein z is1Is a first observed value, z2In order to be the second observed value,in order to fuse the data,is a first observation variance;is the second observation variance.
Preferably, in step S3, a kalman filter is used to perform multi-target tracking on the fusion data, so as to obtain object trajectory data in the area to be early-warned.
Preferably, in step S5, a relatively stationary object with a stationary risk greater than a risk threshold for the target vehicle is screened, and a high risk range of the relatively stationary object with a risk greater than the risk threshold is obtained, and an early warning is given to the target vehicle.
Preferably, in step S5, the relative moving object with the movement risk greater than the risk threshold for the target vehicle is screened, and the high risk range of the relative moving object with the movement risk greater than the risk threshold is obtained, and an early warning is given to the target vehicle.
Compared with the prior art, the invention has the following advantages:
(1) the invention combines different types of roadside sensing equipment to perform data fusion, can integrate the sensing advantages of various roadside sensing data, effectively improves the sensing range and the sensing precision, effectively acquires the object track and the position in the area to be early-warned, and improves the accuracy and precision of driving safety risk early warning;
(2) according to the invention, a driving safety risk field is introduced to obtain a static risk field of a relatively static object and a mobile risk field of a relatively moving object in an area to be early-warned, so that the risk of the object in the area to be early-warned on a target vehicle can be effectively extracted and screened based on a risk threshold, a high-risk object in the area to be early-warned can be accurately and efficiently obtained and early-warned, the safety influence of various traffic factors on the target vehicle can be reflected, the interaction among the target vehicle, the vehicle state and the road environment can be described, and driving risk early warning is carried out for vehicle control;
(3) the invention can distribute high-risk results to the target vehicle, screen objects with large risks and send high-risk ranges to the target vehicle, and designs high-risk range data planning methods for static and dynamic objects respectively, thereby improving the safety and reliability of driving risk early warning.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a schematic layout of the roadside sensing device of the present invention;
FIG. 3 is a schematic illustration of the static hazardous field distribution field strength of the present invention;
FIG. 4 is a schematic diagram of the distributed field strength of the mobile risk field of the present invention;
FIG. 5 shows a static risk field rajThe calculation chart of (1);
FIG. 6 is a schematic view of a traffic safety risk distribution;
FIG. 7 is an exemplary graph of full range data;
FIG. 8 is an exemplary graph of high risk range data.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. Note that the following description of the embodiments is merely a substantial example, and the present invention is not intended to be limited to the application or the use thereof, and is not limited to the following embodiments.
Examples
A target driving safety risk early warning method based on roadside sensing device data fusion is used for carrying out risk early warning on a target vehicle, and comprises the following steps as shown in figure 1:
s1: the method comprises the steps of obtaining sensing data of roadside sensing equipment, wherein the roadside sensing equipment is arranged on the roadside of a region to be early-warned, and a target vehicle runs in the region to be early-warned.
In this embodiment, as shown in fig. 2, the invention relies on two sensor data, namely a video camera and a millimeter wave radar, to realize multi-target detection. In this embodiment, the road side sensing device employs a video camera and a millimeter wave radar, is installed in a region to be pre-warned, and can acquire a full-scene sensing view, as shown in fig. 2, the coincidence rate of the camera view and the millimeter wave radar view range is required to be not lower than 60%. And the time service server is used for carrying out unified time service on the two devices to finish time synchronization.
S2: performing data fusion on the sensing data acquired by the roadside sensing equipment by using a data fusion algorithm to acquire fusion data;
the data fusion algorithm specifically comprises the following steps:
wherein z is1Is a first observed value, z2In order to be the second observed value,in order to fuse the data,is a first observation variance;is the second observation variance. In this embodiment, the first observation value is positioning data of the video camera, and the second observation value is positioning data of the millimeter wave radar.
S3: and performing multi-target tracking on the fusion data by adopting a Kalman filter to acquire object track data in the area to be early-warned.
S4: and acquiring a driving safety risk field of the target vehicle in the area to be early warned based on the multi-target tracking result. Specifically, the driving safety risk field comprises a static risk field with a risk generation source of a relatively static object in the area to be warned and a mobile risk field with a risk generation source of a relatively moving object in the area to be warned.
The invention regards things which can cause risks in the traffic environment as danger generating sources, and takes the things as the center to spread to the periphery, and the field intensity of the risk field can be understood as the magnitude of the risk coefficient at a certain distance from the danger generating sources. The closer the distance to the danger center, the higher the probability of occurrence of an accident, and the farther the distance, the lower the probability of occurrence of an accident, and when the distance approaches 0, it can be considered that a contact collision between the target object and the danger occurrence source occurs, that is, a traffic accident has occurred. The driving safety risks are divided into two categories according to different generation sources:
1) static risk field: the field source is an object which is relatively static in the traffic environment, mainly comprises road surface marking lines such as lane dividing lines and the like, and also comprises hard separation facilities such as a central separation belt and the like. This type of object has two features: under the condition of not considering road construction, the object is in a static state relative to a target object, and the actual meaning of the object is represented as a dangerous boundary and does not have a speed attribute; except for part of hard separation facilities, the objects cause the driver to intentionally leave the position of the object based on legal effect, but even if the driver actually takes the action of crossing the lane line, the traffic accident can not happen instantly.
In this embodiment, the risk source of the static risk field, i.e., the object in the area to be warned relatively still, includes the lane separation line and the hard separation zone.
The calculation formula of the static risk of the relatively static object a to the target vehicle j in the area to be early warned is as follows:
LTais a risk coefficient of the lane marking type, determined by traffic regulations, generally a rigid separation facility>Non-crossing lane dividing line>May cross lane separation lines. The parameters of common facilities and lane lines take values as follows: guardrail or greenbelt center separator: 20-25; sidewalk road edge stone: 18-20; yellow solid or dashed line: 15-18; solid white line: 10-15; white dotted line: 0 to 5.
RaIs a constant greater than 0, represents (x)a,ya) The road condition influencing factors are determined by traffic environment factors such as road adhesion coefficient, road gradient, road curvature and visibility of an area near the object a, and a fixed value is generally selected for a section of road in the actual use process. The data interval generally used is [0.5, 1.5 ]]。
D is the width of the lane, D is the width of the object j, and D is the width of the identification frame of the object j.
rajIs the distance vector between the lane marker a and the object j, in this case (x)j,yj) Is a target objectCentroid of i, (x)a,ya) Then represents (x)j,yj) And (5) making a point where the perpendicular line and the lane line a intersect.
k1Is a constant greater than 0, representing a magnification factor of the distance, since the risk of collision does not generally vary linearly with the distance between the two objects. General k1The value range of (A) is 0.5-1.5.
raj/|rajIn general, even if the field intensity directions of two risk sources at a certain point are opposite to each other in practical application, the risk magnitude at the point cannot be considered to be reduced, and the two risk sources are still superposed according to a scalar.
ERThe larger the value, the higher the risk that the field source a poses on the object i. The field strength distribution results are shown in fig. 3.
2) Moving the risk field: the field source is an object which is in relative movement in the traffic environment, and mainly comprises vehicles, pedestrians, roadblock facilities and the like. This type of object has two features as well: although the object may be stationary relative to the road surface, such as roadside parking, road block facilities, etc., the moving target object still has a relative speed with respect to the reference system. Secondly, collision of the objects is strictly prohibited, otherwise serious traffic accidents are inevitably caused.
In this embodiment, a calculation formula of the movement risk of the relatively moving object b to the target vehicle j in the area to be early warned is as follows:
rbj=(xj-xb,yj-yb)
wherein E isVFor moving risks, G is the magnitude of the risk factor between two objects of unit mass per unit distance, RbIs the object b is located (x)b,yb) Road condition influencing factor of (C), TbjCorrection factor r for type between object b and target vehicle jbjIs the distance vector, k, between the object b and the target vehicle j2Is the magnification factor of the distance, k3For the correction of risks of different speeds, vbjIs the relative speed between the object b and the target vehicle j, and theta is rbj、vbjAngle (x) betweenb,yb) Is the coordinate of object b, (x)j,yj) The coordinates of the target vehicle j.
Where the x-axis is located on the road line and the y-axis is perpendicular to the road line.
rbjIs the distance vector of the field source b to the object j.
k2、k3And G are each a constant greater than 0, k2Is the magnification factor of the distance, k3For the hazard correction of different speeds, G is analogous to the electrostatic force constant, which is the magnitude of the risk factor between two objects per unit mass per unit distance and is used to describe the magnitude of the risk factor between two objects per unit mass per unit distance. General k2Has a value range of 0.5 to 1.5, k3The value range of (A) is 0.05-0.2, and G is usually 0.001.
RbIs the object b is located (x)b,yb) The data interval is [0.5, 1.5 ] for the road condition influencing factors]。
TbjThe risk factors, such as car-to-car collision, car-to-person collision, are different for the type of correction factor between the field source b and the object j. Common type correction parameter values are: vehicle-vehicle frontal collision: 2.5-3; vehicle-vehicle rear-end collision: 1 to 1.5; human-vehicle collision: 2 to 2.5; vehicle-barrier collision: 1.5 to 2.
vbjIs the relative velocity between the field source b and the object j, i.e. the velocity v of the field source bbVelocity v of object jjThe vector sum of (1). Theta is vbjAnd rbjThe angle between the directions is positive in the clockwise direction.
EVThe larger the value, the higher the risk that the field source b poses on the object j. The field strength distribution results are shown in fig. 4.
In this embodiment, for the target vehicle, the risk size calculation process for each object is as follows:
1) through early-stage data acquisition, fusion data of the scanning results of the road side radar vision sensing equipment are constructed, and the specific method comprises the following steps:
a. collecting multi-frame data, dividing each frame of data into n statistical spaces, wherein the value range of n can be 50-100 according to different scanning ranges of equipment;
b. sequentially overlapping next frame data from an initial frame, wherein during overlapping, dynamic objects in the next frame data need to be manually removed, and the condition that a shielded black area does not exist in the data is ensured as much as possible; finally, a more ideal global static background is obtained.
And separating static risk field sources in the static scene in a manual mode, wherein the static risk field sources comprise lane separation lines, central separation zones, road edge areas and the like, and randomly sampling and fitting a plane linear equation of each static risk field source. Generally, more than 100 points are required to be uniformly collected along the linear direction of visual observation, and the collected point positions should not deviate from the target too far.
2) And selecting a certain frame of data as the calculation time, and extracting the previous frame of data as the reference of the moving speed of the object. And respectively identifying all target objects (generally vehicles, pedestrians and the like) in the calculation frame and the previous frame data by using a target detection and tracking algorithm, and establishing a corresponding relation between the target objects in the previous frame and the target objects in the next frame. Generally, the operation efficiency of target detection and tracking should be as low as 1.25f as possible, where f is the device scanning frequency.
And calculating the approximate centroid of the target object by using the labeling frame of the target object, and then calculating the displacement and the moving speed of the centroid of the same object in the two frames before and after. And regarding the newly added object with no previous frame data for calculating the speed, and considering the speed as the standard speed. And the standard speed is the average speed in the scanned road section under the historical statistical condition, and the direction of the average speed is consistent with the lane where the target is located. An example of the detection result is shown in fig. 5.
3) One target vehicle is selected as a risk calculation object. The attributes such as the relative position, the relative speed and the type of the calculation object and other target objects, and the parameters such as the distance between the static risk source and the calculation object in the step 1) and the traffic environment factors such as road conditions are sequentially substituted into the driving safety risk field calculation mechanism to obtain the risk of each object to the target vehicle in the scanning range, so that the driving safety risk distribution taking the calculation object as the core is formed, as shown in fig. 6. The vehicle A is a target vehicle, the dotted line represents the plane distance, the position of each upright post is a relatively static object or a relatively moving object, and the height of each upright post represents the risk.
S5: screening objects with the risk higher than a risk threshold value to the target vehicle in a traffic safety risk field and sending out early warning.
Screening out relatively stationary objects with stationary risks greater than a risk threshold value for the target vehicle in S5, acquiring a high risk range of the relatively stationary objects with the risks greater than the risk threshold value, and sending out early warning to the target vehicle; and screening out the relatively moving object with the movement risk larger than the risk threshold value for the target vehicle, acquiring the high risk range of the relatively moving object with the movement risk larger than the risk threshold value, and sending out early warning to the target vehicle.
Specifically, based on the step S4, the relative stationary objects, the stationary risk and the moving risk to the target vehicle in the relative moving object of the area to be warned are obtained,
in this embodiment, the threshold value is generally not less than 45, and is related to the road segment environment. Taking the screened object with the risk greater than the risk threshold as the center, extracting data of a certain area around the object as a risk range, and as shown in fig. 6 and 7, the specific method comprises the following steps:
1) for the relatively static object with the static risk of the target vehicle larger than the risk threshold value, the high risk range is centered on the plane linear equation of the object, and the left and right intercepted widths are daRegion of/2 is the high risk region, daIs the width of the calculation object;
2) for relatively moving objects with the moving risk of the target vehicle larger than the risk threshold, the high risk range is centered on the centroid of the high risk target, and the interception width is 1.5dbThe rectangular region having a length of (0.5l +0.5l × k) is a dangerous range, in which the 0.5l portion is the side length of the half side far from the calculation object, and the 0.5l × k portion is the side length of the half side near the calculation object. dbIs the width of the high risk target and/is the length of the high risk target. k is a speed correction coefficient not less than 1, depends on the speed of the high-risk target, and has a value interval of: v is formed by (0, 30) km/h, and k is 2; v e (30, 70) km/h,k=3;v>70km/h,k=5。
and sequentially extracting high-risk early warning information ranges according to the risk coefficients of the risk sources, wherein the regions with overlapped high-risk ranges are extracted only once. As shown in fig. 8, the final extracted total range result is provided as the warning information to the target vehicle. In this embodiment, the roadside processor is configured to execute the target driving safety risk early warning method based on data fusion of the roadside sensing device, and the obtained early warning information is distributed with the target vehicle through the 4G module.
The above embodiments are merely examples and do not limit the scope of the present invention. These embodiments may be implemented in other various manners, and various omissions, substitutions, and changes may be made without departing from the technical spirit of the present invention.
Claims (10)
1. A target driving safety risk early warning method based on roadside sensing device data fusion is characterized by being used for carrying out risk early warning on a target vehicle and comprising the following steps:
s1: acquiring sensing data of roadside sensing equipment, wherein the roadside sensing equipment is arranged at the roadside of a region to be early-warned, and a target vehicle runs in the region to be early-warned;
s2: performing data fusion on the sensing data acquired by the roadside sensing equipment to acquire fused data;
s3: performing multi-target tracking on the area to be early-warned according to the fusion data;
s4: acquiring a driving safety risk field of a target vehicle in the area to be early warned based on the multi-target tracking result;
s5: screening objects with the risk higher than a risk threshold value to the target vehicle in a traffic safety risk field and sending out early warning.
2. The roadside perception device data fusion-based target driving safety risk early warning method as claimed in claim 1, wherein the driving safety risk fields comprise a static risk field with a risk generation source of a relatively static object in the area to be early warned and a mobile risk field with a risk generation source of a relatively mobile object in the area to be early warned.
3. The roadside perception device data fusion-based target driving safety risk early warning method as claimed in claim 1, wherein the objects to be in relative rest in the area to be early warned comprise lane separation lines and hard separation zones.
4. The target driving safety risk early warning method based on roadside sensing device data fusion as claimed in claim 1, wherein the calculation formula of the static risk of the target vehicle j relative to the static object a in the region to be early warned is as follows:
wherein E isRFor static risk, LTaRisk factor for the type of lane marker, RaIs the object a is located (x)a,ya) The road condition influencing factor, D is the width of the lane, D is the width of the target vehicle j, rajDistance vector between object a and target vehicle j, k1Magnification factor of distance (x)a,ya) Is the coordinate of object a, (x)j,yj) The coordinates of the target vehicle j.
5. The target driving safety risk early warning method based on roadside sensing device data fusion as claimed in claim 1, wherein the calculation formula of the movement risk of the relatively moving object b to the target vehicle j in the region to be early warned is as follows:
rbj=(xj-xb,yj-yb)
wherein E isVFor moving risks, G is the magnitude of the risk factor between two objects of unit mass per unit distance, RbIs the object b is located (x)b,yb) Road condition influencing factor of (C), TbjCorrection factor r for type between object b and target vehicle jbjIs the distance vector, k, between the object b and the target vehicle j2Is the magnification factor of the distance, k3For the correction of risks of different speeds, vbjIs the relative speed between the object b and the target vehicle j, and theta is rbj、vbjAngle (x) betweenb,yb) Is the coordinate of object b, (x)j,yj) The coordinates of the target vehicle j.
6. The target driving safety risk early warning method based on roadside sensing device data fusion as claimed in claim 1, wherein the roadside sensing device comprises a video camera and a millimeter wave radar.
7. The roadside sensing device data fusion-based target driving safety risk early warning method according to claim 6, wherein the step S2 uses a data fusion algorithm to fuse the sensing data of the video camera and the millimeter wave radar to obtain fused data.
8. The roadside sensing device data fusion-based target driving safety risk early warning method according to claim 7, wherein the data fusion algorithm is specifically as follows:
9. The roadside sensing device data fusion-based target driving safety risk early warning method according to claim 1, wherein in step S3, a kalman filter is adopted to perform multi-target tracking on the fusion data to obtain object trajectory data in the area to be early warned.
10. The roadside sensing device data fusion-based target driving safety risk early warning method according to claim 2, wherein in the step S5, a relatively stationary object with a stationary risk greater than a risk threshold for a target vehicle is screened out, a high risk range of the relatively stationary object with a risk greater than the risk threshold is obtained, and an early warning is given to the target vehicle;
in step S5, the relative moving object with the movement risk greater than the risk threshold for the target vehicle is further screened out, and the high risk range of the relative moving object with the movement risk greater than the risk threshold is obtained, and an early warning is given to the target vehicle.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111413903.9A CN114419874B (en) | 2021-11-25 | 2021-11-25 | Target driving safety risk early warning method based on road side sensing equipment data fusion |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111413903.9A CN114419874B (en) | 2021-11-25 | 2021-11-25 | Target driving safety risk early warning method based on road side sensing equipment data fusion |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114419874A true CN114419874A (en) | 2022-04-29 |
CN114419874B CN114419874B (en) | 2023-12-26 |
Family
ID=81265975
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111413903.9A Active CN114419874B (en) | 2021-11-25 | 2021-11-25 | Target driving safety risk early warning method based on road side sensing equipment data fusion |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114419874B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114973768A (en) * | 2022-05-17 | 2022-08-30 | 重庆文理学院 | Road traffic risk early warning method and system based on driving safety field |
CN115359681A (en) * | 2022-07-20 | 2022-11-18 | 贵州大学 | Optimized layout method of roadside structure light cameras supporting automatic driving |
CN115578864A (en) * | 2022-12-08 | 2023-01-06 | 常州星宇车灯股份有限公司 | Early warning control system and method for danger of getting on and off ramps in automatic driving of expressway |
CN116534052A (en) * | 2023-05-10 | 2023-08-04 | 上海保隆汽车科技股份有限公司 | Driving scene perception data processing method, device, controller and storage medium |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104239741A (en) * | 2014-09-28 | 2014-12-24 | 清华大学 | Travelling risk field-based automobile driving safety assistance method |
CN109660965A (en) * | 2018-11-05 | 2019-04-19 | 中国电子科技集团公司第二十八研究所 | A kind of intelligent road side terminal based on edge calculations |
US20200098252A1 (en) * | 2017-05-27 | 2020-03-26 | Huawei Technologies Co., Ltd. | Driving Intention Determining Method and Apparatus |
CN111204336A (en) * | 2020-01-10 | 2020-05-29 | 清华大学 | Vehicle driving risk assessment method and device |
CN112258850A (en) * | 2020-10-16 | 2021-01-22 | 南京莱斯网信技术研究院有限公司 | Edge side multi-sensor data fusion system of vehicle-road cooperative system |
CN112562405A (en) * | 2020-11-27 | 2021-03-26 | 山东高速建设管理集团有限公司 | Radar video intelligent fusion and early warning method and system |
CN112644498A (en) * | 2021-01-05 | 2021-04-13 | 南京航空航天大学 | Intelligent vehicle safety decision-making method based on novel driving safety field |
CN112644486A (en) * | 2021-01-05 | 2021-04-13 | 南京航空航天大学 | Intelligent vehicle obstacle avoidance trajectory planning method based on novel driving safety field |
CN112896185A (en) * | 2021-01-25 | 2021-06-04 | 北京理工大学 | Intelligent driving behavior decision planning method and system for vehicle-road cooperation |
-
2021
- 2021-11-25 CN CN202111413903.9A patent/CN114419874B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104239741A (en) * | 2014-09-28 | 2014-12-24 | 清华大学 | Travelling risk field-based automobile driving safety assistance method |
US20200098252A1 (en) * | 2017-05-27 | 2020-03-26 | Huawei Technologies Co., Ltd. | Driving Intention Determining Method and Apparatus |
CN109660965A (en) * | 2018-11-05 | 2019-04-19 | 中国电子科技集团公司第二十八研究所 | A kind of intelligent road side terminal based on edge calculations |
CN111204336A (en) * | 2020-01-10 | 2020-05-29 | 清华大学 | Vehicle driving risk assessment method and device |
CN112258850A (en) * | 2020-10-16 | 2021-01-22 | 南京莱斯网信技术研究院有限公司 | Edge side multi-sensor data fusion system of vehicle-road cooperative system |
CN112562405A (en) * | 2020-11-27 | 2021-03-26 | 山东高速建设管理集团有限公司 | Radar video intelligent fusion and early warning method and system |
CN112644498A (en) * | 2021-01-05 | 2021-04-13 | 南京航空航天大学 | Intelligent vehicle safety decision-making method based on novel driving safety field |
CN112644486A (en) * | 2021-01-05 | 2021-04-13 | 南京航空航天大学 | Intelligent vehicle obstacle avoidance trajectory planning method based on novel driving safety field |
CN112896185A (en) * | 2021-01-25 | 2021-06-04 | 北京理工大学 | Intelligent driving behavior decision planning method and system for vehicle-road cooperation |
Non-Patent Citations (1)
Title |
---|
王建强;吴剑;李洋;: "基于人-车-路协同的行车风险场概念、原理及建模", 中国公路学报, no. 01 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114973768A (en) * | 2022-05-17 | 2022-08-30 | 重庆文理学院 | Road traffic risk early warning method and system based on driving safety field |
CN115359681A (en) * | 2022-07-20 | 2022-11-18 | 贵州大学 | Optimized layout method of roadside structure light cameras supporting automatic driving |
CN115578864A (en) * | 2022-12-08 | 2023-01-06 | 常州星宇车灯股份有限公司 | Early warning control system and method for danger of getting on and off ramps in automatic driving of expressway |
CN116534052A (en) * | 2023-05-10 | 2023-08-04 | 上海保隆汽车科技股份有限公司 | Driving scene perception data processing method, device, controller and storage medium |
CN116534052B (en) * | 2023-05-10 | 2023-12-05 | 上海保隆汽车科技股份有限公司 | Driving scene perception data processing method, device, controller and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN114419874B (en) | 2023-12-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114419874B (en) | Target driving safety risk early warning method based on road side sensing equipment data fusion | |
CN112700470B (en) | Target detection and track extraction method based on traffic video stream | |
WO2022206942A1 (en) | Laser radar point cloud dynamic segmentation and fusion method based on driving safety risk field | |
US11380105B2 (en) | Identification and classification of traffic conflicts | |
US7046822B1 (en) | Method of detecting objects within a wide range of a road vehicle | |
CN102765365B (en) | Pedestrian detection method based on machine vision and pedestrian anti-collision warning system based on machine vision | |
EP3224819B1 (en) | Method of controlling a traffic surveillance system | |
Zhang et al. | Pedestrian safety analysis in mixed traffic conditions using video data | |
CN110065494A (en) | A kind of vehicle collision avoidance method based on wheel detection | |
CN104129389A (en) | Method for effectively judging and recognizing vehicle travelling conditions and device thereof | |
EP2224743B1 (en) | Vehicle periphery monitoring device, vehicle, and vehicle periphery monitoring program | |
CN111243274A (en) | Road collision early warning system and method for non-internet traffic individuals | |
WO2019213982A1 (en) | Driver control behavior quantification method and device employing principle of least action | |
Wang et al. | Advanced driver‐assistance system (ADAS) for intelligent transportation based on the recognition of traffic cones | |
CN115019514B (en) | Highway inspection system based on car networking | |
CN108108680A (en) | A kind of front vehicle identification and distance measuring method based on binocular vision | |
CN111055852B (en) | Interested target search area determination method for automatic driving | |
Noh et al. | Vision-based overhead front point recognition of vehicles for traffic safety analysis | |
CN114067287A (en) | Foreign matter identification and early warning system based on vehicle side road side data perception fusion | |
CN117292358A (en) | Target recognition method, device, electronic equipment and computer readable storage medium | |
KR20190064147A (en) | Vehicle's behavior analyzing system using aerial photo and analyzing method using the same | |
CN114926729A (en) | High-risk road section identification system and method based on driving video | |
Formosa et al. | Evaluating the impact of lane marking quality on the operation of Autonomous Vehicles | |
CN109325447A (en) | Non-guide based on video analysis travels detection method | |
CN116811884B (en) | Intelligent driving environment perception analysis method and system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |