CN117818608A - Lane merging identification method, system and automobile - Google Patents

Lane merging identification method, system and automobile Download PDF

Info

Publication number
CN117818608A
CN117818608A CN202211213177.0A CN202211213177A CN117818608A CN 117818608 A CN117818608 A CN 117818608A CN 202211213177 A CN202211213177 A CN 202211213177A CN 117818608 A CN117818608 A CN 117818608A
Authority
CN
China
Prior art keywords
lane
vehicle
merging
width
time
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.)
Pending
Application number
CN202211213177.0A
Other languages
Chinese (zh)
Inventor
吴凡
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuhu Bethel Intelligent Driving Co ltd
Original Assignee
Wuhu Bethel Intelligent Driving Co ltd
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 Wuhu Bethel Intelligent Driving Co ltd filed Critical Wuhu Bethel Intelligent Driving Co ltd
Priority to CN202211213177.0A priority Critical patent/CN117818608A/en
Publication of CN117818608A publication Critical patent/CN117818608A/en
Pending legal-status Critical Current

Links

Classifications

    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/18Propelling the vehicle
    • B60W30/18009Propelling the vehicle related to particular drive situations
    • B60W30/18163Lane change; Overtaking manoeuvres
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/02Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to ambient conditions
    • B60W40/06Road conditions

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Traffic Control Systems (AREA)
  • Image Processing (AREA)

Abstract

The invention discloses a lane merging identification method, a lane merging identification system and an automobile, wherein the lane merging identification method comprises the following steps: s1, determining the width of a lane in front of a running lane where a vehicle is located at the time t+1 based on front lane images acquired by monocular cameras at the time t and the time t+1; s2, using the width of the lane in front of driving for lane merging detection and sending out reminding. The lane merging pre-judging method based on the monocular camera is achieved, vehicles can be led to enter a main lane along with the merging line, and safety of vehicle driving is improved.

Description

Lane merging identification method, system and automobile
Technical Field
The invention belongs to the technical field of vehicle auxiliary driving, and particularly relates to a lane merging identification method, a lane merging identification system and an automobile.
Background
The existing vehicle recognizes the lane line through the monocular camera, the recognized lane line has certain random error and steady-state error, under the condition of no high-precision navigation positioning, whether the front side has the lane merging cannot be accurately judged only through the lane line curve equation, and under the condition of no accurate recognition and lane merging in advance, certain uncertainty and potential safety hazard can be added to auxiliary driving. For example, the function suddenly exits when merging, controls the vehicle to leave the own lane, collides with the guard rail, collides with the vehicle on the adjacent lane, and the like.
Disclosure of Invention
The invention provides a lane merging recognition method, which aims to improve the merging recognition precision of a monocular camera.
The invention is realized in such a way, a lane merging recognition method, the method specifically comprises the following steps:
s1, based on a front lane image p acquired by a monocular camera at time t and time t+1 t Determining the width of a lane in which the vehicle is positioned in front of the driving lane;
s2, using the width of the lane in front of driving for lane merging detection and sending out reminding.
Further, the method for determining the width of the lane in front of the driving of the vehicle comprises the following steps:
s11, front image p t Dividing a lane where the vehicle is located into n lane segments along the running direction of the vehicle;
s12, calculating the average value W of the lane width of each lane segment in the n lane segments t [n]And predicting the average value of the lane width of each lane segment at the time t+1
S13, fitting lane line curves of lane lines on two sides of the vehicle, and calculating the lane width z of each lane segment in the n lane segments at the time t+1 t+1 [n];
S14, predicting average value of lane widthLane width value z t+1 [n]Fusing to form the lane width W of each lane segment in the n lane segments at the time t+1 t+1 [n]。
Further, the lane width W of each lane segment at time t+1 t+1 [n]The calculation formula is specifically as follows:
wherein K [ n ]]In order for the confidence level to be high,is the variance of the lane width of n lane segments at time t+1, R [ n ]]Is the measurement error.
Further, the average value of the lane width at time t+1The calculation formula is specifically as follows:
further, the variance of the lane widths of the n lane segments at time t+1The calculation formula is specifically as follows:
wherein Q is the process noise variance, and the process noise variance Q takes the value and the speed v of the vehicle at the moment t ego Yaw rate ψ ego And (5) correlation.
Further, the lane merging detection specifically includes the following steps:
s21, detecting whether the vehicle moves straight or not, wherein the included angle between the vehicle and the lane where the vehicle is positioned is smaller than an angle threshold value, and if the detected result is yes, executing a step S22;
s22, acquiring a target vehicle in front of a lane where the target vehicle is located from a front image acquired by a monocular camera at the time t+1, detecting whether the target vehicle is located within a set distance in front of the vehicle, and if the detection result is negative, executing a step S24;
s24, detecting whether the preceding vehicle and the lane where the preceding vehicle is located meet the following conditions, and if the detection result is yes, sending out a lane merging prompt;
241 The confidence coefficient of the left and right lane lines of the lane where the own vehicle is located is larger than the confidence coefficient threshold value,
242 The curvature of a short lane line in lane lines on two sides of a lane where the vehicle is located is small, and the curvature of a long lane line is large;
243 The two lane lines are crossed;
244 The lane width of the lane section of the lane where the own vehicle is located is in a decreasing trend.
Further, if the target vehicle is positioned in the set distance in front of the vehicle, detecting whether the front vehicle and the lane where the front vehicle is positioned meet the following conditions, and if the detection result is yes, sending a lane merging prompt;
231 A front car is changing lanes;
232 The confidence coefficient of the left lane line and the right lane line of the lane where the vehicle is located is larger than a confidence coefficient threshold value;
233 The curvature of a short lane line in lane lines at two sides of a lane where the vehicle is located is small, and the curvature of a long lane line is large.
Further, the lane width of the lane where the own vehicle is located is identified while the lane widths of the lanes on the left side and the right side of the lane where the own vehicle is located are identified, and when the lane changing operation of the own vehicle is detected, the lane width of the lane where the own vehicle is located is updated to the lane width of the lane where the own vehicle is located after the lane changing operation.
The present invention is embodied in a lane merging recognition system, the system comprising:
the monocular camera collects the front image of the vehicle and sends the front image to the processor;
the processor performs lane merging detection based on the lane merging recognition method.
The invention is realized in that a vehicle is provided with the lane merging recognition system integrated on the vehicle.
The lane merging pre-judging method based on the monocular camera achieves lane merging pre-judging based on the monocular camera, enables vehicles to be led to enter a main lane along with the merging line, and improves safety of vehicle driving.
Drawings
Fig. 1 is a flow chart of a lane merging recognition method according to an embodiment of the present invention.
Detailed Description
The following detailed description of the embodiments of the invention, given by way of example only, is presented in the accompanying drawings to aid in a more complete, accurate, and thorough understanding of the inventive concepts and aspects of the invention by those skilled in the art.
Fig. 1 is a flow chart of a lane merging identification method according to an embodiment of the present invention, where the method specifically includes the following steps:
a first part: based on the front lane images acquired by the monocular cameras at the time t and the time t+1, determining the lane width of the lane where the vehicle is located at the time t+1 in front of running; a second part: the lane width in front of driving is used for lane merging detection and reminding, and the contents of the two parts are described below;
firstly, determining the lane width change of a lane where the vehicle is located in front of driving;
s11, dividing a lane in a front image acquired by a monocular camera at the moment t (the last moment) and the moment t+1 (the current moment) into n lane segments along the running direction of the vehicle, and extracting n lane segments of a lane where the vehicle at the moment t is located;
in the embodiment of the invention, the lane line image is equally divided along the running direction of the vehicle, a vehicle coordinate system is defined, the midpoint of the rear axle of the vehicle is taken as the origin of coordinates, the running direction of the vehicle is taken as the x-axis in the vehicle coordinate system, the width direction of the vehicle is taken as the y-axis in the vehicle coordinate system, and the average width between two lane lines in the lane section is the lane width of the corresponding lane section.
S12, calculating the average value W of the lane width of each lane segment in the n lane segments t [n]Variance P t [n]At the same time based on the lane width average value W at time t t [n]Variance P t [n]To predict the average value of the lane width of each lane segment in n lane segments at time t+1Variance->
Since the time interval of the adjacent time (t, t+1) is short, the lane width mean shift of the adjacent time is small, and based on this, the prediction method is specifically as follows:
wherein Q is the process noise variance, which takes the value and the speed v of the vehicle at the moment t ego Yaw rate ψ ego Correlation, when ψ ego >0.1,Q=(0.01736×V ego ×dt) 2 +0.0003, when ψ e go 0.,Q=(0.01736×V ego ×dt) 2 +0.0008, wherein the maximum variation in lane width that may occur per meter of forward travel of the vehicle is 0.01736,0.0003, 0.0008 is a steady state error, which is always present even if the vehicle is stationary, when ψ ego The stability of visual recognition decreases with increasing, and therefore, by changing the steady-state portion of the process noise variance, the ability of the filter to filter noise can be increased.
S13, fitting a lane line curve of lane lines on two sides of the vehicle at the t moment, and reading the abscissa x of the midpoints of all lane line segments at the t+1 moment under a vehicle coordinate system t+1 [n]Calculating the lane width z of the midpoint of each lane segment at the time t+1 t+1 [n];
In the embodiment of the invention, the fitting method of the lane line curve is specifically as follows:
extracting lane line segments of lane lines at two sides of the vehicle at the moment t, acquiring coordinates of all lane line segments under a vehicle coordinate system, and fitting a lane line curve y of the lane lines at two sides of the vehicle t [n];
Lane line curve y t [n]Respectively lane line curvesLane lineCurve->Lane line curve->Lane line curve->All are multiple curve equations, lane curve +.>Is>Is the same as the fitting method of (a) by using a lane curve +.>For illustration, the expression is specifically as follows:
wherein,an abscissa representing a midpoint of an nth lane line segment on the left side of the time t, and an abscissa representing a distance from an origin of a vehicle coordinate system in a direction in which the lane line extends, +.>The ordinate representing the midpoint of the nth left lane line segment at time t, the ordinate representing the distance from the origin of the vehicle coordinate system in the lane width direction, C 0 、C 1 、C 2 C (C) 3 Representing the fitting coefficients, the same time,/-at the same time, when special explanation is required>
In the embodiment of the invention, the lane width z t+1 [n]The acquisition method of (a) is specifically as follows:
the abscissa x of each lane line segment at the time of t+1 read in the vehicle coordinate system t+1 [n]Respectively input lane line curvesLane line curve->The longitudinal coordinates of the line segments on the two sides of each lane at the time of t+1 can be obtainedThen->
Due to the measurement error existing in the lane line identification processWherein delta n Representing a measured variance of the nth lane segment, the measured variance being related to the distance.
S4, predicting average value based on lane width of each lane segment in the n lane segments at time t+1Lane width calculation value z t+1 [n]To determine the average value W of the width of each lane segment t+1 [n]The lane width of each lane segment at the time t+1 is the lane width.
Wherein, kn is confidence, and the value range is 0-1.
The invention carries out filtering treatment on the lane width information, and before treatment, the lane width identification error is determined according to camera parameters as follows: 0.005+0.00034738 x 2 (where 0.005 is the C0 error and 0.00034738 is the C2 error), i.e. the 50 meter error is 0.8734 meters. After Kalman filtering treatment, the variance formula is as follows: p (t+1) = ((P (t) +q)/(P (t) +q+r), where Q is the process noise variance, R is the measurement error, and the 50 meter error after iterative convergence is 0.1673 meters. Therefore, the reliability of merging judgment can be greatly improved after filtering treatment.
A lane merging detection method;
s21, before lane merging judgment is carried out, whether the vehicle is in straight running or not is firstly detected, the included angle between the vehicle and the lane where the vehicle is located is smaller than an angle threshold value, when the included angle between the vehicle and the lane where the vehicle is located is too large, false alarm is easily generated due to image distortion, if the detected result is yes, the step S22 is executed, and if the detected result is no, lane merging detection is not carried out, namely the step S22 is not executed;
s22, acquiring a target vehicle in front of a lane where the target vehicle is located from a front image acquired by a monocular camera at the time t+1, detecting whether the target vehicle is located within a set distance in front of the vehicle, if so, executing a step S23, and if not, executing a step S24;
s23, detecting whether the preceding vehicle and the lane where the preceding vehicle is located meet the following conditions, and if the detection result is that the following conditions are met, sending out a lane merging prompt;
231 Front car changing lane): the curvature of the front vehicle movement track and the included angle between the head speed direction and the lane line are larger, and the front vehicle steering lamp is turned on or the front vehicle transverse position deviates from the lane;
232 The confidence coefficient of the left lane line and the right lane line of the lane where the vehicle is located is larger than a confidence coefficient threshold value;
233 The curvature of the short lane line in the lane lines on both sides of the lane where the own vehicle is located is small, for example, less than 0.0005 (unit 1/m), and the curvature of the long lane line is large, for example, more than 0.004 (unit 1/m).
S24, detecting whether the preceding vehicle and the lane where the preceding vehicle is located meet the following conditions, and if the detection result is that the following conditions are met, sending out a lane merging prompt;
241 The confidence coefficient of the left and right lane lines of the lane where the own vehicle is located is larger than the confidence coefficient threshold value,
242 The curvature of the short lane line in the lane lines on both sides of the lane where the own vehicle is located is small, for example, less than 0.0005 (unit 1/m), while the curvature of the long lane line is large, for example, more than 0.004 (unit 1/m);
243 The two lane lines are crossed;
244 The lane width of the lane section of the lane where the own vehicle is located is in a decreasing trend.
In the embodiment of the invention, in order to be closer to the actual driving process, the vehicle needs to change the lane in the driving process, in order to ensure that the vehicle can realize lane merging identification after the lane change, the lane width of the lane at the left side and the right side of the lane is identified while the lane width of the lane at which the vehicle is positioned is identified, the identification method is the same as the identification method of the lane at which the vehicle is positioned, and when the lane changing operation of the vehicle is detected, the lane width of the lane at which the vehicle is positioned is updated to the lane width of the lane at which the vehicle is positioned after the lane changing.
The invention also provides a lane merging recognition system, which comprises:
and the monocular camera collects the front image of the vehicle and sends the front image to the processor, and the processor carries out lane merging detection based on the lane merging recognition method.
The invention also provides an automobile, wherein the lane merging recognition system is integrated on the automobile, the monocular camera is integrated on a front windshield, and the processor is independently arranged on the automobile or integrated on a whole automobile controller of the automobile.
While the present invention has been described by way of example, it should be apparent that the practice of the invention is not limited by the foregoing, but rather is intended to cover various insubstantial modifications of the method concepts and teachings of the invention, either as applied to other applications without modification, or as applied directly to other applications, without departing from the scope of the invention.

Claims (10)

1. The lane merging identification method is characterized by comprising the following steps of:
s1, determining the width of a lane in front of a running lane where a vehicle is located at the time t+1 based on front lane images acquired by monocular cameras at the time t and the time t+1;
s2, using the width of the lane in front of driving for lane merging detection and sending out reminding.
2. The lane merging recognition method according to claim 1, wherein the method of determining the lane width of the lane in which the host vehicle is located in front of the travel comprises the steps of:
s11, dividing a lane where a host vehicle is located in a front image into n lane segments along the running direction of the vehicle;
s12, calculating the average value W of the lane width of each lane segment in the n lane segments t [n]And predicting the average value of the lane width of each lane segment at the time t+1
S13, fitting lane line curves of lane lines on two sides of the vehicle, and calculating the lane width z of each lane segment in the n lane segments at the time t+1 t+1 [n];
S14, predicting average value of lane widthLane width value z t+1 [n]Fusing to form the lane width W of each lane segment in the n lane segments at the time t+1 t+1 [n]。
3. The lane merging recognition method according to claim 2, wherein the lane width W of each lane segment at time t+1 t+1 [n]The calculation formula is specifically as follows:
wherein K [ n ]]In order for the confidence level to be high,is the variance of the lane width of n lane segments at time t+1, R [ n ]]Is the measurement error.
4. A lane merging recognition method according to claim 2 or 3, wherein the average value of the lane widths at time t+1The calculation formula is specifically as follows:
5. the lane merging recognition method as claimed in claim 3, wherein the variance of the lane widths of n lane segments at time t+1The calculation formula is specifically as follows:
wherein Q is the process noise variance, and the process noise variance Q takes the value and the speed v of the vehicle at the moment t ego Yaw rate ψ ego And (5) correlation.
6. The lane merging recognition method as claimed in claim 1, wherein the lane merging detection comprises the steps of:
s21, detecting whether the vehicle moves straight or not, wherein the included angle between the vehicle and the lane where the vehicle is positioned is smaller than an angle threshold value, and if the detected result is yes, executing a step S22;
s22, acquiring a target vehicle in front of a lane where the target vehicle is located from a front image acquired by a monocular camera at the time t+1, detecting whether the target vehicle is located within a set distance in front of the vehicle, and if the detection result is negative, executing a step S24;
s24, detecting whether the preceding vehicle and the lane where the preceding vehicle is located meet the following conditions, and if the detection result is yes, sending out a lane merging prompt;
241 The confidence coefficient of the left and right lane lines of the lane where the own vehicle is located is larger than the confidence coefficient threshold value,
242 The curvature of a short lane line in lane lines on two sides of a lane where the vehicle is located is small, and the curvature of a long lane line is large;
243 The two lane lines are crossed;
244 The lane width of the lane section of the lane where the own vehicle is located is in a decreasing trend.
7. The lane merging recognition method according to claim 1, wherein if a target vehicle is located within a set distance in front of the host vehicle, whether the preceding vehicle and the lane where the target vehicle is located meet the following conditions is detected, and if the detection result is yes, a merging reminder is sent;
231 A front car is changing lanes;
232 The confidence coefficient of the left lane line and the right lane line of the lane where the vehicle is located is larger than a confidence coefficient threshold value;
233 The curvature of a short lane line in lane lines at two sides of a lane where the vehicle is located is small, and the curvature of a long lane line is large.
8. The lane-merging recognition method according to claim 1, wherein the lane width of the lane in which the own vehicle is located is recognized while the lane widths of the lanes on the left and right sides of the lane in which the own vehicle is located are recognized, and when the lane-changing operation of the own vehicle is detected, the lane width of the lane in which the own vehicle is located is updated to the lane width of the lane in which the own vehicle is located after the lane change.
9. A lane merging recognition system, the system comprising:
the monocular camera collects the front image of the vehicle and sends the front image to the processor;
the processor performs lane merging detection based on the lane merging recognition method according to any one of claims 1 to 8.
10. An automobile having integrated thereon the lane merging recognition system of claim 9.
CN202211213177.0A 2022-09-29 2022-09-29 Lane merging identification method, system and automobile Pending CN117818608A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211213177.0A CN117818608A (en) 2022-09-29 2022-09-29 Lane merging identification method, system and automobile

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211213177.0A CN117818608A (en) 2022-09-29 2022-09-29 Lane merging identification method, system and automobile

Publications (1)

Publication Number Publication Date
CN117818608A true CN117818608A (en) 2024-04-05

Family

ID=90504569

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211213177.0A Pending CN117818608A (en) 2022-09-29 2022-09-29 Lane merging identification method, system and automobile

Country Status (1)

Country Link
CN (1) CN117818608A (en)

Similar Documents

Publication Publication Date Title
JP4420011B2 (en) Object detection device
US7557732B2 (en) Vehicle deviation preventing control device
JP4906398B2 (en) In-vehicle road shape identification device, in-vehicle system, road shape identification method and periphery monitoring method
US6292752B1 (en) Device for acquiring lane path indicative data
US9704391B2 (en) Traffic accident detection device and method of detecting traffic accident
KR101545478B1 (en) Sysyem and method for front detecting of vehicle
US20050278112A1 (en) Process for predicting the course of a lane of a vehicle
US11243080B2 (en) Self-position estimation method and self-position estimation device
JP7279053B2 (en) System and method for detecting collision risk between a motor vehicle and a secondary object in the driving lane next to the vehicle when changing lanes
JP2009073462A (en) Operating state determination device and operation support device
CN109017776B (en) Driving assistance system mounted in vehicle
JP2013020458A (en) On-vehicle object discrimination device
JP2008117073A (en) Interruption vehicle detection device
JP3508213B2 (en) Vehicle running state determination device
JP6207952B2 (en) Leading vehicle recognition device
CN110940982B (en) Vehicle front destination identification method and corresponding device
CN110940981B (en) Method for judging whether position of object in front of vehicle is in own lane
KR101417409B1 (en) System and method for preventing accident in parking lot
KR20130051375A (en) Apparatus for compensating vehicle lane of lane keeping assistance system and method
CN117818608A (en) Lane merging identification method, system and automobile
JP3930309B2 (en) Vehicle driving support device
CN114084133B (en) Method and related device for determining following target
JP5682302B2 (en) Traveling road estimation device, method and program
KR100901414B1 (en) System for Preventing Lane Secession
CN113386773A (en) Method and device for judging reliability of visual identification

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