Commercial vehicle platooning following method based on camera perception decision and path planning

Abstract

The application discloses a kind of based on camera perception decision and path planning's commercial vehicle queuing follow-up method, including camera perception evaluation index unit, lane changing decision unit, lane changing path planning unit, queuing vehicle distance keeping unit.Camera perception evaluation index unit is used to calculate spatial distance evaluation factor, traffic participant motion and posture evaluation factor, traffic sign evaluation factor;Lane changing decision unit includes three kinds of lane changing instructions: follow-up travel, wait for lane changing, execute lane changing;Lane changing path planning unit is used to receive the lane changing willingness instruction sent by lane changing decision unit, and after obtaining one of follow-up travel instruction, wait for lane changing instruction, execute lane changing instruction three kinds of instructions, the path cluster of next period is planned according to path planning algorithm;Queuing vehicle distance keeping unit is used to calculate detection distance comprehensive coefficient, accelerator pedal intensity, brake pedal intensity, and vehicle distance is kept in safe range by controlling accelerator pedal intensity and brake pedal intensity.

Description

Technical Field

[0001] This invention relates to a method for commercial vehicle platooning and following based on camera perception decision-making and path planning. Background Technology

[0002] Research on commercial vehicle platooning primarily focuses on reducing wind resistance, fuel consumption, driver workload, and traffic safety. Optimizing vehicle spacing and platooning patterns can significantly reduce the drag coefficient during operation, thereby decreasing fuel consumption and emissions. Currently, vehicles are becoming increasingly intelligent, with autonomous driving technology emerging. Domestic and international commercial vehicle manufacturers and logistics companies are actively exploring and researching commercial vehicle platooning and following technologies, achieving some progress. However, the evaluation metrics for camera-based perception decision-making and path planning methods for commercial vehicle platooning and following are relatively singular. For optimizing vehicle spacing, camera-based perception decision-making and path planning methods for commercial vehicle platooning and following need to consider spatial distance evaluation factors, traffic participant motion and attitude evaluation factors, and traffic sign evaluation factors. Therefore, how to better optimize vehicle spacing has become a pressing technical problem for the applicant. To address these issues, this invention proposes a commercial vehicle platooning and following method based on camera-based perception decision-making and path planning. Summary of the Invention

[0003] The purpose of this invention is to provide a commercial vehicle platooning following method based on camera perception decision-making and path planning, so as to solve the problems faced in the above-mentioned background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a commercial vehicle platooning following method based on camera perception decision-making and path planning, including a camera perception evaluation index unit, a lane change decision unit, a lane change path planning unit, and a platooning distance maintenance unit.

[0005] The commercial vehicles mentioned are driverless vehicles, and the convoy consists of three vehicles.

[0006] The camera perception evaluation index unit is used to calculate the spatial distance evaluation factor K1, the motion and attitude evaluation factor K2 of traffic participants, and the traffic sign evaluation factor K3, thereby calculating the camera perception evaluation index K through the evaluation index formula. The spatial distance evaluation factor K1 depends on the speed v1 of the first vehicle, the speed v2 of the second vehicle, the speed v3 of the third vehicle, and the spatial coordinate transformation factor β. The spatial coordinate transformation factor β depends on the center coordinates (X1, Y1, Z1) of the first vehicle in the geodetic coordinate system, the center coordinates (X2, Y2, Z2) of the second vehicle in the geodetic coordinate system, and the center coordinates (X3, Y3, Z3) of the third vehicle in the geodetic coordinate system. The motion and attitude evaluation factor K2 depends on the speed v1 of the first vehicle, the speed v2 of the second vehicle, the speed v3 of the third vehicle, the acceleration a1 of the first vehicle, the acceleration a2 of the second vehicle, and the acceleration a3 of the third vehicle. The traffic sign evaluation factor K3 depends on the color conversion factor G. ray The road surface noise evaluation comprehensive coefficient N and the road condition evaluation coefficient E, where the color conversion factor G is also included. ray It depends on the three-channel color parameters, namely the first channel color parameter R, the second channel color parameter G, and the third channel color parameter B;

[0007] The lane-changing decision unit includes three lane-changing instructions: following, waiting for a lane change, and executing a lane change. Based on the principle of safety first, the following instruction has the highest priority, the waiting instruction has the next highest priority, and the executing instruction is the lowest priority. The strength of the lane-changing intention is described by designing following intention thresholds δ1, waiting intention thresholds δ2, and executing intention thresholds δ3, where 0 ≤ δ1 < δ2 < δ3 ≤ 1. When the camera perception evaluation index K satisfies 0 ≤ K < δ2, the lane-changing decision unit executes the following instruction; when the camera perception evaluation index K satisfies δ2 ≤ K < δ3, the lane-changing decision unit executes the waiting instruction; and when the camera perception evaluation index K satisfies δ3 ≤ K ≤ 1, the lane-changing decision unit executes the lane-changing instruction.

[0008] The lane-changing path planning unit is used to receive the lane-changing intention instruction sent by the lane-changing decision unit. After receiving one of the three instructions—following the car, waiting for the lane-changing instruction, or executing the lane-changing instruction—it plans the path cluster for the next time period according to the path planning algorithm. Then, it calculates the cost estimate G(x) of the planned path cluster through the cost function H(x). Under the premise of ensuring safe driving, it selects the path with the smallest G(x) as the optimal planned path.

[0009] The platooning distance maintaining unit is used to calculate the detection distance comprehensive coefficient μ, accelerator pedal intensity S, and brake pedal intensity T. The detection distance comprehensive coefficient μ depends on the sensor sensitivity coefficient ε, the coordinate comprehensive judgment coefficient τ, the number of cameras mounted in front of the vehicle n1, and the number of cameras mounted behind the vehicle n2. Among them, the coordinate comprehensive judgment coefficient τ depends on the center coordinate (X) of the first vehicle in the vehicle coordinate system. V1 Y V1 Z V1 ), the center coordinates (X) of the second vehicle in the vehicle coordinate system V2 Y V2 Z V2 ), the center coordinates (X) of the third vehicle in the vehicle coordinate system V3 Y V3 Z V3 The accelerator pedal strength S depends on the actual distance b1 between the first and second vehicles, the actual distance b2 between the second and third vehicles, the speed v1 of the first vehicle, the speed v2 of the second vehicle, and the speed v3 of the third vehicle; the brake pedal strength T depends on the actual distance b1 between the first and second vehicles, the actual distance b2 between the second and third vehicles, the speed v1 of the first vehicle, the speed v2 of the second vehicle, and the speed v3 of the third vehicle.

[0010] The aforementioned commercial vehicle platooning and following method based on camera perception decision-making and path planning is characterized by the following: after completing path planning, it is necessary to control the platooning distance of the commercial vehicles. When controlling the vehicle distance, it is necessary to calculate the detection distance comprehensive coefficient μ, and then calculate the accelerator pedal intensity S and brake pedal intensity T required to maintain the distance based on the detection distance comprehensive coefficient μ. The platooning distance maintaining unit can calculate the detection distance comprehensive coefficient μ according to the following formula:

[0011]

[0012] Where, ω 10 ω 11 Here, ε is the weighting coefficient, τ is the sensor sensitivity coefficient, n1 is the number of cameras mounted at the front of the vehicle, n2 is the number of cameras mounted at the rear of the vehicle, h1 is the ground clearance of the cameras mounted at the front of the vehicle, h2 is the ground clearance of the cameras mounted at the rear of the vehicle, γ1 is the top-down angle of the cameras mounted at the front of the vehicle, γ2 is the top-down angle of the cameras mounted at the rear of the vehicle, θ1 is the bottom-up angle of the cameras mounted at the front of the vehicle, θ2 is the bottom-up angle of the cameras mounted at the rear of the vehicle, and the expression for the coordinate comprehensive judgment coefficient τ is as follows:

[0013]

[0014] Among them, (X) V1 Y V1 ZV1 (X) represents the center coordinates of the first vehicle in the vehicle coordinate system. V2 Y V2 Z V2 (X) represents the center coordinates of the second vehicle in the vehicle coordinate system. V3 Y V3 Z V3 () represents the center coordinates of the third vehicle in the vehicle coordinate system.

[0015] The aforementioned commercial vehicle platooning and following method based on camera perception decision-making and path planning is characterized in that: the camera perception evaluation index unit can calculate the spatial distance evaluation factor K1 according to the following formula:

[0016]

[0017] Where e is a natural constant, v1 is the speed of the first vehicle, v2 is the speed of the second vehicle, v3 is the speed of the third vehicle, and β is a spatial coordinate transformation factor, the value of which depends on the center coordinates of the first vehicle (X1, Y1, Z1), the center coordinates of the second vehicle (X2, Y2, Z2), and the center coordinates of the third vehicle (X3, Y3, Z3) in the geodetic coordinate system. The expression for the spatial coordinate transformation factor β is as follows.

[0018]

[0019] Where ω1, ω2, and ω3 are weighting coefficients.

[0020] The aforementioned commercial vehicle platooning and following method based on camera perception decision-making and path planning is characterized in that: the camera perception evaluation index unit can calculate the motion and attitude evaluation factor K2 of traffic participants according to the following formula:

[0021]

[0022] Wherein, ω4, ω5, and ω6 are weighting coefficients, v1 is the speed of the first vehicle, v2 is the speed of the second vehicle, v3 is the speed of the third vehicle, a1 is the acceleration of the first vehicle, a2 is the acceleration of the second vehicle, a3 is the acceleration of the third vehicle, i0 is the longitudinal slope of the road, i1 is the cross slope of the road, and L0 is the inherent brightness contrast of traffic participants.

[0023] The aforementioned commercial vehicle platooning and following method based on camera perception decision-making and path planning is characterized in that: the camera perception evaluation index unit can calculate the traffic sign evaluation factor K3 according to the following formula:

[0024]

[0025] Where e is the natural constant, ω7, ω8, and ω9 are weighting coefficients, and G... ray Where N is the color conversion factor, E is the comprehensive evaluation coefficient for road surface noise, and G is the road condition evaluation coefficient. ray It depends on the three-channel color parameters, where R is the first channel color parameter, G is the second channel color parameter, and B is the third channel color parameter. ray The expression is as follows:

[0026]

[0027] Where e is the natural constant.

[0028] The commercial vehicle platooning following method based on camera perception decision-making and path planning is characterized in that: the camera perception evaluation index unit can calculate the camera perception evaluation index K according to the following formula:

[0029]

[0030] Among them, K1 is the spatial distance evaluation factor, K2 is the motion and posture evaluation factor of traffic participants, and K3 is the traffic sign evaluation factor.

[0031] The commercial vehicle platooning following method based on camera perception decision-making and path planning is characterized in that: the lane-changing decision unit includes three lane-changing instructions: following, waiting for lane changing, and executing lane changing; according to the principle of safety first, the following instruction is given priority, the waiting instruction is given secondary priority, and the executing instruction is given last priority. The strength of the lane-changing intention is described by designing a following intention threshold δ1, a waiting intention threshold δ2, and an executing intention threshold δ3, where 0≤δ1<δ2<δ3≤1.

[0032] The commercial vehicle platooning following method based on camera perception decision-making and path planning is characterized by the following: After receiving the lane-changing intention instruction sent by the lane-changing decision unit, the lane-changing path planning unit plans a path cluster for the next time period according to the path planning algorithm. Then, it calculates the cost estimate G(x) of the planned path cluster through the cost function H(x). Under the premise of ensuring safe vehicle driving, the path with the smallest G(x) is selected as the optimal planned path. The expression for the cost estimate G(x) of the path cluster is as follows:

[0033] G(x)=c((d x d y ), H(d x d y H(x),

[0034] Wherein, c((d) x d y), H(d x d y )) represents the weighted value of the cost estimate, (d x d y H(d) represents the coordinates of station d. x d y Let H(x) be the cost function value of site d, and let H(x) be the cost function. The expression for the evaluation function value of site d is as follows:

[0035] H(d x d y )=s(d x d y )+t(d x d y ),

[0036] Among them, s(d x d y t(d) represents the cost function value from the starting point to station d. x d y ) represents the cost function value from station d to the destination.

[0037] The aforementioned commercial vehicle platooning and following method based on camera perception decision-making and path planning is characterized in that: after completing path planning, it is necessary to control the platooning distance of the commercial vehicles. The measures implemented by the vehicle mechanism are to control the accelerator pedal intensity S and the brake pedal intensity T. The platooning distance maintaining unit can calculate the accelerator pedal intensity S according to the following formula:

[0038]

[0039] e is a natural constant, b1 is the actual distance between the first and second vehicles, b2 is the actual distance between the second and third vehicles, v1 is the speed of the first vehicle, v2 is the speed of the second vehicle, v3 is the speed of the third vehicle, and μ is the comprehensive coefficient of detection distance.

[0040] The aforementioned commercial vehicle platooning and following method based on camera perception decision-making and path planning is characterized in that: after completing path planning, it is necessary to control the platooning distance of the commercial vehicles. The measures implemented by the vehicle mechanism are to control the accelerator pedal intensity S and the brake pedal intensity T. The platooning distance maintaining unit can calculate the brake pedal intensity T according to the following formula:

[0041]

[0042] e is a natural constant, b1 is the actual distance between the first and second vehicles, b2 is the actual distance between the second and third vehicles, v1 is the speed of the first vehicle, v2 is the speed of the second vehicle, v3 is the speed of the third vehicle, and μ is the comprehensive coefficient of detection distance.

[0043] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0044] 1. A commercial vehicle platooning and following method based on camera perception decision-making and path planning obtains camera perception evaluation indicators based on spatial distance evaluation factors, traffic participant motion and attitude evaluation factors, and traffic sign evaluation factors.

[0045] 2. Determine the range of the lane change intention threshold based on the camera perception evaluation index, and then execute one of the three lane change intentions: following, waiting for lane change, or executing lane change. After receiving the corresponding lane change instruction, select the path with the smallest cost estimate in the path cluster as the optimal planned path for driving. After completing the path planning, calculate the detection distance comprehensive coefficient, and then calculate the accelerator pedal strength and brake pedal strength required to maintain the target distance based on the detection distance comprehensive coefficient. Attached Figure Description

[0046] The invention will be further described below with reference to the accompanying drawings:

[0047] Figure 1 This invention proposes a method for commercial vehicle platooning and following based on camera perception decision-making and path planning. Detailed Implementation

[0048] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0049] like Figure 1 As shown, the present invention is a commercial vehicle platooning following method based on camera perception decision-making and path planning, including a camera perception evaluation index unit, a lane change decision unit, a lane change path planning unit, and a platooning distance maintenance unit.

[0050] The commercial vehicles mentioned are driverless vehicles, and the convoy consists of three vehicles.

[0051] The camera perception evaluation index unit is used to calculate the spatial distance evaluation factor K1, the motion and attitude evaluation factor K2 of traffic participants, and the traffic sign evaluation factor K3, thereby calculating the camera perception evaluation index K through the evaluation index formula. The spatial distance evaluation factor K1 depends on the speed v1 of the first vehicle, the speed v2 of the second vehicle, the speed v3 of the third vehicle, and the spatial coordinate transformation factor β. The spatial coordinate transformation factor β depends on the center coordinates (X1, Y1, Z1) of the first vehicle in the geodetic coordinate system, the center coordinates (X2, Y2, Z2) of the second vehicle in the geodetic coordinate system, and the center coordinates (X3, Y3, Z3) of the third vehicle in the geodetic coordinate system. The motion and attitude evaluation factor K2 depends on the speed v1 of the first vehicle, the speed v2 of the second vehicle, the speed v3 of the third vehicle, the acceleration a1 of the first vehicle, the acceleration a2 of the second vehicle, and the acceleration a3 of the third vehicle. The traffic sign evaluation factor K3 depends on the color conversion factor G. ray The road surface noise evaluation comprehensive coefficient N and the road condition evaluation coefficient E, where the color conversion factor G is also included. ray It depends on the three-channel color parameters, namely the first channel color parameter R, the second channel color parameter G, and the third channel color parameter B;

[0052] The lane-changing decision unit includes three lane-changing instructions: following, waiting for a lane change, and executing a lane change. Based on the principle of safety first, the following instruction has the highest priority, the waiting instruction has the next highest priority, and the executing instruction is the lowest priority. The strength of the lane-changing intention is described by designing following intention thresholds δ1, waiting intention thresholds δ2, and executing intention thresholds δ3, where 0 ≤ δ1 < δ2 < δ3 ≤ 1. When the camera perception evaluation index K satisfies 0 ≤ K < δ2, the lane-changing decision unit executes the following instruction; when the camera perception evaluation index K satisfies δ2 ≤ K < δ3, the lane-changing decision unit executes the waiting instruction; and when the camera perception evaluation index K satisfies δ3 ≤ K ≤ 1, the lane-changing decision unit executes the lane-changing instruction.

[0053] The lane-changing path planning unit is used to receive the lane-changing intention instruction sent by the lane-changing decision unit. After receiving one of the three instructions—following the car, waiting for the lane-changing instruction, or executing the lane-changing instruction—it plans the path cluster for the next time period according to the path planning algorithm. Then, it calculates the cost estimate G(x) of the planned path cluster through the cost function H(x). Under the premise of ensuring safe driving, it selects the path with the smallest G(x) as the optimal planned path.

[0054] The platooning distance maintaining unit is used to calculate the detection distance comprehensive coefficient μ, accelerator pedal intensity S, and brake pedal intensity T. The detection distance comprehensive coefficient μ depends on the sensor sensitivity coefficient ε, the coordinate comprehensive judgment coefficient τ, the number of cameras mounted in front of the vehicle n1, and the number of cameras mounted behind the vehicle n2. Among them, the coordinate comprehensive judgment coefficient τ depends on the center coordinate (X) of the first vehicle in the vehicle coordinate system. V1 Y V1 Z V1 ), the center coordinates (X) of the second vehicle in the vehicle coordinate system V2 Y V2 Z V2 ), the center coordinates (X) of the third vehicle in the vehicle coordinate system V3 Y V3 Z V3 The accelerator pedal strength S depends on the actual distance b1 between the first and second vehicles, the actual distance b2 between the second and third vehicles, the speed v1 of the first vehicle, the speed v2 of the second vehicle, and the speed v3 of the third vehicle; the brake pedal strength T depends on the actual distance b1 between the first and second vehicles, the actual distance b2 between the second and third vehicles, the speed v1 of the first vehicle, the speed v2 of the second vehicle, and the speed v3 of the third vehicle.

[0055] The aforementioned commercial vehicle platooning and following method based on camera perception decision-making and path planning is characterized by the following: after completing path planning, it is necessary to control the platooning distance of the commercial vehicles. When controlling the vehicle distance, it is necessary to calculate the detection distance comprehensive coefficient μ, and then calculate the accelerator pedal intensity S and brake pedal intensity T required to maintain the distance based on the detection distance comprehensive coefficient μ. The platooning distance maintaining unit can calculate the detection distance comprehensive coefficient μ according to the following formula:

[0056]

[0057] Where, ω 10 ω 11 Here, ε is the weighting coefficient, τ is the sensor sensitivity coefficient, n1 is the number of cameras mounted at the front of the vehicle, n2 is the number of cameras mounted at the rear of the vehicle, h1 is the ground clearance of the cameras mounted at the front of the vehicle, h2 is the ground clearance of the cameras mounted at the rear of the vehicle, γ1 is the top-down angle of the cameras mounted at the front of the vehicle, γ2 is the top-down angle of the cameras mounted at the rear of the vehicle, θ1 is the bottom-up angle of the cameras mounted at the front of the vehicle, θ2 is the bottom-up angle of the cameras mounted at the rear of the vehicle, and the expression for the coordinate comprehensive judgment coefficient τ is as follows:

[0058]

[0059] Among them, (X) V1 Y V1 ZV1 (X) represents the center coordinates of the first vehicle in the vehicle coordinate system. V2 Y V2 Z V2 (X) represents the center coordinates of the second vehicle in the vehicle coordinate system. V3 Y V3 Z V3 () represents the center coordinates of the third vehicle in the vehicle coordinate system.

[0060] The aforementioned commercial vehicle platooning and following method based on camera perception decision-making and path planning is characterized in that: the camera perception evaluation index unit can calculate the spatial distance evaluation factor K1 according to the following formula:

[0061]

[0062] Where e is a natural constant, v1 is the speed of the first vehicle, v2 is the speed of the second vehicle, v3 is the speed of the third vehicle, and β is a spatial coordinate transformation factor, the value of which depends on the center coordinates of the first vehicle (X1, Y1, Z1), the center coordinates of the second vehicle (X2, Y2, Z2), and the center coordinates of the third vehicle (X3, Y3, Z3) in the geodetic coordinate system. The expression for the spatial coordinate transformation factor β is as follows.

[0063]

[0064] Where ω1, ω2, and ω3 are weighting coefficients.

[0065] The aforementioned commercial vehicle platooning and following method based on camera perception decision-making and path planning is characterized in that: the camera perception evaluation index unit can calculate the motion and attitude evaluation factor K2 of traffic participants according to the following formula:

[0066]

[0067] Wherein, ω4, ω5, and ω6 are weighting coefficients, v1 is the speed of the first vehicle, v2 is the speed of the second vehicle, v3 is the speed of the third vehicle, a1 is the acceleration of the first vehicle, a2 is the acceleration of the second vehicle, a3 is the acceleration of the third vehicle, i0 is the longitudinal slope of the road, i1 is the cross slope of the road, and L0 is the inherent brightness contrast of traffic participants.

[0068] The aforementioned commercial vehicle platooning and following method based on camera perception decision-making and path planning is characterized in that: the camera perception evaluation index unit can calculate the traffic sign evaluation factor K3 according to the following formula:

[0069]

[0070] Where e is the natural constant, ω7, ω8, and ω9 are weighting coefficients, and G... ray Where N is the color conversion factor, E is the comprehensive evaluation coefficient for road surface noise, and G is the road condition evaluation coefficient. ray It depends on the three-channel color parameters, where R is the first channel color parameter, G is the second channel color parameter, and B is the third channel color parameter. ray The expression is as follows:

[0071]

[0072] Where e is the natural constant.

[0073] The commercial vehicle platooning following method based on camera perception decision-making and path planning is characterized in that: the camera perception evaluation index unit can calculate the camera perception evaluation index K according to the following formula:

[0074]

[0075] Among them, K1 is the spatial distance evaluation factor, K2 is the motion and posture evaluation factor of traffic participants, and K3 is the traffic sign evaluation factor.

[0076] The commercial vehicle platooning following method based on camera perception decision-making and path planning is characterized in that: the lane-changing decision unit includes three lane-changing instructions: following, waiting for lane changing, and executing lane changing; according to the principle of safety first, the following instruction is given priority, the waiting instruction is given secondary priority, and the executing instruction is given last priority. The strength of the lane-changing intention is described by designing a following intention threshold δ1, a waiting intention threshold δ2, and an executing intention threshold δ3, where 0≤δ1<δ2<δ3≤1.

[0077] The commercial vehicle platooning following method based on camera perception decision-making and path planning is characterized by the following: After receiving the lane-changing intention instruction sent by the lane-changing decision unit, the lane-changing path planning unit plans a path cluster for the next time period according to the path planning algorithm. Then, it calculates the cost estimate G(x) of the planned path cluster through the cost function H(x). Under the premise of ensuring safe vehicle driving, the path with the smallest G(x) is selected as the optimal planned path. The expression for the cost estimate G(x) of the path cluster is as follows:

[0078] G(x)=c((d x d y ), H(d x d y H(x),

[0079] Wherein, c((d) x d y), H(d x d y )) represents the weighted value of the cost estimate, (d x d y H(d) represents the coordinates of station d. x d y Let H(x) be the cost function value of site d, and let H(x) be the cost function. The expression for the evaluation function value of site d is as follows:

[0080] H(d x d y )=s(d x d y )+t(d x d y ),

[0081] Among them, s(d x d y t(d) represents the cost function value from the starting point to station d. x d y ) represents the cost function value from station d to the destination.

[0082] The aforementioned commercial vehicle platooning and following method based on camera perception decision-making and path planning is characterized in that: after completing path planning, it is necessary to control the platooning distance of the commercial vehicles. The measures implemented by the vehicle mechanism are to control the accelerator pedal intensity S and the brake pedal intensity T. The platooning distance maintaining unit can calculate the accelerator pedal intensity S according to the following formula:

[0083]

[0084] e is a natural constant, b1 is the actual distance between the first and second vehicles, b2 is the actual distance between the second and third vehicles, v1 is the speed of the first vehicle, v2 is the speed of the second vehicle, v3 is the speed of the third vehicle, and μ is the comprehensive coefficient of detection distance.

[0085] The aforementioned commercial vehicle platooning and following method based on camera perception decision-making and path planning is characterized in that: after completing path planning, it is necessary to control the platooning distance of the commercial vehicles. The measures implemented by the vehicle mechanism are to control the accelerator pedal intensity S and the brake pedal intensity T. The platooning distance maintaining unit can calculate the brake pedal intensity T according to the following formula:

[0086]

[0087] e is a natural constant, b1 is the actual distance between the first and second vehicles, b2 is the actual distance between the second and third vehicles, v1 is the speed of the first vehicle, v2 is the speed of the second vehicle, v3 is the speed of the third vehicle, and μ is the comprehensive coefficient of detection distance.

Claims

1. A method for commercial vehicle platooning and following based on camera perception decision-making and path planning, characterized in that, Includes the following: It includes a camera perception evaluation index unit, a lane change decision unit, a lane change path planning unit, and a platoon distance maintenance unit. The commercial vehicles mentioned are driverless vehicles, and the convoy consists of three vehicles. The camera perception evaluation index unit is used to calculate the spatial distance evaluation factor. Evaluation factors of motion and posture of traffic participants Traffic sign evaluation factors Therefore, the camera perception evaluation index can be calculated using the evaluation index formula. Spatial distance evaluation factor Depending on the speed of the first car The speed of the second car The speed of the third car Spatial coordinate transformation factor Among them, spatial coordinate transformation factor Depends on the center coordinates of the first vehicle in the geodetic coordinate system ( , , ), the center coordinates of the second vehicle in the geodetic coordinate system ( , , ), the center coordinates of the third vehicle in the geodetic coordinate system ( , , ); Evaluation factors of motion and posture of traffic participants Depending on the speed of the first car The speed of the second car The speed of the third car The acceleration of the first car The acceleration of the second car The acceleration of the third car Traffic sign evaluation factors Depends on color conversion factor Comprehensive evaluation coefficient of road surface noise Road condition evaluation coefficient Among them, color conversion factor It depends on the three-channel color parameters, namely the first channel color parameters. Second channel color parameters Third channel color parameters ; The lane-changing decision unit includes three lane-changing instructions: following, waiting to change lanes, and executing a lane change. Based on the principle of safety first, the following instruction has the highest priority, the waiting instruction has the next highest priority, and the executing instruction is the lowest priority. A threshold for the willingness to follow is designed to address this. Waiting for lane-changing willingness threshold , threshold for lane-changing intention To describe the intensity of the willingness to change lanes, among which When the camera perception evaluation index K satisfies At that time, the lane-changing decision unit executes the following driving command, and when the camera perception evaluation index K meets the requirements... When the lane-changing decision unit executes the waiting lane-changing instruction, and the camera perception evaluation index K is satisfied... At that time, the lane-changing decision unit executes the lane-changing command; The lane-changing path planning unit receives lane-changing intention instructions from the lane-changing decision unit. Upon receiving one of three instructions—following, waiting for a lane-changing instruction, or executing a lane-changing instruction—it plans a path cluster for the next time period based on a path planning algorithm, and then uses a cost function. Calculate the cost estimate for the planned path cluster. Select, under the premise of ensuring safe driving of the vehicle The shortest path is taken as the optimal planned path. The platoon spacing maintenance unit is used to calculate the comprehensive coefficient of detection distance. Accelerator pedal strength Brake pedal strength Detection range comprehensive coefficient Depends on the sensor sensitivity coefficient Coordinate comprehensive judgment coefficient Number of cameras mounted on the front of the vehicle Number of rear-facing cameras on the vehicle Among them, the coordinate comprehensive judgment coefficient Depends on the center coordinates of the first vehicle in the vehicle coordinate system ( , , ), the center coordinates of the second vehicle in the vehicle coordinate system ( , , ), the center coordinates of the third vehicle in the vehicle coordinate system ( , , ); Accelerator pedal strength Depends on the actual distance between the first and second vehicles The actual distance between the second and third vehicles The speed of the first car The speed of the second car The speed of the third car Brake pedal strength Depends on the actual distance between the first and second vehicles The actual distance between the second and third vehicles The speed of the first car The speed of the second car The speed of the third car ; After completing the route planning, it is necessary to control the platooning distance of commercial vehicles. When controlling the distance between vehicles, it is necessary to calculate the comprehensive coefficient of detection distance. Then, based on the comprehensive coefficient of detection distance Calculate the accelerator pedal pressure required to maintain a safe following distance. and brake pedal strength The platoon spacing maintaining unit can calculate the comprehensive detection distance coefficient according to the following formula. : , in, , These are weighting coefficients. This is the sensor sensitivity coefficient. The coordinate comprehensive judgment coefficient, The number of cameras installed in front of the vehicle. The number of cameras installed at the rear of the vehicle. The height of the camera mounted at the front of the vehicle relative to the ground. The height of the rear-mounted camera above the ground. A camera is mounted in front of the vehicle for a top-down view. A camera is mounted at the rear of the vehicle for a top-down view. A camera with an upward angle is installed in front of the vehicle. A rear-view camera is mounted on the vehicle, and the coordinates are used to make a comprehensive judgment coefficient. The expression is as follows: ， in,( , , () represents the center coordinates of the first vehicle in the vehicle coordinate system. , , ) represents the center coordinates of the second vehicle in the vehicle coordinate system. , , () represents the center coordinates of the third vehicle in the vehicle coordinate system.

2. The commercial vehicle platooning and following method based on camera perception decision-making and path planning according to claim 1, characterized in that: The camera perception evaluation index unit can calculate the spatial distance evaluation factor according to the following formula. : ， in, It is a natural constant. The speed of the first car The speed of the second car, The speed of the third car This is a spatial coordinate transformation factor, the value of which depends on the center coordinates of the first vehicle in the geodetic coordinate system. , , ), the center coordinates of the second vehicle in the geodetic coordinate system ( , , ), the center coordinates of the third vehicle in the geodetic coordinate system ( , , ), spatial coordinate transformation factor The expression is as follows; ， in, These are the weighting coefficients.

3. The commercial vehicle platooning method based on camera perception decision-making and path planning according to claim 1, characterized in that: The camera perception evaluation index unit can calculate the motion and attitude evaluation factor of traffic participants according to the following formula. : ， in, , , These are weighting coefficients. The speed of the first car The speed of the second car, The speed of the third car For the acceleration of the first vehicle, For the acceleration of the second car, For the acceleration of the third vehicle, The road has a longitudinal slope. For the cross slope of the road, The inherent brightness contrast of traffic participants.

4. The commercial vehicle platooning method based on camera perception decision-making and path planning according to claim 1, characterized in that: The camera perception evaluation index unit can calculate the traffic sign evaluation factor according to the following formula. : ， in, It is a natural constant. , , These are weighting coefficients. Color conversion factor This is the comprehensive coefficient for evaluating road surface noise. Road condition evaluation coefficient, color conversion factor Depending on the three-channel color parameters, For the first channel color parameters, For the second channel color parameters, For the third channel color parameters, The expression is as follows: ， in, It is a natural constant.

5. The commercial vehicle platooning method based on camera perception decision-making and path planning according to claim 1, characterized in that: The camera perception evaluation index unit can calculate the camera perception evaluation index according to the following formula. : , in, Spatial distance evaluation factor, For the evaluation factors of motion and posture of traffic participants, This is a traffic sign evaluation factor.

6. The commercial vehicle platooning method based on camera perception decision-making and path planning according to claim 1, characterized in that: The lane-changing decision unit includes three types of lane-changing instructions: following the car, waiting to change lanes, and executing lane change; Based on the principle of safety first, the following instruction is given priority, the waiting for the lane change instruction is given second priority, and executing the lane change instruction is given the lowest priority. A threshold for the willingness to follow the car is designed accordingly. Waiting for lane-changing willingness threshold , threshold for lane-changing intention To describe the intensity of the willingness to change lanes, among which .

7. The commercial vehicle platooning method based on camera perception decision-making and path planning according to claim 1, characterized in that: After receiving the lane-changing intention instruction from the lane-changing decision unit, the lane-changing path planning unit plans the path cluster for the next time period according to the path planning algorithm, and then uses the cost function. Calculate the cost estimate for the planned path cluster. Select, under the premise of ensuring safe driving of the vehicle The path with the shortest minimum is used as the optimal planned path, and the cost estimate of the path family is... The expression is as follows: ， in, This represents the weighted value of the cost estimate. Let d be the coordinates of station d. The cost function value for site d. The expression for the cost function value of site d is as follows: ， in, Let d be the cost function value from the starting point to station d. The cost function value from station d to the destination.

8. The commercial vehicle platooning method based on camera perception decision-making and path planning according to claim 1, characterized in that: After completing the route planning, it is necessary to control the platooning distance of commercial vehicles. The vehicle mechanism implements this by controlling the intensity of the accelerator pedal. and brake pedal strength The platoon spacing maintaining unit can calculate the accelerator pedal intensity according to the following formula. : , It is a natural constant. This represents the actual distance between the first and second vehicles. This represents the actual distance between the second and third vehicles. The speed of the first car The speed of the second car, The speed of the third car This is the comprehensive coefficient for detection distance.

9. The commercial vehicle platooning method based on camera perception decision-making and path planning according to claim 1, characterized in that: After completing the route planning, it is necessary to control the platooning distance of commercial vehicles. The vehicle mechanism implements this by controlling the intensity of the accelerator pedal. and brake pedal strength The platoon spacing maintaining unit can calculate the brake pedal strength according to the following formula. : ， It is a natural constant. This represents the actual distance between the first and second vehicles. This represents the actual distance between the second and third vehicles. The speed of the first car The speed of the second car, The speed of the third car This is the comprehensive coefficient for detection distance.

Images