Driving path boundary generation method and device, vehicle and storage medium

Abstract

The invention relates to a driving path boundary generation method and device, a vehicle and a storage medium, and relates to the technical field of intelligent driving. The method comprises the following steps: acquiring an initial vehicle pose and a target final position of a target vehicle; based on the initial vehicle pose and the target end point position, an initial driving track is generated through three-time optimal curve planning; based on the initial driving track, determining the maximum transverse position of the target vehicle deviating from the reference line; and based on the actual driving boundary condition and the maximum transverse position of the target vehicle, modifying the boundary of the initial driving track to generate a target driving path boundary. According to the invention, on the basis of ensuring the driving capability, a wider boundary range is provided for the intelligent driver to carry out subsequent path planning, and a more flexible and more comfortable path plan is generated.

Description

Technical Field

[0001] The present disclosure relates to the field of intelligent driving technology, and in particular, to a method, apparatus, vehicle, and storage medium for generating a driving path boundary. Background Art

[0002] Path planning refers to the problem of finding a reasonable path from a starting point to an ending point on a map, and has wide applications in fields such as computer science, artificial intelligence, and robotics, such as autonomous driving, UAV path planning, and robot path tracking.

[0003] There are various path planning methods, and common ones include hard constraints such as osqp and Ipopt. In related technologies, when generating the driving path boundary for intelligent driving of a vehicle, it can be generated by adding a lateral safety distance on the basis of the obtained lane boundary. However, the driving path boundary generated in this way can only ensure drivability, and in the case where the vehicle needs to encroach on the adjacent lane or slightly detour for more reasonable passage, it will cause the intelligent driving to be unable to plan a reasonable path, which is not flexible enough and also reduces the probability of successful path planning for a more comfortable driving.

[0004] It should be noted that the information disclosed in the above background art section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The present disclosure provides a method, apparatus, vehicle, and storage medium for generating a driving path boundary.

[0006] According to a first aspect of an embodiment of the present disclosure, a method for generating a driving path boundary is provided, including:

[0007] Obtaining an initial vehicle pose and a target end position of a target vehicle;

[0008] Based on the initial vehicle pose and the target end position, generating an initial driving trajectory by using a cubic optimal curve planning;

[0009] Based on the initial driving trajectory, determining a maximum lateral position of the target vehicle deviating from a reference line;

[0010] Based on the actual driving boundary condition of the target vehicle and the maximum lateral position, modifying the boundary of the initial driving trajectory to generate a target driving path boundary.

[0011] In some embodiments of the present disclosure, determining the maximum lateral position of the target vehicle deviating from the reference line based on the initial driving trajectory includes:

[0012] Based on the initial driving trajectory, an array of trajectory points of the target vehicle is calculated; wherein, the array of trajectory points includes a plurality of trajectory point data groups, and each trajectory point data group includes the position of the front axle center and the position of the rear axle center of the target vehicle in this trajectory point.

[0013] Based on the plurality of front axle center positions and the plurality of rear axle center positions in the plurality of trajectory point data groups, the maximum lateral position of the target vehicle deviating from the reference line is determined.

[0014] In some embodiments of the present disclosure, calculating an array of trajectory points of the target vehicle based on the initial driving trajectory includes:

[0015] Sampling the initial driving trajectory to obtain a plurality of trajectory points;

[0016] Taking each of the trajectory points as the plurality of rear axle center positions of the target vehicle, and calculating the front axle center positions corresponding to the plurality of rear axle center positions based on the plurality of rear axle center positions;

[0017] Based on the front axle center positions corresponding to the plurality of rear axle centers and the plurality of rear axle center positions, the array of trajectory points is obtained.

[0018] In some embodiments of the present disclosure, determining the maximum lateral position of the target vehicle deviating from the reference line based on the plurality of front axle center positions and the plurality of rear axle center positions in the plurality of trajectory point data groups includes:

[0019] Obtaining the reference line position corresponding to each of the trajectory points; wherein, the reference line includes the center line of the lane;

[0020] Based on the reference line position corresponding to each of the trajectory points, the plurality of front axle center positions and the plurality of rear axle center positions, calculate a plurality of first lateral distances by which each of the front axle center positions deviates from the reference line and a plurality of second lateral distances by which each of the rear axle center positions deviates from the reference line;

[0021] Select the maximum lateral distance with the largest value among the plurality of first lateral distances and the plurality of second lateral distances, and use the position point corresponding to the maximum lateral distance as the maximum lateral position.

[0022] In some embodiments of the present disclosure, modifying the boundary of the initial driving trajectory based on the actual driving boundary condition of the target vehicle and the maximum lateral position to generate a target driving path boundary includes:

[0023] In response to the actual driving boundary condition of the target vehicle being a crossable boundary, generate the boundary of the initial driving trajectory based on the vehicle width of the target vehicle and the initial driving trajectory;

[0024] Modify the boundary of the initial driving trajectory by using the maximum lateral position to generate a first driving path boundary;

[0025] Based on the driving obstacle information, correct the first driving path boundary without collision risk to obtain the target driving path boundary.

[0026] In some embodiments of the present disclosure, modifying the boundary of the initial driving trajectory by using the maximum lateral position to generate a first driving path boundary includes:

[0027] Calculate multiple offsets of the boundary of the initial driving trajectory corresponding to each trajectory point from the reference line;

[0028] Compare each offset with the maximum lateral position;

[0029] In response to the maximum lateral position being greater than the first offset, expand the boundary corresponding to the first offset by using the maximum lateral position to generate the first driving path boundary;

[0030] Wherein, the multiple offsets include the first offset.

[0031] In some embodiments of the present disclosure, the provided driving path boundary generation method further includes:

[0032] Calculate the positions of the self-vehicle obstacle boxes corresponding to the multiple rear axle centers based on the multiple rear axle center positions;

[0033] Generate the driving obstacle information based on the self-vehicle obstacle box positions.

[0034] In some embodiments of the present disclosure, the provided driving path boundary generation method further includes:

[0035] Based on the actual driving boundary condition of the target vehicle, determine whether all the actual driving boundaries corresponding to the initial driving trajectory are non-collision type boundaries;

[0036] If not, the actual driving boundary condition of the target vehicle is a crossable boundary.

[0037] In some embodiments of the present disclosure, the provided driving path boundary generation method further includes:

[0038] In response to there being some non-collision type boundaries in the actual driving boundaries corresponding to the initial driving trajectory, generate non-collision boundary information;

[0039] Generate the driving obstacle information based on the non-collision boundary information.

[0040] In some embodiments of the present disclosure, based on the initial vehicle pose and the target end position, an initial driving trajectory is generated using a cubic optimal curve planning, including:

[0041] Setting constraint conditions; the constraint conditions include: the lateral acceleration and lateral jerk of the target vehicle at the target end are zero;

[0042] Using the cubic optimal curve planning to generate the initial driving trajectory that satisfies the initial vehicle pose, the constraint conditions, and the target end position.

[0043] In some embodiments of the present disclosure, it further includes:

[0044] Obtaining obstacle information around the target vehicle;

[0045] Based on the obstacle information, using spatio-temporal joint decision-making to calculate the avoidance distance by which the target vehicle deviates from the reference line after a preset time duration;

[0046] Combining the reference line and the avoidance distance to generate the target end position.

[0047] In some embodiments of the present disclosure, it further includes:

[0048] Judging the driving scenario of the target vehicle based on the initial vehicle pose and the reference line;

[0049] Among them, obtaining the initial vehicle pose and the target end position of the target vehicle includes:

[0050] In response to the driving scenario of the target vehicle being an expandable boundary scenario, obtaining the initial vehicle pose and the target end position of the target vehicle.

[0051] In some embodiments of the present disclosure, judging the driving scenario of the target vehicle based on the initial vehicle pose and the reference line includes:

[0052] Based on the initial vehicle pose and the reference line, calculating the angle between the target vehicle and the reference line. If the angle is greater than a preset angle threshold, the driving scenario of the target vehicle is a non-expandable boundary scenario;

[0053] Based on the initial vehicle pose, determining that the target vehicle is in a preset turning driving situation, then the driving scenario of the target vehicle is a non-expandable boundary scenario;

[0054] Based on the reference line, determining the curvature of the reference line within a preset range in front of the target vehicle. If the curvature is greater than a preset curvature threshold, the driving scenario of the target vehicle is a non-expandable boundary scenario;

[0055] Otherwise, the driving scenario of the target vehicle is an expandable boundary scenario.

[0056] According to a second aspect of the embodiments of the present disclosure, there is provided a driving path boundary generation device, including:

[0057] A data acquisition unit, configured to acquire the initial vehicle pose and the target end position of the target vehicle;

[0058] An initial driving trajectory generation unit, configured to generate an initial driving trajectory by using a cubic optimal curve planning based on the initial vehicle pose and the target end position;

[0059] A maximum lateral position determination unit, configured to determine the maximum lateral position of the target vehicle deviating from the reference line based on the initial driving trajectory;

[0060] A driving path boundary generation unit, configured to modify the boundary of the initial driving trajectory based on the actual driving boundary condition of the target vehicle and the maximum lateral position, and generate a target driving path boundary.

[0061] According to a third aspect of the embodiments of the present disclosure, there is provided a vehicle, including:

[0062] A processor;

[0063] A memory for storing instructions executable by the processor;

[0064] Wherein, the processor is configured to: implement the driving path boundary generation method described in the first aspect above.

[0065] According to a fourth aspect of the embodiments of the present disclosure, there is provided a non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by a processor of a mobile terminal, enabling the mobile terminal to execute the driving path boundary generation method described in the first aspect above.

[0066] The technical solutions provided by the embodiments of the present disclosure may include the following beneficial effects:

[0067] The present disclosure obtains the initial vehicle pose and the target end position of a target vehicle; generates an initial driving trajectory using a cubic optimal curve planning based on the initial vehicle pose and the target end position; determines the maximum lateral position at which the target vehicle deviates from a reference line based on the initial driving trajectory; and modifies the boundary of the initial driving trajectory based on the actual driving boundary condition of the target vehicle and the maximum lateral position to generate a target driving path boundary. Based on the maximum lateral position at which the target vehicle deviates from the reference line and the actual driving boundary condition on the basis of the optimal driving trajectory generated by the cubic optimal curve planning, the boundary of the initial driving trajectory is extended as much as possible to obtain the target driving path boundary, so as to ensure that the target driving path boundary is drivable and can also provide a wider boundary range for subsequent path planning by intelligent driving, thereby generating a more flexible and more comfortable path planning.

[0068] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The accompanying drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure. Obviously, the accompanying drawings in the following description are only some embodiments of the present disclosure, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0070] Figure 1 is a flowchart of a method for generating a driving path boundary according to some embodiments of the present disclosure Figure 1 .

[0071] Figure 2 is a flowchart of a method for generating a driving path boundary according to some embodiments of the present disclosure Figure 2 .

[0072] Figure 3 is a flowchart of a method for generating a driving path boundary according to some embodiments of the present disclosure Figure 3 .

[0073] Figure 4 is a flowchart of an implementation process of step S104 according to some embodiments of the present disclosure.

[0074] Figure 5 is a flowchart of an implementation process of step S106 according to some embodiments of the present disclosure.

[0075] Figure 6 is a flowchart of an implementation process of step S502 according to some embodiments of the present disclosure.

[0076] Figure 7 It is a flowchart of an implementation process of step S504 shown according to some embodiments of the present disclosure.

[0077] Figure 8 It is a flowchart of an implementation process of step S108 shown according to some embodiments of the present disclosure Figure 1 .

[0078] Figure 9 It is a flowchart of an implementation process of step S804 shown according to some embodiments of the present disclosure.

[0079] Figure 10 It is a flowchart of an implementation process of step S108 shown according to some embodiments of the present disclosure Figure 2 .

[0080] Figure 11 It is a flowchart of an implementation process of step S108 shown according to some embodiments of the present disclosure Figure 3 .

[0081] Figure 12 It is a flowchart of an implementation process of step S108 shown according to some embodiments of the present disclosure Figure 4 .

[0082] Figure 13 It is a block diagram of a driving path boundary generation device shown according to some embodiments of the present disclosure.

[0083] Figure 14 It is a block diagram of a vehicle shown according to an exemplary embodiment of the present disclosure. Detailed implementation manners

[0084] Some embodiments of the present disclosure will be described in detail herein, and examples thereof are shown in the accompanying drawings. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. Various changes, modifications, and equivalents of the methods, apparatuses, and / or systems described herein will become apparent after understanding the present disclosure. For example, the order of operations described herein is merely an example and is not limited to those set forth herein, but may be changed as will be apparent after understanding the present disclosure, except for operations that must be performed in a specific order. Additionally, descriptions of features known in the art may be omitted for increased clarity and conciseness.

[0085] The implementation manners described in some embodiments of the present disclosure below do not represent all implementation manners consistent with the present disclosure. Instead, they are merely examples of apparatuses and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0086] The following will describe in detail the specific implementation manners of the embodiments of the present disclosure with reference to the accompanying drawings.

[0087] Figure 1 is a flowchart of a driving path boundary generation method shown according to some embodiments of the present disclosure Figure 1 , as Figure 1 shown, the driving path boundary generation method can be applied to a vehicle.

[0088] Figure 1 The driving path boundary generation method shown as above includes the following steps.

[0089] In step S102, obtain the initial vehicle pose and the target end position of the target vehicle.

[0090] It should be noted that the initial vehicle pose of the target vehicle refers to the vehicle pose information at the initial moment. The vehicle pose information refers to the data describing the position and attitude of the vehicle in space, which may include position data and attitude data, such as yaw angle (Yaw), pitch angle (Pitch), roll angle (Roll), etc. It can be obtained by means of a combination of global positioning system (GPS) and inertial measurement unit (IMU), lidar, vehicle-mounted sensors, etc.

[0091] In some exemplary embodiments of the present disclosure, the target end position refers to the ideal position where it is desired for the target vehicle to be after a preset duration during path planning. For example, in a path planning scenario of slightly crossing the line, the target end position can be the position 2 seconds after the initial moment, and its position is inside the lane line of the vehicle body and in a non-crossing state.

[0092] In step S104, based on the initial vehicle pose and the target end position, use cubic optimal curve planning to generate an initial driving trajectory.

[0093] It should be noted that the cubic optimal curve refers to a smooth curve constructed using a cubic polynomial or cubic spline, which is optimal under certain conditions. In autonomous driving, cubic polynomials are often used to plan trajectories, which can ensure the continuity of position, speed, and acceleration. For example, given the position, speed, and acceleration of the starting point and end point of the target vehicle, the coefficients of a cubic polynomial can be obtained by solving an appropriate system of equations, thereby defining a smooth trajectory.

[0094] In step S106, based on the initial driving trajectory, determine the maximum lateral position of the target vehicle deviating from the reference line. [[ID=3,2]]

[0095] It should be noted that the reference line is a reference quantity during the vehicle driving process, which is an idealized trajectory or center line guiding the vehicle to drive along a predetermined path. This line usually defines the path that the vehicle should follow and serves as the basis for generating the actual driving trajectory. It can be determined based on map data, sensor fusion, etc. For example, in an urban road scenario, the reference line can be the center line of the lane. In a rural path scenario, the reference line can be the center line of the actual road. It can be defined according to the actual driving scenario, and the embodiments of the present disclosure do not make limitations here.

[0096] In step S108, based on the actual driving boundary conditions and the maximum lateral position of the target vehicle, the boundary of the initial driving trajectory is modified to generate the target driving path boundary.

[0097] In some embodiments of the present disclosure, the maximum lateral position represents the maximum lateral distance that can ensure the target vehicle can deviate from the reference line during the process of reaching the target end position. The boundary of the initial driving trajectory can be adaptively expanded through the maximum lateral distance, and the expansion of the boundary is corrected by combining the actual driving boundary conditions of the target vehicle, so as to obtain the target driving path boundary that can ensure safe driving and has an expanded range.

[0098] It can be seen from the above steps that the driving path boundary generation method provided by the embodiments of the present disclosure includes: obtaining the initial vehicle pose and the target end position of the target vehicle; based on the initial vehicle pose and the target end position, using a cubic optimal curve to plan to generate an initial driving trajectory; based on the initial driving trajectory, determining the maximum lateral position of the target vehicle deviating from the reference line; based on the actual driving boundary conditions and the maximum lateral position of the target vehicle, modifying the boundary of the initial driving trajectory to generate the target driving path boundary. Based on the optimal driving trajectory generated by the cubic optimal curve planning, and based on the maximum lateral position of the target vehicle deviating from the reference line and the actual driving boundary conditions, the boundary of the initial driving trajectory is expanded as much as possible to obtain the target driving path boundary, so as to ensure that on the basis of the drivability of the target driving path boundary, a wider boundary range can be provided for the intelligent driving to perform subsequent path planning, and a more flexible and more comfortable path planning can be generated.

[0099] Figure 2 is a flowchart of a driving path boundary generation method shown according to an exemplary embodiment of the present disclosure Figure 2 。

[0100] In the embodiments of the present disclosure, Figure 2 in the driving path boundary generation method shown, steps S206 to S210 correspond to Figure 1 steps S104 to S108 in the driving path boundary generation method shown, and will not be repeated here.

[0101] In the embodiments of the present disclosure, inFigure 1 Based on the driving path boundary generation method shown above, Figure 2 the driving path boundary generation method shown above may further include the following steps.

[0102] In step S202, based on the initial vehicle pose and the reference line, determine the driving scenario of the target vehicle.

[0103] In step S204, in response to the driving scenario of the target vehicle being an expandable boundary scenario, obtain the initial vehicle pose of the target vehicle and the target end position.

[0104] It should be noted that when the vehicle is driving, there are certain scenarios where boundary expansion is not required or cannot be performed. In such cases, it can be judged through step 202. If it is judged that the driving scenario of the target vehicle is an expandable boundary scenario, the subsequent steps are executed. If the driving scenario of the target vehicle is a non-expandable boundary scenario, the execution logic is directly ended, reducing the processing flow of the vehicle controller and saving the operation consumption of the controller.

[0105] In some embodiments of the present disclosure, the specific implementation process of step S202 may include:

[0106] Based on the initial vehicle pose and the reference line, calculate the angle between the target vehicle and the reference line. If the angle is greater than a preset angle threshold, for example, it can be 30 degrees, it indicates that the deviation between the target vehicle and the lane is large at this time, and the driving scenario of the target vehicle is a non-expandable boundary scenario.

[0107] Based on the initial vehicle pose, determine that the target vehicle is in a preset turning driving situation, then the driving scenario of the target vehicle is a non-expandable boundary scenario. It should be noted that the preset turning driving situations may include turning scenarios such as turning left, turning right, and U-turn.

[0108] Based on the reference line, determine the curvature of the reference line within a preset range in front of the target vehicle. If the curvature is greater than a preset curvature threshold, the driving scenario of the target vehicle is a non-expandable boundary scenario. For example, if the curvature of the reference line within 2m in front of the target vehicle is greater than 0.05, the driving scenario of the target vehicle is a non-expandable boundary scenario.

[0109] If any of the above situations is satisfied, the driving scenario of the target vehicle is a non-expandable boundary scenario; otherwise, the driving scenario of the target vehicle is an expandable boundary scenario. [[ID=2{8]]

[0110] Figure 3 is a flowchart of a driving path boundary generation method shown according to an exemplary embodiment of the present disclosure <{ Figure 3 .

[0111] In the embodiments of the present disclosure, Figure 3In the driving path boundary generation method shown, steps S308 to S314 correspond to steps S102 to S108 in the driving path boundary generation method shown, and will not be repeated here. Figure 1 In the driving path boundary generation method shown, steps S102 to S108 correspond to steps S308 to S314 in the driving path boundary generation method shown, and will not be repeated here.

[0112] In the embodiments of the present disclosure, based on Figure 1 the driving path boundary generation method shown, Figure 3 the driving path boundary generation method shown may further include the following steps.

[0113] In step S302, obtain the obstacle information around the target vehicle.

[0114] In step S304, based on the obstacle information, use spatio-temporal joint decision-making to calculate the avoidance distance by which the target vehicle deviates from the reference line after a preset time duration.

[0115] In step S306, combine the reference line and the avoidance distance to generate the target end position.

[0116] It should be noted that when there are no obstacles around the target vehicle, it is expected that the physical position of the target vehicle returns to the reference line. When obstacles need to be avoided, the target end position may deviate from the reference line by a certain distance, which usually depends on the position of the obstacles to be avoided, and a safety distance threshold, such as 0.2 m, is added to calculate the final target end position.

[0117] In some embodiments of the present disclosure, the obstacles may include dynamic obstacles, which may include: other vehicles: including cars, trucks, motorcycles, etc. in front, behind, and on both sides, pedestrians: people walking on the road or on the roadside, cyclists: people riding bicycles or other two-wheeled vehicles, animals: wild animals or livestock that occasionally appear on the road, etc. And static obstacles, which may include: roadblocks: such as fences and cones in construction areas, buildings: building structures on both sides of the road, trees and vegetation: green belts or natural landscapes by the roadside, traffic facilities, etc. The obstacle information around the target vehicle can be determined by means such as sensor monitoring, obstacle detection and classification, and obstacle tracking. The implementation process is diverse and will not be elaborated in the embodiments of the present disclosure.

[0118] In some exemplary embodiments of the present disclosure, based on obstacle information, the avoidance distance by which the target vehicle deviates from the reference line after a preset time duration is calculated using spatio-temporal joint decision-making. It should be noted that the preset time duration is predefined and is generally the minimum time unit for driving path planning. For example, when driving path planning generally plans and predicts the driving path within a 3-second period, the target end position represents the position where the target vehicle is expected to appear after 3 seconds. The avoidance distance calculated based on spatio-temporal joint decision-making can be used to adjust the position of the vehicle on the road to ensure safe and efficient driving. This avoidance distance is obtained by considering information in two dimensions: time (preset time duration) and space (i.e., the specific position of the vehicle on the road). Based on this avoidance distance and the position of the reference line, the target end position can be determined.

[0119] In some exemplary embodiments of the present disclosure, as Figure 4 shown, it is a flowchart of the implementation process of step S104 provided by some exemplary embodiments of the present disclosure, including the following steps.

[0120] In step S402, constraint conditions are set.

[0121] It should be noted that the constraint conditions include: the lateral acceleration and lateral jerk of the target vehicle at the target end point are zero.

[0122] In step S404, using cubic optimal curve planning, an initial driving trajectory that satisfies the initial vehicle pose, constraint conditions, and target end position is generated.

[0123] In some embodiments of the present disclosure, based on the initial vehicle pose, the initial state of the vehicle can be determined, which may include: initial position, initial attitude, initial lateral velocity, initial lateral acceleration, and the lateral jerk is set to 0. Based on the target end position, the end state of the vehicle can be determined, which may include end position, end attitude, end lateral velocity of 0, end lateral acceleration of 0, and end lateral jerk of 0. The key position information for setting the generated curve is mainly the lateral offset distance relative to the reference line. Based on the initial state, constraint conditions, and end state, using the method of generating a cubic optimal curve, a curve that can satisfy the initial state, constraint conditions, and end state and reach the end position fastest is generated, that is, the initial driving trajectory.

[0124] In some embodiments of the present disclosure, after generating the initial driving trajectory, the maximum lateral position by which the target vehicle deviates from the reference line can be determined according to the initial driving trajectory. The specific implementation process is as Figure 5 shown, including the following steps.

[0125] In step 502, based on the initial driving trajectory, an array of trajectory points of the target vehicle is calculated.

[0126] It should be noted that the trajectory point array contains multiple trajectory point data groups, and each trajectory point data group contains the front axle center position and the rear axle center position of the target vehicle in that trajectory point. It can be understood that in the initially generated driving trajectory, the target vehicle is regarded as a particle, but the vehicle has its own shape, volume, and size, and the actual situation of the target vehicle needs to be considered. Therefore, it is necessary to determine the corresponding positions of the front axle center and the rear axle center of the target vehicle at each trajectory point. The front axle center refers to the center point of the front wheel axis of the vehicle, usually located at the midpoint of the contact surface of the two front wheels, that is, the midpoint of the connection line of the center points of the front wheel tires in contact with the ground. The front axle center is an important reference point for analyzing the longitudinal and lateral dynamics of the vehicle. The rear axle center refers to the center point of the rear wheel axis of the vehicle, usually located at the midpoint of the contact surface of the two rear wheels, that is, the midpoint of the connection line of the center points of the rear wheel tires in contact with the ground. In the autonomous driving system, the rear axle center is often used as a reference point for the vehicle kinematic model.

[0127] In some exemplary embodiments of the present disclosure, the implementation process of step S502 is as Figure 6 shown, including the following steps.

[0128] In step S602, sampling is performed on the initial driving trajectory to obtain multiple trajectory points.

[0129] In some embodiments of the present disclosure, sampling can be performed every 0.2 seconds on the initial driving trajectory. The obtained sampling points are also the positions where the target vehicle reaches after traveling for 0.2 seconds on this initial driving trajectory, and multiple trajectory points are obtained. It should be noted that 0.2 seconds is only an example, and the duration can be adjusted according to actual needs, and the present disclosure does not make any limitations here.

[0130] In step S604, each trajectory point is used as the multiple rear axle center positions of the target vehicle, and the front axle center positions corresponding to the multiple rear axle centers are calculated based on the multiple rear axle center positions.

[0131] In some embodiments of the present disclosure, for each trajectory point, when the trajectory point is used as the rear axle center position of the target vehicle, the position of the front axle center corresponding to the rear axle center can be calculated based on the rear axle center position, the current orientation of the target vehicle, and the wheelbase.

[0132] In step S606, a trajectory point array is obtained based on the front axle center positions corresponding to the multiple rear axle centers and the multiple rear axle center positions.

[0133] In some embodiments of the present disclosure, the front axle center position and the rear axle center position at each trajectory point are stored in the form of a pair of numbers, for example, (A1, B1), where 1 represents the first trajectory point, A1 is the numerical value of the front axle center position at the first trajectory point, and B1 is the numerical value of the rear axle center position at the first trajectory point. A plurality of pairs of numbers form an array of trajectory points. For example, {(A1, B1), (A2, B2),...}.

[0134] In step 504, based on the plurality of front axle center positions and the plurality of rear axle center positions in the plurality of trajectory point data groups, determine the maximum lateral position at which the target vehicle deviates from the reference line.

[0135] It should be noted that the reference line includes the center line of the lane.

[0136] In some exemplary embodiments of the present disclosure, the implementation process of step S504 is as Figure 7 shown and includes the following steps.

[0137] In step S702, obtain the reference line position corresponding to each trajectory point.

[0138] In step S704, based on the reference line position corresponding to each trajectory point, the plurality of front axle center positions, and the plurality of rear axle center positions, calculate a plurality of first lateral distances by which each front axle center position deviates from the reference line and a plurality of second lateral distances by which each rear axle center position deviates from the reference line.

[0139] In step S706, select the maximum lateral distance among the plurality of first lateral distances and the plurality of second lateral distances, and use the position point corresponding to the maximum lateral distance as the maximum lateral position.

[0140] It can be understood that the center line of the lane as the reference line varies in different sections of the lane, and it is necessary to obtain the reference line position corresponding to each trajectory point. For each trajectory point, based on the reference line position corresponding to the trajectory point and the front axle center position at the trajectory point, a first lateral distance can be calculated, and based on the reference line position corresponding to the trajectory point and the rear axle center position at the trajectory point, a second lateral distance can be calculated. Thus, a plurality of first lateral distances and a plurality of second lateral distances can be obtained, compare the numerical values, and select the largest numerical value as the maximum lateral distance.

[0141] It should be noted that the lateral direction mentioned in the embodiments of the present disclosure is the direction perpendicular to the reference line.

[0142] Correspondingly, in some exemplary embodiments of the present disclosure, as Figure 8 shown, the implementation process flow of step S108 provided by some exemplary embodiments of the present disclosure Figure 1 includes the following steps.

[0143] In step S802, in response to the actual driving boundary condition of the target vehicle being a crossable boundary, based on the vehicle width of the target vehicle and the initial driving trajectory, the boundary of the initial driving trajectory is generated.

[0144] It should be noted that when the actual driving boundary condition of the target vehicle is a crossable boundary, the boundary can be expanded. And the generation of the initial driving trajectory is based on regarding the target vehicle as a mass point, that is, the driving trajectory of the equivalent center point of the target vehicle. Therefore, it is necessary to combine the half vehicle width of the target vehicle and the initial driving trajectory to generate the boundary of the initial driving trajectory, so as to ensure that the vehicle remains within a safe range during driving and avoid collisions with other obstacles.

[0145] In some embodiments of the present disclosure, the normal vector at each trajectory point can be calculated. For each trajectory point, move a distance of half the vehicle width in the outer direction (away from the road center) and the inner direction (towards the road center) along the normal vector direction respectively to obtain two boundary points. Connect all the outer boundary points to form an outer boundary line, and connect all the inner boundary points to form an inner boundary line. These two lines together constitute the boundary of the initial driving trajectory. It can be understood that the above process of generating the boundary of the initial driving trajectory is only an example and is not used to limit the protection scope of the embodiments of the present disclosure.

[0146] In step S804, the boundary of the initial driving trajectory is modified using the maximum lateral position to generate the first driving path boundary.

[0147] In some exemplary embodiments of the present disclosure, the implementation process of step S804 is as Figure 9 shown and includes the following steps.

[0148] In step S902, calculate multiple offsets of the boundary of the initial driving trajectory corresponding to each trajectory point from the reference line.

[0149] It should be noted that calculate multiple offsets of the outer boundary of the initial driving trajectory corresponding to each trajectory point from the reference line.

[0150] In step S904, compare each offset with the maximum lateral position.

[0151] In step S906, in response to the maximum lateral position being greater than the first offset, expand the boundary corresponding to the first offset using the maximum lateral position to generate the first driving path boundary.

[0152] During specific implementation, if the maximum lateral position is less than or equal to the offset, the outer boundary of the initial driving trajectory corresponding to this trajectory point remains unchanged. If the maximum lateral position is greater than the offset, the offset is denoted as the first offset, and multiple offsets include the first offset. The outer boundary of the initial driving trajectory at this trajectory point is extended by using the maximum lateral position to replace the first offset.

[0153] In step S806, based on the driving obstacle information, the first driving path boundary is corrected for collision-free risk to obtain the target driving path boundary.

[0154] It should be noted that after obtaining the expanded first driving path boundary, collision-free risk correction is required to ensure that the driving path planned thereby has no collision risk and ensure safe driving.

[0155] Based on the driving obstacle information, the first driving path boundary can be corrected for collision-free risk to obtain the target driving path boundary without collision risk.

[0156] In some exemplary embodiments of the present disclosure, the driving obstacle information may include static obstacle information during driving. Correspondingly, as Figure 10 shown, it is the implementation process flow of step S108 provided by some exemplary embodiments of the present disclosure Figure 2 , Figure 10 shown, steps S1002 to S1004 and step S1010 in the implementation process are corresponding to steps S802 to S806 in the implementation process shown in Figure 8 shown, and will not be repeated here.

[0157] In the embodiments of the present disclosure, on the basis of the implementation process shown in Figure 8 shown, Figure 10 the implementation process shown may further include the following steps.

[0158] In step S1006, the self-vehicle obstacle box positions corresponding to multiple rear axle center positions are calculated based on the multiple rear axle center positions.

[0159] In step S1008, driving obstacle information is generated based on the self-vehicle obstacle box positions.

[0160] It should be noted that the self-vehicle obstacle box refers to a three-dimensional space region around an obstacle in the vehicle coordinate system, which is usually used to represent the size and position of the obstacle for tasks such as collision detection and path planning. Generally, it is a static obstacle during the driving of the target vehicle, such as stones, warning triangles, tree branches and other non-collidable objects.

[0161] In some exemplary embodiments of the present disclosure, the driving obstacle information may include non-collision boundary information during driving. The non-collision boundary information refers to the boundary area or range that defines the area that the target vehicle must avoid entering during its driving to ensure safe driving. For example, it may be a lane boundary such as a curb, a fence, or a wire mesh that the target vehicle cannot collide with.

[0162] Correspondingly, as Figure 11 shown, it is the implementation process flow of step S108 provided by some exemplary embodiments of the present disclosure Figure 3 , Figure 11 In the shown implementation process, steps S1104 to S1108 correspond to steps S802 to S806 in the Figure 8 shown implementation process, and will not be repeated here.

[0163] In the embodiments of the present disclosure, based on the Figure 8 shown implementation process, Figure 11 the shown implementation process may further include the following steps.

[0164] In step S1102, based on the actual driving boundary situation of the target vehicle, it is determined whether all the actual driving boundaries corresponding to the initial driving trajectory are non-collision type boundaries; if not, the actual driving boundary situation of the target vehicle is a crossable boundary.

[0165] It should be noted that the actual driving boundary situation of the target vehicle within the preset time period, that is, the actual driving boundary situation between the initial position and the target end position, is obtained to determine whether all the actual driving boundaries corresponding to the initial driving trajectory are non-collision type boundaries.

[0166] For better illustration, by way of example, if the actual driving boundary situation of the target vehicle between the initial position and the target end position is all lane lines, then they are all collidable type boundaries. If the actual driving boundary situation of the target vehicle between the initial position and the target end position is part lane lines and part curb, it can still be determined as a crossable boundary. If the actual driving boundary situation of the target vehicle between the initial position and the target end position is all curbs, then they are all non-collision type boundaries, and at this time the actual driving boundary situation of the target vehicle is a non-crossable boundary.

[0167] In response to the actual driving boundary situation of the target vehicle being a crossable boundary, the boundary of the initial driving trajectory can be expanded to obtain the target driving path boundary. Otherwise, the boundary of the obtained initial driving trajectory can be directly output, as there will be a collision risk and no expansion is required.

[0168] It can be understood that when there are some non-collision type boundaries in the actual driving boundary corresponding to the initial driving trajectory, there will still be a constraint on the first driving path boundary, and driving obstacle information can be generated based on the non-collision boundary information for collision-free risk correction.

[0169] As Figure 12 shown, it is the implementation process flow of step S108 provided by some exemplary embodiments of the present disclosure Figure 4 , [[ID= In the implementation process shown, steps S1202 to S1206, step S1212 correspond to steps S1102 to S1108 in the implementation process shown, and will not be repeated here. ​

[0170] In the embodiments of the present disclosure, on the basis of the implementation process shown in ​ ​ shown, the implementation process shown can further include the following steps.

[0171]

[0172] In step S1208, in response to the existence of some non-collision type boundaries in the actual driving boundary corresponding to the initial driving trajectory, non-collision boundary information is generated.

[0173] In step S1210, driving obstacle information is generated based on the non-collision boundary information.

[0174] In some embodiments of the present disclosure, generating non-collision boundary information based on the existing some non-collision type boundaries may include information such as the position of the non-collision boundary. And driving obstacle information is generated based on the non-collision boundary information to perform collision-free risk correction on the first driving path boundary to obtain a target driving path boundary, ensuring the safe driving of the target vehicle.

[0175] In some exemplary embodiments of the present disclosure, the driving obstacle information may include non-collision boundary information during driving and static obstacle information during driving. The implementation process is similar to the above embodiments, and the embodiments of the present disclosure will not be elaborated here. Combining all non-collision information when the target vehicle is driving in the first driving path boundary to perform collision-free risk correction on the first driving path boundary, maximizing the driving safety of the generated target driving path boundary and ensuring that the expanded boundary is safe and reliable.

[0175] It should be noted that after generating the target driving path boundary, the trajectory of the target vehicle for intelligent driving can also be generated according to the generated target driving path boundary, so as to improve the success rate of intelligent driving planning. By expanding the boundary, the space of the drivable area during the trajectory generation process is maximally increased to ensure that the constraint limits of the planning algorithm are looser, thereby increasing the reliability of the algorithm. This is to avoid the situation where the lane boundary of the vehicle is too narrow, but the actual vehicle can pass in a way of slightly pressing the line, or the path planning algorithm may still fail when there is passable space around, so as to avoid affecting the overall planning quality.

[0176] It can be understood that the driving path boundary generation method provided in the above embodiments of the present disclosure is only for generating the driving path boundary during a period of time in intelligent driving decision-making, generally the minimum time unit of driving path planning. The actual intelligent driving decision can repeatedly execute the driving path boundary generation method provided in the above embodiments. For example, after generating the target driving path boundary of the target vehicle within 3 seconds at the first position point, when the target vehicle actually reaches the second position point, the above driving path boundary generation method is re-executed to generate the target driving path boundary within 3 seconds at the second position point, and so on.

[0177] The following is an embodiment of the device of the present disclosure, which can be used to execute the method embodiment of the present disclosure. For the details not disclosed in the embodiment of the device of the present disclosure, please refer to the method embodiment of the present disclosure.

[0178] ​ is a block diagram of a driving path boundary generation device shown according to some embodiments of the present disclosure. Referring to ​ , the device includes: a data acquisition unit 1301, an initial driving trajectory generation unit 1302, a maximum lateral position determination unit 1303, and a driving path boundary generation unit 1304.

[0179] The data acquisition unit 1301 is configured to acquire the initial vehicle pose and the target end position of the target vehicle;

[0180] The initial driving trajectory generation unit 1302 is configured to generate an initial driving trajectory based on the initial vehicle pose and the target end position by using a cubic optimal curve planning;

[0181] The maximum lateral position determination unit 1303 is configured to determine the maximum lateral position of the target vehicle deviating from the reference line based on the initial driving trajectory;

[0182] The driving path boundary generation unit 1304 is configured to modify the boundary of the initial driving trajectory based on the actual driving boundary condition of the target vehicle and the maximum lateral position to generate a target driving path boundary.

[0183] In some exemplary embodiments of the present disclosure, it further includes: a target end position generation unit, configured to:

[0184] Obtain obstacle information around the target vehicle;

[0185] Based on the obstacle information, use spatio-temporal joint decision-making to calculate the avoidance distance by which the target vehicle deviates from the reference line after a preset duration;

[0186] Combine the reference line and the avoidance distance to generate the target end position.

[0187] In some exemplary embodiments of the present disclosure, the initial driving trajectory generation unit 1302 is configured to: set constraint conditions; the constraint conditions include: the lateral acceleration and lateral jerk of the target vehicle at the target end are zero; use cubic optimal curve planning to generate an initial driving trajectory that satisfies the initial vehicle pose, constraint conditions, and target end position.

[0188] In some exemplary embodiments of the present disclosure, the maximum lateral position determination unit 1303 is configured to:

[0189] Based on the initial driving trajectory, calculate an array of trajectory points of the target vehicle; wherein, the array of trajectory points contains multiple trajectory point data groups, and each trajectory point data group contains the front axle center position and the rear axle center position of the target vehicle in that trajectory point;

[0190] Based on the multiple front axle center positions and multiple rear axle center positions in the multiple trajectory point data groups, determine the maximum lateral position by which the target vehicle deviates from the reference line.

[0191] In some exemplary embodiments of the present disclosure, the maximum lateral position determination unit 1303 is configured to:

[0192] Sample the initial driving trajectory to obtain multiple trajectory points;

[0193] Take each trajectory point as the multiple rear axle center positions of the target vehicle, and calculate the front axle center positions corresponding to the multiple rear axle center positions based on the multiple rear axle center positions;

[0194] Based on the front axle center positions corresponding to the multiple rear axle center positions and the multiple rear axle center positions, obtain the array of trajectory points.

[0195] In some exemplary embodiments of the present disclosure, the maximum lateral position determination unit 1303 is configured to:

[0196] Obtain the reference line position corresponding to each trajectory point; wherein, the reference line includes the lane center line;

[0197] Based on the reference line positions corresponding to each trajectory point, multiple front axle center positions, and multiple rear axle center positions, calculate multiple first lateral distances by which each front axle center position deviates from the reference line and multiple second lateral distances by which each rear axle center position deviates from the reference line;

[0198] Select the maximum lateral distance among the multiple first lateral distances and the multiple second lateral distances, and use the position point corresponding to the maximum lateral distance as the maximum lateral position.

[0199] In some exemplary embodiments of the present disclosure, the driving path boundary generation unit 1304 is configured to:

[0200] In response to the actual driving boundary condition of the target vehicle being a crossable boundary, generate a boundary of the initial driving trajectory based on the vehicle width of the target vehicle and the initial driving trajectory;

[0201] Modify the boundary of the initial driving trajectory using the maximum lateral position to generate a first driving path boundary;

[0202] Based on the driving obstacle information, perform a collision - free risk correction on the first driving path boundary to obtain a target driving path boundary.

[0203] In some exemplary embodiments of the present disclosure, the driving path boundary generation unit 1304 is configured to:

[0204] Calculate multiple offsets by which the boundary of the initial driving trajectory corresponding to each trajectory point deviates from the reference line;

[0205] Compare each offset with the maximum lateral position;

[0206] In response to the maximum lateral position being greater than the first offset, use the maximum lateral position to expand the boundary corresponding to the first offset to generate a first driving path boundary;

[0207] Wherein, the multiple offsets include a first offset.

[0208] In some exemplary embodiments of the present disclosure, the driving path boundary generation unit 1304 is configured to:

[0209] Based on multiple rear axle center positions, calculate self - vehicle obstacle box positions corresponding to multiple rear axles;

[0210] Generate driving obstacle information based on the self - vehicle obstacle box positions.

[0211] In some exemplary embodiments of the present disclosure, the driving path boundary generation unit 1304 is configured to:

[0212] Based on the actual driving boundary situation of the target vehicle, determine whether all of the actual driving boundaries corresponding to the initial driving trajectory are non-collision type boundaries;

[0213] If not, the actual driving boundary situation of the target vehicle is a crossable boundary.

[0214] In some exemplary embodiments of the present disclosure, the driving path boundary generation unit 1304 is configured to:

[0215] In response to the existence of some non-collision type boundaries in the actual driving boundaries corresponding to the initial driving trajectory, generate non-collision boundary information;

[0216] Generate driving obstacle information based on the non-collision boundary information.

[0217] In some exemplary embodiments of the present disclosure, it further includes: a driving scenario determination unit, configured to: judge the driving scenario of the target vehicle based on the initial vehicle pose and the reference line. Correspondingly, the data acquisition unit 1301 is configured to: in response to the driving scenario of the target vehicle being an expandable boundary scenario, acquire the initial vehicle pose of the target vehicle and the target end position.

[0218] In some exemplary embodiments of the present disclosure, the driving scenario determination unit is configured to:

[0219] Based on the initial vehicle pose and the reference line, calculate the angle between the target vehicle and the reference line. If the angle is greater than a preset angle threshold, the driving scenario of the target vehicle is a non-expandable boundary scenario;

[0220] Based on the initial vehicle pose, determine that the target vehicle is in a preset turning driving situation, then the driving scenario of the target vehicle is a non-expandable boundary scenario;

[0221] Based on the reference line, determine the curvature of the reference line within a preset range in front of the target vehicle. If the curvature is greater than a preset curvature threshold, the driving scenario of the target vehicle is a non-expandable boundary scenario;

[0222] Otherwise, the driving scenario of the target vehicle is an expandable boundary scenario.

[0223] Regarding the device in the above embodiments, the specific manners in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0224] ​ It is a block diagram of a vehicle 1400 shown according to an exemplary embodiment. For example, the vehicle 1400 can be a hybrid vehicle, or a non-hybrid vehicle, an electric vehicle, a fuel cell vehicle, or other types of vehicles. The vehicle 1400 can be an autonomous vehicle, a semi-autonomous vehicle, or a non-autonomous vehicle.

[0225] Referring to ​ , vehicle 1400 may include various subsystems. For example, an infotainment system 1410, a perception system 1420, a decision control system 1430, a drive system 1440, and a computing platform 1450. Among them, vehicle 1400 may also include more or fewer subsystems, and each subsystem may include multiple components. In addition, each subsystem and each component of vehicle 1400 may be interconnected by wired or wireless means.

[0226] In some embodiments, the infotainment system 1410 may include a communication system, an entertainment system, and a navigation system, etc.

[0227] The perception system 1420 may include several sensors for sensing information about the environment around vehicle 1400. For example, the perception system 1420 may include a global positioning system (the global positioning system may be a GPS system, a Beidou system, or other positioning systems), an inertial measurement unit (IMU), lidar, millimeter-wave radar, ultrasonic radar, and a camera device.

[0228] The decision control system 1430 may include a computing system, a vehicle controller, a steering system, an accelerator, and a braking system.

[0229] The drive system 1440 may include components that provide motive power for vehicle 1400. In one embodiment, the drive system 1440 may include an engine, an energy source, a powertrain, and wheels. The engine may be one or a combination of an internal combustion engine, an electric motor, and an air compression engine. The engine can convert the energy provided by the energy source into mechanical energy.

[0230] Some or all functions of vehicle 1400 are controlled by the computing platform 1450. The computing platform 1450 may include at least one processor 1451 and a memory 1452. The processor 1451 may execute instructions 1453 stored in the memory 1452.

[0231] The processor 1451 may be any conventional processor, such as a commercially available CPU. The processor may also include, such as a data processor (Graphic Process Unit, GPU), a field programmable gate array (Field Programmable Gate Array, FPGA), a system on chip (System on Chip, SOC), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), or a combination thereof.

[0232] The memory 1452 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0233] In addition to the instructions 1453, the memory 1452 can also store data, such as road maps, route information, data such as the position, direction, speed of the vehicle, etc. The data stored in the memory 1452 can be used by the computing platform 1450.

[0234] In an embodiment of the present disclosure, the processor 1451 can execute the instructions 1453 to complete all or part of the steps of the above-mentioned driving path boundary generation method.

[0235] In some embodiments of the present disclosure, a non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by a processor of a mobile terminal, enables the mobile terminal to execute the above-mentioned driving path boundary generation method.

[0236] Those skilled in the art will readily conceive of other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure, which follow the general principles of the present disclosure and include known common knowledge or conventional technical means in the technical field not disclosed in the present disclosure. The specification and embodiments are only to be considered as exemplary, and the true scope and spirit of the present disclosure are pointed out by the following claims.

[0237] It should be understood that the present disclosure is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present disclosure is only limited by the appended claims.

Claims

1. A method for generating a driving path boundary, characterized in that, Including: Obtain the initial vehicle pose and the target end position of the target vehicle; Based on the initial vehicle pose and the target end position, generate an initial driving trajectory using a cubic optimal curve planning; Based on the initial driving trajectory, determine the maximum lateral position of the target vehicle deviating from the reference line; Based on the actual driving boundary condition of the target vehicle and the maximum lateral position, modify the boundary of the initial driving trajectory to generate a target driving path boundary.

2. The driving path boundary generation method according to claim 1, characterized in that, Based on the initial driving trajectory, determining the maximum lateral position of the target vehicle deviating from the reference line includes: Based on the initial driving trajectory, calculate an array of trajectory points of the target vehicle; wherein, the array of trajectory points contains multiple trajectory point data groups, and each trajectory point data group contains the front axle center position and the rear axle center position of the target vehicle in this trajectory point; Based on the multiple front axle center positions and the multiple rear axle center positions in the multiple trajectory point data groups, determine the maximum lateral position of the target vehicle deviating from the reference line.

3. The driving path boundary generation method according to claim 2, wherein Based on the initial driving trajectory, calculating an array of trajectory points of the target vehicle includes: Sample the initial driving trajectory to obtain multiple trajectory points; Take each of the trajectory points as the multiple rear axle center positions of the target vehicle, and calculate the front axle center positions corresponding to the multiple rear axle center positions based on the multiple rear axle center positions; Based on the front axle center positions corresponding to the multiple rear axle center positions and the multiple rear axle center positions, obtain the array of trajectory points.

4. The driving path boundary generation method according to claim 3, wherein Based on the multiple front axle center positions and the multiple rear axle center positions in the multiple trajectory point data groups, determining the maximum lateral position of the target vehicle deviating from the reference line includes: Obtain the reference line position corresponding to each of the trajectory points; wherein, the reference line includes the lane center line; Based on the reference line position corresponding to each of the trajectory points, the multiple front axle center positions and the multiple rear axle center positions, calculate multiple first lateral distances of each front axle center position deviating from the reference line and multiple second lateral distances of each rear axle center position deviating from the reference line; Select the maximum lateral distance with the largest value among the multiple first lateral distances and the multiple second lateral distances, and take the position point corresponding to the maximum lateral distance as the maximum lateral position.

5. The method for generating a driving path boundary according to claim 3, wherein Based on the actual driving boundary condition of the target vehicle and the maximum lateral position, modifying the boundary of the initial driving trajectory to generate a target driving path boundary includes: In response to the actual driving boundary condition of the target vehicle being a crossable boundary, generate the boundary of the initial driving trajectory based on the vehicle width of the target vehicle and the initial driving trajectory; Use the maximum lateral position to modify the boundary of the initial driving trajectory to generate a first driving path boundary; Based on the driving obstacle information, perform a collision-free risk correction on the first driving path boundary to obtain the target driving path boundary.

6. The driving path boundary generation method according to claim 5, wherein Using the maximum lateral position to modify the boundary of the initial driving trajectory to generate a first driving path boundary includes: Calculate multiple offsets of the boundaries of the initial driving trajectory corresponding to each of the trajectory points from the reference line; Compare each of the offsets with the maximum lateral position; In response to the maximum lateral position being greater than the first offset, expand the boundary corresponding to the first offset using the maximum lateral position to generate the first driving path boundary; Wherein, the multiple offsets include the first offset.

7. The driving path boundary generation method according to claim 5, wherein Further includes: Calculate the self-vehicle obstacle box positions corresponding to the multiple rear axle center positions based on the multiple rear axle center positions; Generate the driving obstacle information based on the self-vehicle obstacle box positions.

8. The driving path boundary generation method according to claim 5, characterized in that, Further includes: Based on the actual driving boundary condition of the target vehicle, determine whether all of the actual driving boundaries corresponding to the initial driving trajectory are non-collision type boundaries; If not, then the actual driving boundary condition of the target vehicle is a crossable boundary.

9. The driving path boundary generation method according to claim 8, wherein Further includes: In response to there being some non-collision type boundaries in the actual driving boundaries corresponding to the initial driving trajectory, generate non-collision boundary information; Generate the driving obstacle information based on the non-collision boundary information.

10. The driving path boundary generation method according to claim 1, wherein Generate an initial driving trajectory using cubic optimal curve planning based on the initial vehicle pose and the target end position, including: Set constraint conditions; the constraint conditions include: the lateral acceleration and lateral jerk of the target vehicle at the target end are zero; Use the cubic optimal curve planning to generate the initial driving trajectory that satisfies the initial vehicle pose, the constraint conditions, and the target end position.

11. The driving path boundary generation method according to claim 1, characterized in that Further includes: Obtain the obstacle information around the target vehicle; Based on the obstacle information, calculate the avoidance distance by which the target vehicle offsets from the reference line after a preset time period using spatio-temporal joint decision-making; Combine the reference line and the avoidance distance to generate the target end position.

12. The driving path boundary generation method according to claim 1, wherein Further includes: Judge the driving scenario of the target vehicle based on the initial vehicle pose and the reference line; Wherein, obtaining the initial vehicle pose and the target end position of the target vehicle includes: In response to the driving scenario of the target vehicle being an expandable boundary scenario, obtain the initial vehicle pose and the target end position of the target vehicle.

13. The driving path boundary generation method according to claim 12, wherein Judging the driving scenario of the target vehicle based on the initial vehicle pose and the reference line includes: Based on the initial vehicle pose and the reference line, calculate the angle between the target vehicle and the reference line, and if the angle is greater than a preset angle threshold, then the driving scenario of the target vehicle is a non-expandable boundary scenario; Based on the initial vehicle pose, determine that the target vehicle is in a preset steering driving situation, then the driving scenario of the target vehicle is a non-expandable boundary scenario; Based on the reference line, determine the curvature of the reference line within a preset range in front of the target vehicle, and if the curvature is greater than a preset curvature threshold, then the driving scenario of the target vehicle is a non-expandable boundary scenario; Otherwise, the driving scenario of the target vehicle is an expandable boundary scenario.

14. A driving path boundary generation device, characterized in that, Includes: A data acquisition unit for acquiring the initial vehicle pose and the target end position of the target vehicle; An initial driving trajectory generation unit, configured to generate an initial driving trajectory based on the initial vehicle pose and the target end position by using a cubic optimal curve planning; A maximum lateral position determination unit, configured to determine the maximum lateral position of the target vehicle deviating from the reference line based on the initial driving trajectory; A driving path boundary generation unit, configured to modify the boundary of the initial driving trajectory based on the actual driving boundary condition of the target vehicle and the maximum lateral position to generate a target driving path boundary.

15. A vehicle, characterized in that, Comprising: A processor; A memory for storing instructions executable by the processor; Wherein, the processor is configured to: Implement the driving path boundary generation method according to any one of claims 1 to 13.

16. A non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by a processor of a mobile terminal, enabling the mobile terminal to execute a driving path boundary generation method according to any one of claims 1 to 13.

Images