Vehicle trajectory planning method and device, intelligent terminal and storage medium

Abstract

The invention relates to the technical field of intelligent driving, in particular to a vehicle trajectory planning method and device, an intelligent terminal and a storage medium. The method comprises the following steps: acquiring vehicle information, driving information, driving environment information and a trajectory planning starting point, and adding the trajectory planning starting point into a candidate node set; a preset cost function is utilized to calculate cost values of nodes in the candidate node set, the node with the minimum cost value is determined as a first target node, and the preset cost function comprises a heuristic cost function determined based on the direction error and the lane reference line; track searching is carried out according to the vehicle information, the driving information and a preset searching decision, and a first target child node with the first target node as a father node is generated; determining whether the trajectory search is completed based on a preset end condition; and when it is determined that the trajectory search is completed, determining a planned trajectory of the vehicle based on a preset ending condition. By adopting the method, the vehicle trajectory planning speed can be optimized, and the trajectory quality can be improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology, and in particular to a vehicle trajectory planning method, device, intelligent terminal and storage medium. Background Art

[0002] In recent years, with the development of artificial intelligence technology, vehicle autonomous driving has also become a research hotspot. Trajectory planning is an important part of autonomous driving. Trajectory planning refers to local path planning. It is necessary to plan a collision-free and smooth feasible trajectory based on the vehicle's current posture provided by the positioning module, lane boundaries, lane reference lines, driving space boundaries provided by the map module, obstacle information provided by the perception module, traffic lights and other elements to ensure safe and comfortable driving of autonomous vehicles.

[0003] Structured roads are one of the main scenarios faced by autonomous vehicles. Structured roads generally refer to highways, urban trunk roads and other roads with obvious traffic signs and road elements. The background environment of such roads is relatively simple, and the geometric characteristics of the roads are relatively obvious. Vehicles often travel at a relatively high speed on structured roads. How to plan a high-quality trajectory in a short period of time is an issue that needs to be considered urgently. Summary of the invention

[0004] In view of this, the embodiments of the present application provide a vehicle trajectory planning method, device, intelligent terminal and storage medium, which can optimize the speed of vehicle trajectory planning and improve the trajectory quality.

[0005] A first aspect of an embodiment of the present application provides a vehicle trajectory planning method, comprising:

[0006] Acquire vehicle information, driving information, driving environment information, and a trajectory planning starting point, and add the trajectory planning starting point to a candidate node set;

[0007] Calculating the cost values ​​of the nodes in the candidate node set by using a preset cost function, and determining the node with the smallest cost value in the current candidate node set as the first target node, wherein the preset cost function includes a heuristic cost function determined based on the direction error and the lane reference line;

[0008] Performing a trajectory search according to the vehicle information, the driving information and a preset search decision to generate a first target child node with the first target node as a parent node;

[0009] When the number of the first target sub-nodes reaches a preset number of sub-nodes, determining whether the trajectory search is completed based on a preset end condition;

[0010] When it is determined that the trajectory search is completed, a planned trajectory of the vehicle is determined based on the preset end condition.

[0011] A second aspect of an embodiment of the present application provides a vehicle trajectory planning device, the device comprising:

[0012] A starting point and information acquisition unit, used to acquire vehicle information, driving information, driving environment information and a trajectory planning starting point, and add the trajectory planning starting point to a candidate node set;

[0013] A parent node selection unit, configured to calculate the cost values ​​of the nodes in the candidate node set by using a preset cost function, and determine the node with the smallest cost value in the current candidate node set as the first target node, wherein the preset cost function includes a heuristic cost function determined based on a direction error and a lane reference line;

[0014] A first search unit, configured to perform a trajectory search according to the vehicle information, the driving information and a preset search decision, and generate a first target child node with the first target node as a parent node;

[0015] a completion judgment unit, configured to determine whether the trajectory search is completed based on a preset end condition when the number of the first target sub-nodes reaches a preset number of sub-nodes;

[0016] The trajectory planning unit is used to determine the planned trajectory of the vehicle based on the preset end condition when it is determined that the trajectory search is completed.

[0017] A third aspect of an embodiment of the present application provides an intelligent terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the vehicle trajectory planning method provided in the first aspect of the embodiment of the present application are implemented.

[0018] The fourth aspect of the embodiments of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the vehicle trajectory planning method provided in the first aspect of the embodiments of the present application.

[0019] The fifth aspect of the embodiments of the present application provides a computer program product. When the computer program product is run on a terminal device, the terminal device executes the steps of the vehicle trajectory planning method described in the first aspect of the embodiments of the present application.

[0020] In the embodiment of the present application, vehicle information, driving information, driving environment information and trajectory planning starting point are obtained, and the trajectory planning starting point is added to the candidate node set, and the cost value of the node in the candidate node set is calculated using a preset cost function, and the preset cost function includes a heuristic cost function determined based on the direction error and the lane reference line, which is conducive to improving the solution speed of trajectory planning and avoiding unnecessary search redundancy. The node with the smallest cost value in the current candidate node set is determined as the first target node, and then the trajectory search is performed according to the vehicle information, the driving information and the preset search decision, and the first target child node with the first target node as the parent node is generated. When the number of the first target child nodes reaches the preset number of child nodes, it is determined whether the trajectory search is completed based on the preset end condition. When it is determined that the trajectory search is completed, the planned trajectory of the vehicle is determined based on the preset end condition. The present application scheme uses a preset cost function including an improved heuristic cost function to calculate the node cost value, which can reduce the calculation complexity. At the same time, the node for trajectory planning is determined by accurately calculating the cost value, which can effectively improve the quality of the trajectory. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0022] Figure 1 is a flow chart of the implementation of the vehicle trajectory planning method provided in the embodiment of the present application;

[0023] Figure 2 This is a specific implementation flow chart of calculating node cost values ​​using a preset cost function in the vehicle trajectory planning method provided in the embodiment of the present application;

[0024] Figure 3 It is another specific implementation flow chart of calculating node cost values ​​by using a preset cost function in the vehicle trajectory planning method provided in the embodiment of the present application;

[0025] Figure 4 This is another specific implementation flowchart of calculating node cost values ​​using a preset cost function in the vehicle trajectory planning method provided in an embodiment of the present application;

[0026] Figure 5 It is a specific implementation flow chart of generating a first target child node with the first target node as the parent node in the vehicle trajectory planning method provided in an embodiment of the present application;

[0027] Figure 5.1It is a schematic diagram of a scene of a front wheel turning angle search range in a vehicle trajectory planning method provided in an embodiment of the present application;

[0028] Figure 6 This is a flowchart for implementing continued search in the vehicle trajectory planning method provided in an embodiment of the present application;

[0029] Figure 7 is a flow chart of an implementation of a vehicle trajectory planning method provided by another embodiment of the present application;

[0030] Figure 8 is a structural block diagram of a vehicle trajectory planning device provided in an embodiment of the present application;

[0031] Fig. 9 It is a schematic diagram of a smart terminal provided in an embodiment of the present application. DETAILED DESCRIPTION

[0032] In the following description, specific details such as specific device structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0033] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0034] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0035] It should be understood that each method embodiment of the present application provides a vehicle trajectory planning method applicable to various types of smart terminals that need to perform trajectory planning, specifically smart vehicle terminals. The present application does not impose any limitation on the type of smart terminals.

[0036] The vehicle trajectory planning method provided in this application is exemplarily described below in conjunction with specific embodiments.

[0037] Figure 1 The implementation process of the vehicle trajectory planning method provided in an embodiment of the present application is shown, and the method process may include the following steps S101 to S105.

[0038] Step S101: Acquire vehicle information, driving information, driving environment information and a trajectory planning starting point, and add the trajectory planning starting point to a candidate node set.

[0039] Vehicle information includes various types of vehicle parameter information, such as the vehicle's front wheel turning angle range, maximum turning radius, vehicle wheelbase, centripetal acceleration limit threshold, etc. Driving information includes the vehicle's current actual position, driving speed, heading angle, front wheel turning angle, etc. Driving environment information includes lane boundaries, road boundaries, lane reference lines, driving space, obstacles, etc. Lane boundaries, road boundaries, lane reference lines, driving space, etc. can be determined through map information, and obstacle information can be obtained through the vehicle's sensors. Among them, driving space refers to the range within which the vehicle is allowed to travel.

[0040] In this embodiment, the trajectory planning starting point is added to the candidate node set, and the candidate node set is used to store candidate nodes. The above trajectory planning starting point is determined according to the current actual position in the above driving information.

[0041] Initialize the starting point of trajectory planning. The coordinates of the starting point of trajectory planning are the center point of the vehicle body. The xy coordinates and heading angle of the starting point of trajectory planning are set to 0 in the vehicle body coordinate system. The front wheel turning angle of the starting point of trajectory planning is obtained from the chassis, the value is set to 0, and the parent node is set to empty. It should be noted that the search space of trajectory planning in this application needs to be determined in the Frenet coordinate system and then converted to the vehicle body xy coordinate system.

[0042] Step S102: Calculate the cost values ​​of the nodes in the candidate node set using a preset cost function, and determine the node with the smallest cost value in the current candidate node set as the first target node, wherein the preset cost function includes a heuristic cost function determined based on the direction error and the lane reference line.

[0043] In this embodiment, a node with the smallest cost value is selected from the current candidate node set as the first target node, and the first target node is the parent node used as the search starting point.

[0044] The initial candidate node set and searched node set are empty sets. The node information in the candidate node set includes cost value, xy coordinates, heading angle, curvature, front wheel angle, grid, parent node, etc.

[0045] In the hybrid A-star algorithm, the heuristic cost function is the most important part. The core of this application lies in the optimization and improvement of the heuristic cost function. The traditional hybrid A-star algorithm uses the Manhattan distance or Euclidean distance from the node to the end point as the heuristic function. Since the vehicle is required to drive forward at high speed and smoothly on the structured road, and the planning needs to make feedback based on the current road conditions, the trajectory is adjusted horizontally or vertically, and there is no need to approach a certain point. The method of using a fixed point as the target is not applicable on structured roads. Therefore, this application introduces the idea of ​​the artificial potential field method to guide the trajectory search and determines the heuristic cost function based on the direction error and the lane reference line.

[0046] As a possible implementation manner of the present application, the direction error includes a heading angle error and a lateral error; the heuristic cost function includes a gravity function, and the cost value includes a gravity value; the gravity value is obtained based on the gravity function; Figure 2 A specific implementation process of calculating the cost value of the nodes in the candidate node set by using a preset cost function in the method provided in the embodiment of the present application is shown, and is described in detail as follows:

[0047] A1: Obtain a heading angle error from the node to a reference node, and a lateral error from the node to the reference node, where the reference node refers to a projection point of the node on the lane reference line.

[0048] Take node xy as the parent node, the parent node coordinates as the starting point of the vector, extend a fixed sampling span in the heading angle direction of the parent node, generate a virtual node at the end point of the vector, and the foot of the virtual node to the lane reference line is the reference node of all child nodes of the parent node.

[0049] A2: Determine a gravity value according to the heading angle error and the lateral error.

[0050] The gravity is provided by the lane reference line to guide the vehicle to travel along the reference line. The reference gravity is divided into two parts, namely the heading angle error and the lateral error from the node to the lane reference line. In this embodiment, the calculation formula of the gravity function is as follows (1):

[0051] AttrPotential=ω phi_attr ×(e |phi_diff| -1)+ω lat_attr ×|lat_diff|+ω lat_phi *phi_diff*lat_diff (1)

[0053] Where AttrPotential is the gravity value, phi_diff represents the heading angle error from the node to the reference node, lat_diff represents the lateral error from the node to the reference node, ω phi_attrrepresents the weight corresponding to phi_diff, ω lat_attr Indicates the weight corresponding to lat_diff, ω lat_phi Represents the cost of multiplying phi_diff and lat_diff.

[0054] On the one hand, considering that as the deviation of the heading angle from the reference line increases linearly, the difficulty and danger of the vehicle returning to the reference line increase exponentially, the vehicle should be allowed to have a certain heading angle deviation from the lane reference line in order to find a better route. When the heading angle error is small, phi_diff accounts for a small proportion of the total cost, and when the heading angle error is large, the influence of phi_diff on the total cost should be dominant; on the other hand, considering the need to integrate the obstacle avoidance function in this algorithm, the restriction on the lateral error is relatively tolerant. Furthermore, considering that the vehicle has a certain lateral error and heading angle error to the lane reference line, when the lateral error and the heading angle error have different signs, it is easier to get close to the lane reference line in the generation of subsequent nodes, and if the two have the same sign, it is more difficult to get close to the reference line. The two nodes with the same lateral error and heading angle error modulus have obvious differences in the degree of this relationship, so this item compensates for the product of the two.

[0055] In some implementations, since the reference line gravity can guide the vehicle to drive close to the lane reference line when there is no repulsive force provided by obstacles, the third term in the calculation formula of the gravity function in the above formula (1) can be omitted in actual implementation. Therefore, in this embodiment, the calculation formula of the gravity function is as follows:

[0056] AttrPotential=ω phi_attr ×(e |phi_diff| -1)+ω lat_attr ×|lat_diff| (2)

[0057] Where AttrPotential is the gravity value, phi_diff represents the heading angle error from the node to the reference node, lat_diff represents the lateral error from the node to the reference node, ω phi_attr represents the weight corresponding to phi_diff, ω lat_attr Indicates the weight corresponding to lat_diff.

[0058] As a possible implementation manner of the present application, the driving environment information further includes obstacles and driving space, the heuristic cost function further includes a repulsion function; the cost value further includes a repulsion value; the repulsion value is obtained based on the repulsion function; Figure 3 Another specific implementation process of calculating the cost value of the node in the candidate node set by using a preset cost function in the method provided in the embodiment of the present application is shown, and is described in detail as follows:

[0059] B1: Obtain the repulsive obstacle set of the node and the resistance of the driving space boundary in the driving space to the node.

[0060] In this embodiment, at the current parent node position, obstacles behind, obstacles outside the driving space, and obstacles beyond a certain distance in front are filtered out, and the remaining obstacles are added to the repulsive obstacle set of the corresponding child node.

[0061] B2: Determine the resistance from the obstacle to the node in the obstacle according to the obstacle and the driving space.

[0062] B3: Determine a repulsion value according to the obstacle, the repulsion obstacle set, the resistance from the obstacle to the node, and the resistance of the driving space boundary to the node.

[0063] In this embodiment, the repulsive force is provided by the obstacle and the driving space, and the purpose is to guide the vehicle to avoid obstacles and avoid being too close to the obstacle and the boundary of the driving space. The calculation formula of the repulsive force function is as follows (3):

[0064] RepuPotential = ω ob_repu ×max(ObRepu(o)) +ω bo_repu ×BoRepu o∈O (3)

[0065] Wherein, RepuPotential represents the repulsion value, o represents an obstacle, O represents a repulsion obstacle set, ObRepu(o) represents a first resistance, which is the resistance of obstacle o to the node, and BoRepu represents a second resistance, which is the resistance of the driving space boundary to the node.

[0066] In a possible implementation manner, the first resistance ObRepu(o) is determined according to the following formula (4.1):

[0067]

[0068] Among them, angle o Indicates the minimum angle difference between the heading angle of the node and the boundary of the obstacle o, distance o represents the distance from the obstacle o to the node, k o Indicates the repulsion gain value corresponding to different obstacle types.

[0069] In a possible implementation manner, the second resistance BoRepu is determined according to the following formula (4.2):

[0070]

[0071] Wherein, distance_bo represents the shortest distance from the node to the boundary of the driving space, and max_distance_bo represents the maximum influence distance of the repulsive force of the driving space boundary.

[0072] The design of obstacle repulsion calculation takes into account that the danger level of obstacles increases sharply as the vehicle approaches, and is strongly related to the angle of the obstacle relative to the vehicle's heading. Therefore, the cost function uses distance as the denominator and decays as an exponential function of the angle. The maximum value of all obstacle repulsion is taken and added to the repulsion calculation. The boundary repulsion is calculated based on the distance from the node to the boundary to prevent the vehicle from being too close to the boundary.

[0073] In the embodiment of the present application, the lane reference line calculates the attraction based on the lateral distance to the reference node and the heading angle error, and the two factors are coupled. The obstacle calculates the repulsion based on the distance and angle to the node, and the repulsion decays with the angle offset. The calculation of the node cost can reduce the calculation time and improve the efficiency of the cost calculation, reflecting the different degrees of influence of obstacles in different directions on structured roads on vehicle driving. The repulsion provided by the driving space boundary prevents the vehicle from being too close to the driving space boundary, which can optimize the trajectory quality and improve the safety of the trajectory.

[0074] B4: Determine a heuristic cost value according to the attraction value and the repulsion value.

[0075] In this embodiment, the heuristic cost function is composed of an attraction function and a repulsion function. The heuristic cost value is determined according to the heuristic cost function, and the node with the smallest heuristic cost value in the current candidate node set is determined as the first target node.

[0076] The calculation formula of the heuristic cost function is as follows (5):

[0077] HeuCost = AttrPotential + RepuPotential (5)

[0078] Among them, HeuCost represents the heuristic cost value, AttrPotential represents the attraction value, and RepuPotential represents the repulsion value. The calculation of the attraction value and repulsion value refers to the above content.

[0079] In an embodiment of the present application, before a parent node generates a child node, the heuristic conditions of the child node are calculated based on the parent node to determine the gravitational reference node and repulsive obstacle set of the child node, so as to avoid repeatedly calculating these conditions for each child node. This can effectively reduce the algorithm running time and thus improve the speed of trajectory planning.

[0080] As a possible implementation of this application, Figure 4Another specific implementation process of calculating the cost value of the node in the candidate node set by using a preset cost function in the method provided in the embodiment of the present application is shown, and is described in detail as follows:

[0081] C1: Calculate the heuristic cost value using the heuristic cost function. The calculation of the heuristic cost value refers to the above content.

[0082] C2: Calculate the path cost value using the path cost function.

[0083] The path cost value of a node is the path cost value of the parent node plus the path cost function value. As a possible implementation method of the present application, in the method provided in the embodiment of the present application, a specific implementation process of calculating the path cost value using the path cost function specifically includes:

[0084] C11: Obtain the front wheel steering angle corresponding to the node, the change value of the front wheel steering angle from the parent node of the node to the node, and the displacement from the parent node of the node to the projection point of the node on the lane reference line along the direction of the lane reference line.

[0085] C12: Determine the path cost value according to the front wheel steering angle, the change value and the displacement.

[0086] In this embodiment, the calculation formula of the above path cost function is as follows (6):

[0087] TrajCost = ω yaw *|yaw| + ω d_yaw *|d_yaw| - ω d_s *d_s (6)

[0088] Wherein, TrajCost represents the path cost, yaw represents the front wheel turning angle of the node, d_yaw represents the change in the front wheel turning angle from the parent node to the node, d_s represents the displacement from the parent node to the projection point of the node on the lane reference line along the direction of the lane reference line, ω yaw represents the weight of yaw in the cost function of the path, ω d_yaw represents the weight of d_yaw in the cost function of the path, ω d_s Represents the weight of d_s in the path cost function.

[0089] In a possible implementation, the path curvature is obtained, the path curvature is used to replace the front wheel turning angle in the above formula (6), and the path cost value is calculated based on the replaced formula (6).

[0090] In the embodiment of the present application, the cost value is increased according to the steering wheel angle and the change in the steering wheel angle, the cumulative amplitude of the steering wheel rotation in the path is reduced as much as possible, and the cost value is reduced according to the displacement along the lane reference line direction, and the guidance algorithm gives priority to the nodes farther along the lane reference line direction. This is beneficial to improve the solution speed of trajectory planning and avoid unnecessary search redundancy. At the same time, because the s coordinate span on the inside of the curve is larger than that on the outside of the curve for the same displacement along the s direction in the Frenet coordinate system, the embodiment of the present application can also guide the algorithm to give priority to nodes on the inside of the curve when the road is curved.

[0091] C2: Determine the cost value of the node according to the sum of the heuristic cost value and the path cost value.

[0092] In this embodiment, the cost value of the node is determined by combining the heuristic cost value and the path cost value, which further improves the accuracy of the cost value calculation, thereby more effectively ensuring that the vehicle travels within the lane while improving the obstacle avoidance capability.

[0093] Step S103: performing a trajectory search according to the vehicle information, the driving information and a preset search decision, and generating a first target child node with the first target node as a parent node.

[0094] In this embodiment, the search space of trajectory planning needs to be determined in the Frenet coordinate system and then converted to the vehicle body xy coordinate system. The preset search decision includes the following:

[0095] 1. Based on the map information, with the current lane as the center, if there are adjacent lanes in the same direction on both sides that are allowed to use the lane according to traffic regulations, the trajectory search space will be expanded horizontally to both sides. When the expansion width on one side reaches half the lane width or reaches the boundary of the driving space, the expansion will stop;

[0096] 2. According to the adjacent lane with expanded width, check the rear vehicles in the lane. If the distance from the rear vehicle to the vehicle is less than max(safe_distance, (obstacle_speed-adc_speed)*relative_safe_ratio, obstacle_speed*absolute_back_safe_ratio), then expand a lateral buffer distance (different distances are preset according to the obstacle type) to the rear vehicle, and reduce the trajectory search space according to the l-direction coordinates of the rear vehicle close to the edge of the current direction of the vehicle after expansion. The search space is reduced to the current lane boundary at most. safe_distance represents the minimum safe distance, relative_safe_ratio represents the relative safe distance ratio, absolute_back_safe_ratio represents the absolute safe distance ratio, obstacle_speed represents the speed of the rear vehicle, adc_speed represents the speed of the vehicle, and adc represents the intelligent driving vehicle.

[0097] 3: Convert the boundary of the driving space and the coordinates of the current lane reference line to the vehicle body xy coordinate system with the center point of the vehicle body as the origin.

[0098] As a possible implementation of this application, Figure 5 In the method embodiment provided by the embodiment of the present application, a specific implementation process of performing a trajectory search according to the vehicle information, the driving information and the preset search decision to generate a first target child node with the first target node as the parent node is shown, which is detailed as follows:

[0099] D1: Determine the front wheel turning angle range to be searched according to the vehicle information, the driving information and the preset search decision.

[0100] In a possible implementation manner, the vehicle information includes a maximum angle of a front wheel steering angle of the vehicle, a centripetal acceleration limit threshold, and a vehicle wheelbase, and the driving information includes a current driving speed; the above-mentioned determining the searched front wheel steering angle range based on the vehicle information, the driving information, and a preset search decision includes:

[0101] D11: Obtain the front wheel turning angle corresponding to the first target node;

[0102] D12: determining a maximum front wheel turning angle at the current driving speed according to the current driving speed, the centripetal acceleration limit threshold of the vehicle and the wheelbase of the vehicle;

[0103] In this example, based on vehicle kinematics, the maximum front wheel steering angle max_yaw at the current driving speed is determined according to the following formula (7): v :

[0104] max_yaw v = arctan(cent_acc * wheel_base / v2) (7)

[0105] Among them, cent_acc represents the centripetal acceleration limit threshold, wheel_base represents the vehicle wheelbase, and v represents the current driving speed.

[0106] D13: Determine a maximum front wheel steering angle change value according to the current driving speed;

[0107] D14: Determine the searched front wheel turning angle range according to the maximum front wheel turning angle, the front wheel turning angle corresponding to the target node, the maximum front wheel turning angle at the current driving speed, and the maximum front wheel turning angle change value.

[0108] D2: Based on the preset sampling span and the preset number of sub-nodes, the sub-nodes are uniformly searched in the front wheel turning angle range. Specifically, the node information searched includes the coordinates, heading angle, front wheel turning angle and curvature of the sub-nodes.

[0109] In this embodiment, based on vehicle kinematics, a child node is generated from the first target node, the maximum angle of the front wheel steering angle max_yaw is obtained from the vehicle information, and the maximum front wheel steering angle max_yaw allowed at the current driving speed is calculated according to the current driving speed v, the centripetal acceleration limit threshold cent_acc, and the vehicle wheelbase wheel_base. v , obtain the front wheel steering angle curr_yaw of the first target node, and determine the front wheel steering angle range of the child node search as [max(-max_yaw, -max_yaw) according to the maximum front wheel steering angle change value max_d_yaw allowed at the current driving speed v , curr_yaw-max_d_yaw), min(max_yaw, max_yaw v , curr_yaw+max_d_yaw)], within the front wheel turning angle range, according to the preset number of child nodes next_node_num of the child nodes, the child nodes are evenly searched within the front wheel turning angle search range with a certain forward sampling span, and the coordinates, heading angle, front wheel turning angle and curvature of the child nodes are calculated.

[0110] Exemplarily, the search range of the front wheel angle is shown as follows: Figure 5.1 In the figure, sector 1 represents the sampling range corresponding to the maximum front wheel steering angle interval [-max_yaw, max_yaw] of the vehicle, and sector 2 represents the maximum front wheel steering angle [-max_yaw v , max_yaw v] corresponds to the sampling range, and sector 3 represents the sampling range corresponding to the maximum front wheel steering angle change value [curr_yaw-max_d_yaw, curr_yaw+max_d_yaw] under the front wheel steering angle of the current node.

[0111] D3: Verify whether the searched child node is valid.

[0112] In a possible implementation manner, in the method embodiment provided in the embodiment of the present application, a specific implementation process of verifying whether the searched child node is valid specifically includes:

[0113] D31: Determine whether there is an equivalent node of the searched child node in the target set, the target set includes the candidate node set and the searched node set, and the searched node set is used to store the node selected as the parent node from the candidate node set.

[0114] In this embodiment, equivalent nodes refer to nodes with the same grid. Two nodes are equivalent nodes if their grids are equal in three dimensions.

[0115] D32: If there is no equivalent node of the child node in the candidate node set and the searched node set, verify whether the vehicle polygon generated on the child node meets preset conditions, wherein the preset conditions include not colliding with obstacles and not touching the boundary of the trajectory search space.

[0116] D33: If it is verified that the child node meets the preset condition, the child node is determined to be valid.

[0117] If the child node does not meet the preset condition, that is, the child node collides with an obstacle and / or touches the boundary of the trajectory search space, the child node is determined to be invalid. The child node that is verified to be invalid is stored in the searched node set.

[0118] D4: Determine the child node verified to be valid as the first target child node with the first target node as the parent node.

[0119] In this embodiment, the first target child node is a valid child node that takes the first target node as a parent node.

[0120] For the child nodes that are verified to be valid, their cost values ​​are calculated. The method for calculating the child node cost value is the same as the method for calculating the node cost value using the preset cost function mentioned above, which will not be repeated here.

[0121] In an embodiment of the present application, the searched sub-nodes are effectively verified, invalid sub-nodes are filtered out, and valid sub-nodes are retained, which can improve the effectiveness of the search.

[0122] Step S104: When the number of the first target sub-nodes reaches a preset number of sub-nodes, determining whether the trajectory search is completed based on a preset end condition.

[0123] In this embodiment, when the number of valid child nodes of the first target node as a parent node reaches a preset number of child nodes, it is determined whether the trajectory search is completed. The above preset end condition includes a first preset end condition related to the first target node and a second preset end condition related to the candidate node set.

[0124] As a possible implementation manner of the present application, the preset end condition is any one of a first preset end condition and a second preset end condition, the first preset end condition is related to the first target node, and the second preset end condition is related to the candidate node set; the driving environment information includes a lane reference line, and when the preset end condition is the first preset end condition, determining whether the trajectory search is completed is a specific implementation process, specifically including:

[0125] E1: Obtain the displacement from the trajectory planning starting point to the first target node along the lane reference line. The displacement is the length of the displacement of the first target node from the trajectory planning starting point projected onto the lane reference line.

[0126] E2: If the displacement has reached a preset track length, it is determined that the track search is completed. The preset track length is generally a preset maximum search track length.

[0127] When the preset end condition is the second preset end condition, a specific implementation process of determining whether the trajectory search is completed specifically includes:

[0128] E3: Move the node selected as the parent node in the candidate node set to the searched node set.

[0129] Exemplarily, the candidate node set is open_set, and the searched node set is close_set. After determining the node with the smallest cost value (ie, the first target node or the second target node) from the open_set, the first target node is moved from the open_set to the close_set.

[0130] E4: When the candidate node set is an empty set, it is determined that the trajectory search is completed.

[0131] As a possible implementation of the present application, Figure 6 As shown, the method embodiment provided in the embodiment of the present application also includes:

[0132] F1: When the number of the first target child nodes does not reach the preset number of child nodes, continue the trajectory search according to the vehicle information, the driving information and the preset search decision, and generate a new first target child node with the first target node as the parent node.

[0133] F2: Determine whether the number of first target child nodes with the first target node as a parent node reaches a preset number of child nodes.

[0134] F3: If it is determined that the preset number of child nodes is reached, determine whether the trajectory search is completed.

[0135] In an embodiment of the present application, when the number of first target child nodes with the first target node as the parent node does not meet the requirement, the trajectory search continues to be performed to generate a new first target child node with the first target node as the parent node to ensure the integrity of the search, thereby improving the effectiveness of the search.

[0136] Step S105: When it is determined that the trajectory search is completed, the planned trajectory of the vehicle is determined based on the preset end condition.

[0137] In this embodiment, the vehicle plans a trajectory and outputs nodes including position, heading angle and curvature as the planned path. In order to facilitate the subsequent processing of other submodules in the planning module such as trajectory smoothing and speed planning, the path composed of nodes finally needs to be converted to a set of trajectory points in the Frenet coordinate system and output. The converted trajectory points contain information such as sl coordinates and curvature in the Frenet coordinate system.

[0138] As a possible implementation mode of the present application, a specific implementation process of determining the planned trajectory of the vehicle based on the preset end condition specifically includes:

[0139] G1: When the preset end condition is the first preset end condition, a backtracking path of the first target node is obtained, where the backtracking path refers to a path from the first target child node backtracking to the parent node step by step to the starting point of the trajectory planning.

[0140] G2: Based on the backtracking path, determine the planned trajectory of the vehicle.

[0141] In this embodiment, by reversing the backtracking path, a path generated from the trajectory planning starting point, sequentially and reversely passing through the parent nodes step by step until the first target child node is determined as the planned trajectory of the vehicle.

[0142] As a possible implementation mode of the present application, in the method embodiment provided in the embodiment of the present application, another specific implementation process of determining the planned trajectory of the vehicle based on the preset end condition specifically includes:

[0143] G3: When the preset end condition is the second preset end condition, the searched nodes that meet the condition along the lane reference line are traced back to the parent nodes step by step until the trajectory planning starting point. The node that meets the condition refers to the node farthest from the vehicle along the lane reference line.

[0144] G4: Generate a trajectory from the starting point of the trajectory planning to the node that meets the conditions. Specifically, starting from the trajectory planning starting point, the path generated is determined as the planned trajectory of the vehicle by sequentially going through the parent node of the node farthest from the vehicle along the lane reference line in reverse order until the node farthest from the vehicle along the lane reference line.

[0145] In an embodiment of the present application, if the end condition of the trajectory search is that the first target node reaches a preset trajectory length, then the parent node of the current first target node is traced back step by step until the trajectory planning starting point, and the trajectory from the trajectory planning starting point to the current first target node is generated as the planned trajectory of the vehicle; if the end condition is that open_set is an empty set, then the parent node of the searched farthest node along the reference line direction is traced back step by step until the trajectory planning starting point, and the trajectory planning starting point is traced back in reverse order through the parent node of the farthest node step by step until the farthest node, and a trajectory from the trajectory planning starting point to the node farthest from the vehicle along the lane reference line direction is generated, and this trajectory is the planned trajectory of the vehicle.

[0146] As a possible implementation of this application, Figure 7 Another method embodiment provided by the present application is shown, which specifically includes:

[0147] S201: updating the candidate node set, wherein the updating includes adding the first target child node after calculating the cost value to the candidate node set.

[0148] S202: When it is determined that the trajectory search is not completed, select a second target node from the updated candidate node set as a parent node, where the second target node is a node with the smallest cost value in the updated candidate node set.

[0149] S203: Performing a trajectory search according to the vehicle information, the driving information and the preset search decision, generating a second target subnode of the second target node, and adding the second target subnode to the candidate node set.

[0150] S204: When the number of the second target sub-nodes reaches a preset number of sub-nodes, determining whether the trajectory search is completed based on the preset end condition.

[0151] S205: When it is determined that the trajectory search is completed, determining a planned trajectory of the vehicle based on the preset end condition.

[0152] In an embodiment of the present application, the candidate node set is updated according to the searched first target child node. When it is determined that the trajectory search is not completed, the node with the smallest cost value is reselected from the updated candidate node set as the new parent node to continue the trajectory search.

[0153] As can be seen from the above, in the embodiments of the present application, the present application integrates the idea of ​​artificial potential field method under the framework of hybrid A-star algorithm, extracts the structured road trajectory planning features, and develops a trajectory planning algorithm suitable for structured roads by combining the advantages of both; the lateral distance and heading angle error from the trajectory node to the reference node are used as gravity to guide the vehicle along the reference line while avoiding excessive dependence on the reference line, and can quickly generate high-quality trajectories while expanding the search space, reducing redundant searches for trajectories; obstacles and lane boundaries are used to provide repulsive forces, and gravity is combined to calculate a reasonable trajectory for obstacle avoidance, and the obstacle avoidance function is integrated while overcoming the combinatorial explosion dilemma in the trajectory search process, overcoming the disturbance of vehicle behavior near the obstacle avoidance condition threshold, and effectively optimizing the trajectory quality.

[0154] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0155] Corresponding to the vehicle trajectory planning method described in the above embodiment, Figure 8 A structural block diagram of a vehicle trajectory planning device provided in an embodiment of the present application is shown. For ease of explanation, only the parts related to the embodiment of the present application are shown.

[0156] Reference Figure 8 The device is applied to an intelligent terminal, and includes: a starting point and information acquisition unit 81, a parent node selection unit 82, a first search unit 83, a completion judgment unit 84, and a trajectory planning unit 85, wherein:

[0157] A starting point and information acquisition unit 81 is used to acquire vehicle information, driving information, driving environment information and a trajectory planning starting point, and add the trajectory planning starting point to a candidate node set;

[0158] A parent node selection unit 82, configured to calculate the cost values ​​of the nodes in the candidate node set by using a preset cost function, and determine the node with the smallest cost value in the current candidate node set as the first target node, wherein the preset cost function includes a heuristic cost function determined based on a direction error and a lane reference line;

[0159] A first search unit 83, configured to perform a trajectory search according to the vehicle information, the driving information and a preset search decision, and generate a first target child node with the first target node as a parent node;

[0160] A completion judgment unit 84, configured to determine whether the trajectory search is completed based on a preset end condition when the number of the first target sub-nodes reaches a preset number of sub-nodes;

[0161] The trajectory planning unit 85 is used to determine the planned trajectory of the vehicle based on the preset end condition when it is determined that the trajectory search is completed.

[0162] As a possible implementation of the present application, the first search unit 83 includes:

[0163] A front wheel steering angle range determination module, used to determine the searched front wheel steering angle range according to the vehicle information, the driving information and a preset search decision;

[0164] A subnode search module, configured to uniformly search for subnodes in the front wheel turning angle range based on a preset sampling span and the preset number of subnodes;

[0165] The subnode verification module is used to verify whether the searched subnode is valid;

[0166] The first target child node determination module is used to determine a child node that is verified to be valid as a first target child node that takes the first target node as a parent node.

[0167] As a possible implementation manner of the present application, the vehicle information includes the maximum angle of the front wheel turning angle of the vehicle, the centripetal acceleration limit threshold and the vehicle wheelbase, and the driving information includes the current driving speed; the front wheel turning angle range determination module is specifically used for:

[0168] Obtaining the front wheel turning angle corresponding to the first target node;

[0169] Determining a maximum front wheel turning angle at the current driving speed according to the current driving speed, the centripetal acceleration limit threshold of the vehicle, and the vehicle wheelbase;

[0170] Determining a maximum front wheel steering angle change value according to the current driving speed;

[0171] The searched front wheel turning angle range is determined according to the maximum front wheel turning angle, the front wheel turning angle corresponding to the target node, the maximum front wheel turning angle at the current driving speed, and the maximum front wheel turning angle change value.

[0172] As a possible implementation manner of the present application, the direction error includes a heading angle error and a lateral error; the heuristic cost function includes a gravity function; the cost value includes a gravity value; the gravity value is obtained based on the gravity function; the parent node selection unit 82 includes:

[0173] An error acquisition module, used to acquire a heading angle error from the node to a reference node, and a lateral error from the node to the reference node;

[0174] The gravity value determination module is used to determine the gravity value according to the heading angle error and the lateral error.

[0175] As a possible implementation of the present application, the calculation formula of the gravity function is as follows:

[0176] AttrPotential=ω phi_attr ×(e |phi_diff| -1)+ω lat_attr ×|lat_diff|

[0177] Where AttrPotential is the gravity value, phi_diff represents the heading angle error from the node to the reference node, lat_diff represents the lateral error from the node to the reference node, ω phi_attr represents the weight corresponding to phi_diff, ω lat_attr Indicates the weight corresponding to lat_diff.

[0178] As a possible implementation manner of the present application, the driving environment information further includes obstacles and driving space, the heuristic cost function further includes a repulsion function; the cost value further includes a repulsion value; the repulsion value is obtained based on the repulsion function; the above-mentioned parent node selection unit 82 also includes:

[0179] Obtaining a repulsive obstacle set of the node and a resistance of a driving space boundary in the driving space to the node;

[0180] Determining, according to the obstacle and the driving space, a resistance from the obstacle to the node;

[0181] Determining a repulsion value according to the obstacle, the repulsion obstacle set, the resistance from the obstacle to the node, and the resistance from the driving space boundary to the node;

[0182] A heuristic cost value is determined according to the attraction value and the repulsion value.

[0183] As a possible implementation of the present application, the calculation formula of the repulsion function is as follows:

[0184] RepuPotential=ω ob_repu ×max(ObRepu(o))+ω bo_repu ×BoRepu o∈O

[0185] Wherein, RepuPotential represents the repulsion value, o represents an obstacle, O represents a repulsion obstacle set, ObRepu(o) represents a first resistance, which is the resistance of obstacle o to the node, and BoRepu represents a second resistance, which is the resistance of the driving space boundary to the node.

[0186] As a possible implementation manner of the present application, the first resistance ObRepu(o) is determined according to the following formula:

[0187]

[0188] Among them, angle o Indicates the minimum angle difference between the heading angle of the node and the boundary of the obstacle o, distance o represents the distance from the obstacle o to the node, k o Indicates the repulsion gain value corresponding to different obstacle types;

[0189] The second resistance BoRepu is determined according to the following formula:

[0190]

[0191] Wherein, distance_bo represents the shortest distance from the node to the boundary of the driving space, and max_distance_bo represents the maximum influence distance of the repulsive force of the driving space boundary.

[0192] As a possible implementation of the present application, the preset cost function also includes a path cost function; the parent node selection unit 82 includes:

[0193] A heuristic cost value calculation module, used to calculate the heuristic cost value using the heuristic cost function;

[0194] A path cost value calculation module, used to calculate the path cost value using the path cost function;

[0195] The cost value determination module is used to determine the cost value of the node according to the sum of the heuristic cost value and the path cost value.

[0196] As a possible implementation of the present application, the above-mentioned path cost calculation module is specifically used for:

[0197] Obtaining the front wheel steering angle corresponding to the node, the change value of the front wheel steering angle from the parent node of the node to the node, and the displacement of the projection point from the parent node of the node to the node on the lane reference line along the direction of the lane reference line;

[0198] The path cost value is determined according to the front wheel turning angle, the change value and the displacement.

[0199] As a possible implementation manner of the present application, the preset end condition is any one of a first preset end condition and a second preset end condition, the first preset end condition is related to the first target node, and the second preset end condition is related to the candidate node set; the driving environment information includes a lane reference line; the completion judgment unit 84 includes:

[0200] a displacement acquisition module, configured to acquire a displacement of the first target subnode from the lane reference line when the preset end condition is the first preset end condition;

[0201] A first completion determination module, configured to determine that the trajectory search is completed if the displacement has reached a preset trajectory length;

[0202] A node transfer module, used for transferring the node selected as the parent node in the candidate node set to the searched node set when the preset end condition is the second preset end condition;

[0203] The second completion determination module is used to determine that the trajectory search is completed when the candidate node set is an empty set.

[0204] As a possible implementation of the present application, the trajectory planning unit 85 includes:

[0205] A first path backtracking module, used for obtaining a backtracking path of the first target node when the preset end condition is the first preset end condition, wherein the backtracking path refers to a path from the first target child node backtracking to the parent node step by step to the starting point of the trajectory planning;

[0206] The first trajectory planning module is used to determine the planned trajectory of the vehicle based on the backtracking path.

[0207] A second path backtracking module, configured to, when the preset end condition is the second preset end condition, backtrack the parent nodes step by step from the searched nodes that meet the conditions along the lane reference line direction to the starting point of the trajectory planning; wherein the node that meets the conditions refers to the node farthest from the vehicle along the lane reference line direction;

[0208] The first trajectory planning module is used to generate a trajectory from the trajectory planning starting point to the node that meets the conditions.

[0209] As a possible implementation of the present application, the vehicle trajectory planning device further includes:

[0210] A set updating unit, configured to update the candidate node set, wherein the updating includes adding the first target child node after calculating the cost value to the candidate node set;

[0211] The parent node selection unit 82 is further configured to select a second target node from the updated candidate node set as a parent node when it is determined that the trajectory search is not completed, wherein the second target node is a node with the smallest cost value in the updated candidate node set;

[0212] The first search unit 83 is further configured to perform a trajectory search according to the vehicle information, the driving information and the preset search decision, generate a second target subnode of the second target node, and add the second target subnode to the candidate node set;

[0213] The completion judgment unit 84 is further configured to determine whether the trajectory search is completed based on the preset end condition when the number of the second target sub-nodes reaches a preset number of sub-nodes;

[0214] The trajectory planning unit 85 is further configured to determine a planned trajectory of the vehicle based on the preset end condition when it is determined that the trajectory search is completed.

[0215] As a possible implementation of the present application, the vehicle trajectory planning device further includes:

[0216] a second search unit, configured to, when the number of the first target child nodes does not reach the preset number of child nodes, continue to perform a trajectory search according to the vehicle information, the driving information and a preset search decision, and generate a new first target child node with the first target node as a parent node;

[0217] The number judgment unit is used to judge whether the number of the first target child nodes with the first target node as the parent node reaches a preset number of child nodes; if it is judged that the preset number of child nodes is reached, determine whether the trajectory search is completed.

[0218] The present application also provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, Figures 1 to 7 Represents the steps of any vehicle trajectory planning method.

[0219] The embodiment of the present application also provides an intelligent terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, Figures 1 to 7 Represents the steps of any vehicle trajectory planning method.

[0220] The embodiment of the present application also provides a computer program product, when the computer program product is run on a terminal device, the terminal device executes the following Figures 1 to 7 Represents the steps of any vehicle trajectory planning method.

[0221] Fig. 9 is a schematic diagram of a smart terminal provided by an embodiment of the present application. Fig. 9 As shown, the intelligent terminal 9 of this embodiment includes: a processor 90, a memory 91, and a computer program 92 stored in the memory 91 and executable on the processor 90. When the processor 90 executes the computer program 92, the steps in the above-mentioned embodiments of the vehicle trajectory planning method are implemented, for example Figure 1 Alternatively, when the processor 90 executes the computer program 92, the functions of each module / unit in the above-mentioned device embodiments are realized, for example, Figure 8 The functions of the units 81 to 85 are shown.

[0222] The computer program 92 may be divided into one or more modules / units, which are stored in the memory 91 and executed by the processor 90 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 92 in the smart terminal 9.

[0223] The processor 90 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0224] The memory 91 may be an internal storage unit of the smart terminal 9, such as a hard disk or memory of the smart terminal 9. The memory 91 may also be an external storage device of the smart terminal 9, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the smart terminal 9. Further, the memory 91 may also include both an internal storage unit of the smart terminal 9 and an external storage device. The memory 91 is used to store the computer program and other programs and data required by the smart terminal. The memory 91 may also be used to temporarily store data that has been output or is to be output.

[0225] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned device can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0226] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0227] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0228] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0229] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0230] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0231] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0232] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0233] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A vehicle trajectory planning method, characterized in that: The method comprises: Acquire vehicle information, driving information, driving environment information, and a trajectory planning starting point, and add the trajectory planning starting point to a candidate node set; Calculating the cost values ​​of the nodes in the candidate node set by using a preset cost function, and determining the node with the smallest cost value in the current candidate node set as the first target node, wherein the preset cost function includes a heuristic cost function determined based on the direction error and the lane reference line; Performing a trajectory search according to the vehicle information, the driving information and a preset search decision to generate a first target child node with the first target node as a parent node; When the number of the first target sub-nodes reaches a preset number of sub-nodes, determining whether the trajectory search is completed based on a preset end condition; When it is determined that the trajectory search is completed, a planned trajectory of the vehicle is determined based on the preset end condition.

2. The method according to claim 1, characterized in that The performing a trajectory search according to the vehicle information, the driving information and a preset search decision to generate a first target child node with the first target node as a parent node includes: Determining a front wheel turning angle range to be searched according to the vehicle information, the driving information and a preset search decision; Based on a preset sampling span and the preset number of sub-nodes, uniformly searching for sub-nodes in the front wheel turning angle range; Verify whether the searched child node is valid; The child node verified to be valid is determined as a first target child node having the first target node as a parent node.

3. The method according to claim 2, characterized in that The vehicle information includes the maximum angle of the front wheel turning angle of the vehicle, the centripetal acceleration limit threshold and the vehicle wheelbase, and the driving information includes the current driving speed; The determining of the searched front wheel turning angle range according to the vehicle information, the driving information and the preset search decision includes: Obtaining the front wheel turning angle corresponding to the first target node; Determining a maximum front wheel turning angle at the current driving speed according to the current driving speed, a centripetal acceleration limit threshold of the vehicle, and the vehicle wheelbase; Determining a maximum front wheel steering angle change value according to the current driving speed; The searched front wheel turning angle range is determined according to the maximum front wheel turning angle, the front wheel turning angle corresponding to the target node, the maximum front wheel turning angle at the current driving speed, and the maximum front wheel turning angle change value.

4. The method according to claim 1, characterized in that The direction error includes a heading angle error and a lateral error; the heuristic cost function includes a gravity function; the cost value includes a gravity value; the gravity value is obtained based on the gravity function; The calculating the cost value of the nodes in the candidate node set by using a preset cost function includes: Obtaining a heading angle error from the node to a reference node and a lateral error from the node to the reference node, wherein the reference node refers to a projection point of the node on the lane reference line; The gravity value is determined according to the heading angle error and the lateral error.

5. The method according to claim 4, characterized in that The calculation formula of the gravitational function is as follows: AttrPotential=ω phi_attr ×(e |phi_diff| -1)+ω lat_attr ×lat_diff Where AttrPotential is the gravity value, phi_diff represents the heading angle error from the node to the reference node, lat_diff represents the lateral error from the node to the reference node, ω phi_attr represents the weight corresponding to phi_diff, ω lat_attr Indicates the weight corresponding to lat_diff.

6. The method according to claim 4, characterized in that The driving environment information further includes obstacles and driving space, the heuristic cost function further includes a repulsion function; the cost value further includes a repulsion value; the repulsion value is obtained based on the repulsion function; The step of calculating the cost value of the nodes in the candidate node set by using a preset cost function further includes: Obtaining a repulsive obstacle set of the node and a resistance of a driving space boundary in the driving space to the node; Determining, according to the obstacle and the driving space, a resistance from the obstacle to the node; Determining the repulsion value according to the obstacle, the repulsion obstacle set, the resistance from the obstacle to the node, and the resistance of the driving space boundary to the node; A heuristic cost value is determined according to the attraction value and the repulsion value.

7. The method according to claim 6, characterized in that The calculation formula of the repulsion function is as follows: RepuPotential=ω ob_repu ×max(ObRepu(o))+ω bo_repu ×BoRepu o∈O Wherein, RepuPotential represents the repulsion value, o represents an obstacle, O represents a repulsion obstacle set, ObRepu(o) represents a first resistance, which is the resistance of obstacle o to the node, and BoRepu represents a second resistance, which is the resistance of the driving space boundary to the node.

8. The method according to claim 7, characterized in that The first resistance ObRepu(o) is determined according to the following formula: Among them, angle o Indicates the minimum angle difference between the heading angle of the node and the boundary of the obstacle o, distance o represents the distance from the obstacle o to the node, k o Indicates the repulsion gain value corresponding to different obstacle types; The second resistance BoRepu is determined according to the following formula: Wherein, distance_bo represents the shortest distance from the node to the boundary of the driving space, and max_distance_bo represents the maximum influence distance of the repulsive force of the driving space boundary.

9. The method according to claim 1, characterized in that The preset cost function also includes a path cost function; the step of calculating the cost value of the nodes in the candidate node set by using the preset cost function includes: Calculating a heuristic cost value using the heuristic cost function; Calculating a path cost value using the path cost function; The cost value of the node is determined according to the sum of the heuristic cost value and the path cost value.

10. The method according to claim 9, characterized in that The calculating the path cost value by using the path cost function includes: Obtaining the front wheel steering angle corresponding to the node, the change value of the front wheel steering angle from the parent node of the node to the node, and the displacement from the parent node of the node to the reference node along the lane reference line, where the reference node refers to the projection point of the node on the lane reference line; The path cost value is determined according to the front wheel turning angle, the change value and the displacement.

11. A vehicle trajectory planning device, characterized in that: The device comprises: A starting point and information acquisition unit, used to acquire vehicle information, driving information, driving environment information and a trajectory planning starting point, and add the trajectory planning starting point to a candidate node set; A parent node selection unit, configured to calculate the cost values ​​of the nodes in the candidate node set by using a preset cost function, and determine the node with the smallest cost value in the current candidate node set as the first target node, wherein the preset cost function includes a heuristic cost function determined based on a direction error and a lane reference line; A first search unit, configured to perform a trajectory search according to the vehicle information, the driving information and a preset search decision, and generate a first target child node with the first target node as a parent node; a completion judgment unit, configured to determine whether the trajectory search is completed based on a preset end condition when the number of the first target sub-nodes reaches a preset number of sub-nodes; The trajectory planning unit is used to determine the planned trajectory of the vehicle based on the preset end condition when it is determined that the trajectory search is completed.

12. An intelligent terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the vehicle trajectory planning method according to any one of claims 1 to 10 are implemented.

13. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the vehicle trajectory planning method according to any one of claims 1 to 10 are implemented.

Images