Path planning method, vehicle and computer readable storage medium

Abstract

Embodiments of the present application provide a path planning method, a vehicle and a computer readable storage medium. The method comprises: obtaining ego vehicle state data, a plurality of candidate reference lines, attribute information corresponding to the plurality of candidate reference lines, an obstacle trajectory and an ego vehicle predicted trajectory; performing analysis and processing on the ego vehicle state data, the plurality of candidate reference lines, the obstacle trajectory and the ego vehicle predicted trajectory to obtain an analysis result; determining an initial recommended lane based on the ego vehicle state data, the ego vehicle predicted trajectory and a safety reference line; determining an initial locking target based on ego vehicle position information and the plurality of candidate reference lines; and determining a target recommended lane based on the plurality of candidate reference lines, the attribute information corresponding to the plurality of candidate reference lines, a reference line type, the ego vehicle position information, the initial recommended lane and the initial locking target. The present application solves the technical problem of low path planning accuracy of the path planning method provided in the related art.

Description

Technical Field

[0001] This application relates to the field of intelligent driving technology, and more specifically, to a path planning method, a vehicle, and a computer-readable storage medium. Background Technology

[0002] In related technologies, when intelligent driving systems plan routes in complex intersection scenarios, they often select reference lines based on preset rules (such as prioritizing the center lane or selecting lines along the direction of the current lane) or based on simple geometric features (such as the principle of minimum lateral distance). However, the above methods in related technologies rely solely on the initial set of reference lines provided by the mapping, and output a single reference line to the planning module after filtering. When facing dynamic obstacles and complex multi-lane road conditions, this can easily lead to inaccurate route planning results and even collision risks.

[0003] There is currently no good solution to the above problems. Summary of the Invention

[0004] This application provides a route planning method, a vehicle, and a computer-readable storage medium to at least solve the technical problem of low route planning accuracy in the route planning methods provided in the related art.

[0005] According to one aspect of the embodiments of this application, a path planning method is provided, comprising: acquiring vehicle state data, multiple candidate reference lines, attribute information corresponding to the multiple candidate reference lines, obstacle trajectories, and a predicted trajectory of the vehicle, wherein the vehicle state data includes vehicle position information, and the candidate reference lines are used to represent the optional driving paths of the vehicle within an intersection area; analyzing and processing the vehicle state data, multiple candidate reference lines, obstacle trajectories, and predicted trajectory of the vehicle to obtain analysis results, wherein the analysis results are used to determine the reference line types corresponding to the multiple candidate reference lines, the reference line types including safe reference lines and high-risk reference lines, and the collision risk value corresponding to the safe reference lines being less than or equal to... For a preset risk threshold, the collision risk value corresponding to a high-risk reference line is greater than the preset risk threshold. Based on the vehicle's status data, predicted trajectory, and safety reference lines, an initial recommended lane is determined. Based on the vehicle's position information and multiple candidate reference lines, an initial locking target is determined, wherein the initial locking target is determined based on multiple lateral projection values ​​between the vehicle's position information and multiple candidate reference lines. The lateral projection values ​​are used to represent the geometric offset of the candidate reference lines relative to the vehicle's position information in the lateral direction. Based on multiple candidate reference lines, their corresponding attribute information, reference line type, vehicle position information, the initial recommended lane, and the initial locking target, a target recommended lane is determined.

[0006] Optionally, the vehicle status data also includes: real-time vehicle speed and vehicle acceleration. The vehicle status data, multiple candidate reference lines, obstacle trajectories, and predicted vehicle trajectory are analyzed and processed to obtain the following analysis results: Based on multiple candidate reference lines and obstacle trajectories, an obstacle trajectory envelope is constructed, whereby the obstacle trajectory envelope represents the spatial coverage area corresponding to the obstacle trajectory within a preset time period; Each candidate reference line and obstacle trajectory envelope among the multiple candidate reference lines is analyzed and processed to obtain a first lateral coordinate value and a second lateral coordinate value of the obstacle trajectory envelope relative to each candidate reference line, whereby the first lateral coordinate value is less than the second lateral coordinate value. The obstacle trajectory and the predicted trajectory of the vehicle are analyzed and processed to obtain a first time and a second time. The first time represents the preset time within which the obstacle travels from its current position to the predicted collision point, and the second time represents the preset time within which the vehicle travels from its current position to the predicted collision point. The first time is determined based on the current speed and acceleration of the obstacle, and the second time is determined based on the real-time speed and acceleration of the vehicle. The predicted collision point is the spatiotemporal conflict point corresponding to the obstacle trajectory and the predicted trajectory of the vehicle. The first lateral coordinate value, the second lateral coordinate value, the first time, and the second time are analyzed and processed to obtain the analysis results.

[0007] Optionally, the first lateral coordinate value, the second lateral coordinate value, the first time, and the second time are analyzed and processed to obtain the following analysis results: in response to the first lateral coordinate value being less than a first preset threshold and the second lateral coordinate value being greater than a second preset threshold, it is determined that the obstacle is in a lateral crossing state, wherein the first preset threshold is less than the second preset threshold; in response to the obstacle being in a lateral crossing state and the difference between the first time and the second time being less than a preset time threshold, the candidate reference line is determined to be a high-risk reference line; in response to the obstacle not being in a lateral crossing state, and / or the difference between the first time and the second time being greater than or equal to a preset time threshold, the candidate reference line is determined to be a safe reference line.

[0008] Optionally, the vehicle status data further includes: the vehicle's driving direction. Based on the vehicle status data, the vehicle's predicted trajectory, and the safety reference line, determining the initial recommended lane includes: performing equidistant sampling processing on the safety reference line according to the vehicle's driving direction to obtain an equidistant sampling sequence, wherein the equidistant sampling sequence includes a first sampling point and a second sampling point, and the distance between the second sampling point and the starting point of the safety reference line is greater than the distance between the first sampling point and the starting point of the safety reference line; in response to the distance between the first sampling point and the second sampling point being less than a preset distance threshold, deleting the second sampling point to obtain a center point sequence corresponding to the safety reference line; and determining the initial recommended lane based on the center point sequence and the vehicle's predicted trajectory.

[0009] Optionally, determining the initial recommended lane based on the center point sequence and the vehicle's predicted trajectory includes: analyzing and processing multiple predicted trajectory points in the center point sequence and the vehicle's predicted trajectory to obtain multiple first center trajectory points and multiple second center trajectory points, wherein the distance between the first center trajectory points and the predicted trajectory points is less than a preset distance threshold, and the first center trajectory points and the second center trajectory points are adjacent trajectory points in the center point sequence; performing linear interpolation on the multiple first center trajectory points and the multiple second center trajectory points to obtain multiple interpolated center points; determining multiple target distances based on the multiple interpolated center points and the multiple predicted trajectory points, wherein the target distance is the lateral and vertical distance between the interpolated center points and the corresponding predicted trajectory points; determining the matching degree between the center point sequence and the vehicle's predicted trajectory based on the average value of the multiple target distances, wherein the matching degree is inversely proportional to the average value; and determining the initial recommended lane based on the matching degree between the center point sequence and the vehicle's predicted trajectory, wherein the matching degree between the center point sequence and the vehicle's predicted trajectory corresponding to the initial recommended lane is less than a preset matching degree threshold.

[0010] Optionally, determining the initial locking target based on the vehicle's location information and multiple candidate reference lines includes: determining a set of lateral projection values ​​based on the vehicle's location information and multiple candidate reference lines, wherein the set of lateral projection values ​​includes multiple lateral projection values; sorting the multiple candidate reference lines in ascending order based on the multiple lateral projection values ​​to obtain a recommended reference line sequence; and determining the initial locking target based on the recommended reference line sequence, wherein the initial locking target is a candidate reference line corresponding to a preset sequence identifier in the recommended reference line sequence, and the preset sequence identifier is a first value.

[0011] Optionally, the attribute information corresponding to multiple candidate reference lines includes: the number of candidate reference lines and the lock point corresponding to the candidate reference lines. Based on multiple candidate reference lines, the attribute information corresponding to multiple candidate reference lines, reference line type, vehicle position information, initial recommended lane, and initial lock target, determining the target recommended lane includes: in response to the number of high-risk reference lines being less than the number of candidate reference lines corresponding to the current frame, and the vehicle's real-time speed being greater than or equal to a preset speed threshold, determining the target recommended lane as the initial recommended lane; or, in response to the number of high-risk reference lines being less than the number of candidate reference lines corresponding to the current frame, and the vehicle's real-time speed being less than a preset speed threshold, determining the target recommended lane as the initial recommended lane. The recommended lane is the lane corresponding to the initial locked target; or, in response to the number of high-risk reference lines being greater than or equal to the number of candidate reference lines in the current frame, and the vehicle's real-time speed being greater than or equal to a preset speed threshold, the target recommended lane is determined as the initial recommended lane; or, in response to the number of high-risk reference lines being greater than or equal to the number of candidate reference lines in the current frame, and the vehicle's real-time speed being less than a preset speed threshold, the target recommended lane is determined based on the locking point corresponding to the candidate reference line and the vehicle's position information, wherein the locking point corresponding to the candidate reference line is the starting point of the first non-virtual segment among the candidate reference lines, and the locking point is determined based on the boundary attributes corresponding to the candidate reference line.

[0012] Optionally, determining the target recommended lane based on the locking point corresponding to the candidate reference line and the vehicle's position information includes: establishing a local coordinate system based on the position information and orientation information corresponding to the locking point; determining the target recommended lane based on the vehicle's position information and the lateral deviation value of the interpolation center point in the local coordinate system, wherein the lateral deviation value corresponding to the target recommended lane is less than a preset deviation threshold.

[0013] According to another aspect of the embodiments of this application, a path planning device is also provided, comprising: an acquisition module, configured to acquire vehicle state data, multiple candidate reference lines, attribute information corresponding to the multiple candidate reference lines, obstacle trajectories, and a predicted trajectory of the vehicle, wherein the vehicle state data includes vehicle position information, and the candidate reference lines are used to represent the optional driving paths of the vehicle within an intersection area; and an analysis module, configured to analyze and process the vehicle state data, multiple candidate reference lines, obstacle trajectories, and the predicted trajectory of the vehicle to obtain analysis results, wherein the analysis results are used to determine the reference line types corresponding to the multiple candidate reference lines, the reference line types including safe reference lines and high-risk reference lines, and the collision risk value corresponding to the safe reference lines is less than or equal to a preset risk threshold. The system consists of three modules: a high-risk reference line and a safety reference line. The first module determines the initial recommended lane based on the vehicle's status data, predicted trajectory, and safety reference line. The second module determines the initial locking target based on the vehicle's position information and multiple candidate reference lines. The initial locking target is determined based on multiple lateral projection values ​​between the vehicle's position information and the candidate reference lines, where the lateral projection values ​​represent the geometric offset of the candidate reference lines relative to the vehicle's position information in the lateral direction. The third module determines the target recommended lane based on multiple candidate reference lines, their corresponding attribute information, reference line type, vehicle position information, the initial recommended lane, and the initial locking target.

[0014] Optionally, the vehicle status data also includes: the vehicle's real-time speed and acceleration. The analysis module is further used to: construct an obstacle trajectory envelope based on multiple candidate reference lines and obstacle trajectories, wherein the obstacle trajectory envelope represents the spatial coverage area corresponding to the obstacle trajectory within a preset time; analyze and process each candidate reference line and obstacle trajectory envelope among the multiple candidate reference lines to obtain a first lateral coordinate value and a second lateral coordinate value of the obstacle trajectory envelope relative to each candidate reference line, wherein the first lateral coordinate value is less than the second lateral coordinate value; analyze and process the obstacle trajectory and the vehicle's predicted trajectory to obtain a first time and a second time, wherein the first time represents the preset time for the obstacle to travel from its current position to the predicted collision point within a preset time, and the second time represents the preset time for the vehicle to travel from its current position to the predicted collision point within a preset time. The first time is determined based on the obstacle's current speed and acceleration, and the second time is determined based on the vehicle's real-time speed and acceleration. The predicted collision point is the spatiotemporal conflict point corresponding to the obstacle trajectory and the vehicle's predicted trajectory; analyze and process the first lateral coordinate value, the second lateral coordinate value, the first time, and the second time to obtain the analysis result.

[0015] Optionally, the analysis module is further configured to: determine that the obstacle is in a lateral crossing state in response to a first lateral coordinate value being less than a first preset threshold and a second lateral coordinate value being greater than a second preset threshold, wherein the first preset threshold is less than the second preset threshold; determine the candidate reference line as a high-risk reference line in response to the obstacle being in a lateral crossing state and the difference between a first time and a second time being less than a preset time threshold; and determine the candidate reference line as a safe reference line in response to the obstacle not being in a lateral crossing state, and / or the difference between a first time and a second time being greater than or equal to a preset time threshold.

[0016] Optionally, the vehicle status data also includes: the vehicle's driving direction. The first determining module is further configured to: perform equidistant sampling processing on the safety reference line according to the vehicle's driving direction to obtain an equidistant sampling sequence, wherein the equidistant sampling sequence includes a first sampling point and a second sampling point, and the distance between the second sampling point and the starting point of the safety reference line is greater than the distance between the first sampling point and the starting point of the safety reference line; in response to the distance between the first sampling point and the second sampling point being less than a preset distance threshold, delete the second sampling point to obtain a center point sequence corresponding to the safety reference line; and determine the initial recommended lane based on the center point sequence and the vehicle's predicted trajectory.

[0017] Optionally, the first determining module is further configured to: analyze and process multiple predicted trajectory points in the center point sequence and the vehicle's predicted trajectory to obtain multiple first center trajectory points and multiple second center trajectory points, wherein the distance between the first center trajectory points and the predicted trajectory points is less than a preset distance threshold, and the first center trajectory points and the second center trajectory points are adjacent trajectory points in the center point sequence; perform linear interpolation processing on the multiple first center trajectory points and the multiple second center trajectory points to obtain multiple interpolated center points; determine multiple target distances based on the multiple interpolated center points and the multiple predicted trajectory points, wherein the target distance is the lateral and vertical distance between the interpolated center points and the corresponding predicted trajectory points; determine the matching degree between the center point sequence and the vehicle's predicted trajectory based on the average value of the multiple target distances, wherein the matching degree is inversely proportional to the average value; and determine an initial recommended lane based on the matching degree between the center point sequence and the vehicle's predicted trajectory, wherein the matching degree between the center point sequence corresponding to the initial recommended lane and the vehicle's predicted trajectory is less than a preset matching degree threshold.

[0018] Optionally, the second determining module is further configured to: determine a set of lateral projection values ​​based on the vehicle's position information and multiple candidate reference lines, wherein the set of lateral projection values ​​includes multiple lateral projection values; sort the multiple candidate reference lines in ascending order based on the multiple lateral projection values ​​to obtain a recommended reference line sequence; and determine an initial locking target based on the recommended reference line sequence, wherein the initial locking target is a candidate reference line corresponding to a preset sequence identifier in the recommended reference line sequence, and the preset sequence identifier is a first value.

[0019] Optionally, the attribute information corresponding to multiple candidate reference lines includes: the number of candidate reference lines and the lock point corresponding to the candidate reference line. The third determining module is further configured to: determine the target recommended lane as the initial recommended lane in response to the number of high-risk reference lines being less than the number of candidate reference lines in the current frame, and the real-time speed of the vehicle being greater than or equal to a preset speed threshold; or, determine the target recommended lane as the lane corresponding to the initial locking target in response to the number of high-risk reference lines being less than the number of candidate reference lines in the current frame, and the real-time speed of the vehicle being less than a preset speed threshold; or, determine the target recommended lane as the initial recommended lane in response to the number of high-risk reference lines being greater than or equal to the number of candidate reference lines in the current frame, and the real-time speed of the vehicle being greater than or equal to a preset speed threshold; or, determine the target recommended lane based on the lock point corresponding to the candidate reference line and the vehicle's position information in response to the number of high-risk reference lines being greater than or equal to the number of candidate reference lines in the current frame, and the real-time speed of the vehicle being less than a preset speed threshold, wherein the lock point corresponding to the candidate reference line is the starting point of the first non-virtual segment among the candidate reference lines, and the lock point is determined based on the boundary attributes corresponding to the candidate reference line.

[0020] Optionally, the third determining module is further configured to: establish a local coordinate system based on the location and orientation information corresponding to the locking point; and determine the target recommended lane based on the vehicle's location information and the lateral deviation value of the interpolation center point in the local coordinate system, wherein the lateral deviation value corresponding to the target recommended lane is less than a preset deviation threshold.

[0021] According to another aspect of the embodiments of this application, a vehicle is also provided, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods in various embodiments of this application when it runs.

[0022] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0023] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0024] According to another aspect of the embodiments of this application, a computer program product is also provided, including a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods in various embodiments of this application.

[0025] According to another aspect of the embodiments of this application, a computer program is also provided, which, when executed by a processor, implements the methods of the various embodiments of this application.

[0026] In this embodiment, by acquiring vehicle state data, multiple candidate reference lines, attribute information corresponding to the multiple candidate reference lines, obstacle trajectories, and predicted vehicle trajectories, and analyzing and processing the vehicle state data, multiple candidate reference lines, obstacle trajectories, and predicted vehicle trajectories to obtain analysis results, an initial recommended lane is determined based on the vehicle state data, predicted vehicle trajectories, and safety reference lines. An initial locking target is determined based on the vehicle position information and multiple candidate reference lines. Finally, a target recommended lane is determined based on the multiple candidate reference lines, their corresponding attribute information, reference line types, vehicle position information, the initial recommended lane, and the initial locking target. This achieves the goal of accurately selecting safe and feasible reference lines, dynamically avoiding trajectory conflict risks, intelligently matching the vehicle's future driving intentions, and stably locking the optimal path in complex dynamic intersection scenarios. This improves the accuracy of path planning for autonomous vehicles in intersection areas and solves the technical problem of low path planning accuracy in related technologies. Attached Figure Description

[0027] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0028] Figure 1 This is a flowchart of a path planning method according to one embodiment of this application;

[0029] Figure 2 This is a schematic diagram of a path planning method according to one embodiment of this application;

[0030] Figure 3 This is a schematic diagram of another path planning method according to one embodiment of this application;

[0031] Figure 4 This is a schematic diagram of another path planning method according to one embodiment of this application;

[0032] Figure 5 This is a schematic diagram of another path planning method according to one embodiment of this application;

[0033] Figure 6 This is a schematic diagram of another path planning method according to one embodiment of this application;

[0034] Figure 7This is a structural block diagram of a path planning device according to one embodiment of the present application. Detailed Implementation

[0035] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0036] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0037] In related technologies, when intelligent driving systems plan routes in complex intersection scenarios, they often select reference lines based on preset rules (such as prioritizing the center lane or selecting lines along the direction of the current lane) or based on simple geometric features (such as the principle of minimum lateral distance). However, the above methods in related technologies rely solely on the initial set of reference lines provided by the mapping, and output a single reference line to the planning module after filtering. When facing dynamic obstacles and complex multi-lane road conditions, this can easily lead to inaccurate route planning results and even collision risks.

[0038] Specifically, the obstacle conflict judgment in related technologies does not integrate trajectory trends and time dimensions, but only judges risk based on the current position of the obstacle. It fails to combine the obstacle trajectory with the vehicle's predicted trajectory for dynamic cross-analysis, easily missing potential collisions where "there is currently no positional overlap but the trajectories are about to intersect" (e.g., a vehicle approaching laterally is currently in an adjacent lane, but its trajectory points towards the vehicle's reference line, making it impossible to identify this risk in advance). This may lead to the selection of a reference line that conflicts with the obstacle. Simultaneously, the path planning methods in related technologies often neglect to predict the degree of matching between the vehicle's future motion state (predicted trajectory) and the reference path, resulting in insufficient foresight in reference line selection. Furthermore, related technologies lack a reference line locking mechanism based on scene continuity and lack inter-frame consistency and adaptive adjustment mechanisms. When the vehicle's perception capabilities / environment change, such as changes in the number of lanes at an intersection (e.g., incorrect connection to the oncoming lane, perception changing from 4 lanes to 2 lanes) or the temporary appearance of obstacles, it is prone to frequent reference line switching, leading to unstable vehicle driving posture (e.g., swaying left and right, frequent acceleration fluctuations), reducing ride comfort and driving safety. Furthermore, the path planning methods in related technologies only sort reference lines by lateral distance without considering the topological relationship of the vehicle's current lane and the subsequent lanes (such as the subsequent lanes corresponding to the optimal exit of the vehicle's current lane). This makes it easy to select non-optimal reference lines. At the same time, in multi-lane and complex intersection scenarios, simple reference line selection methods based on preset rules are often unable to cope with various situations, resulting in unsatisfactory path planning results.

[0039] According to an embodiment of this application, a method embodiment for path planning is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0040] This method embodiment can be executed in an electronic device or similar computing device that includes memory and a processor. Taking operation on a computer terminal as an example, the computer terminal may include one or more processors (processors may include, but are not limited to, central processing units (CPUs), graphics processing units (GPUs), digital signal processing (DSP) chips, microcontroller units (MCUs), field-programmable gate arrays (FPGAs), neural network processors (NPUs), tensor processors (TPUs), artificial intelligence (AI) type processors, etc.) and memory for storing data. Optionally, the computer terminal may also include transmission devices, input / output devices, and display devices for communication functions. Those skilled in the art will understand that the above structural description is merely illustrative and does not limit the structure of the computer terminal. For example, the computer terminal may include more or fewer components than described above, or have a different configuration than described above.

[0041] The memory can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the path planning method in this embodiment. The processor executes various functional applications and data processing by running the computer program stored in the memory, thereby implementing the aforementioned path planning method. The memory may include high-speed random access memory (RAM) and non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the mobile terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks (LANs), mobile communication networks, and combinations thereof.

[0042] The transmission device is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the mobile terminal's communication provider. In one example, the transmission device includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0043] Display devices can be, for example, touchscreen liquid crystal displays (LCDs) and touch displays (also referred to as "touchscreens" or "touch displays"). The LCD allows users to interact with the user interface of the mobile terminal. In some embodiments, the mobile terminal has a graphical user interface (GUI), which allows users to interact with the GUI through finger contact and / or gestures on a touch-sensitive surface. Optional human-computer interaction functions include: creating web pages, drawing, word processing, creating electronic documents, playing games, video conferencing, instant messaging, sending and receiving emails, call interfaces, playing digital video, playing digital music, and / or web browsing, etc. Executable instructions for performing the above human-computer interaction functions are configured / stored in one or more processor-executable computer program products or readable storage media.

[0044] This embodiment provides a path planning method. Figure 1 This is a flowchart of a path planning method according to one embodiment of this application, such as... Figure 1 As shown, the process includes the following steps:

[0045] Step S11: Obtain vehicle status data, multiple candidate reference lines, attribute information corresponding to the multiple candidate reference lines, obstacle trajectory, and vehicle predicted trajectory. The vehicle status data includes vehicle position information, and the candidate reference lines are used to represent the vehicle's optional driving paths within the intersection area.

[0046] The aforementioned vehicle state data is used to characterize the vehicle's motion state in the current road environment and can be output by the fusion of the vehicle perception system and the positioning system. Specifically, the aforementioned vehicle state data may include, but is not limited to, the vehicle's global position coordinates, heading angle, real-time speed, acceleration, yaw rate, and the longitudinal distance and lateral offset of the vehicle's center of mass relative to the road reference frame in the Frenet coordinate system at the current moment.

[0047] The aforementioned candidate reference lines refer to the set of discrete geometric curves that represent the vehicle's selectable driving paths, generated jointly by a high-precision map and a vehicle perception system after the vehicle enters the intersection area.

[0048] The attribute information corresponding to the aforementioned candidate reference lines refers to the lane topology attributes associated with each candidate reference line. Specifically, the attribute information corresponding to the aforementioned candidate reference lines may include, but is not limited to, lane ID, lane type (physical lane or virtual lane), whether the lane boundary is a virtual line, lane width, lane curvature change rate, connection relationship between preceding and succeeding lanes, exit direction, whether it is a variable lane, whether it has a conflicting topology with oncoming traffic flow, and the state of the reference line in the current frame.

[0049] The obstacle trajectory mentioned above refers to the temporal motion history and predicted path of obstacles identified and tracked by the vehicle perception system within the intersection area. Specifically, the obstacle trajectory can consist of a series of three-dimensional spatial location points arranged by timestamps. The attribute information corresponding to each point may include, but is not limited to, the obstacle's global position, velocity vector, acceleration vector, heading angle, size parameters, and obstacle type. For example, the obstacle type may include, but is not limited to, static obstacles and dynamic obstacles.

[0050] For example, target detection algorithms can be used to identify dynamic obstacles such as vehicles, pedestrians, and non-motorized vehicles in the intersection area. Combined with multi-frame observation data obtained by the vehicle-mounted perception system, Kalman filtering, particle filtering, or deep learning trackers can be used to perform data association and trajectory prediction on the dynamic obstacles, thereby estimating the time sequence of their position, velocity, acceleration, and heading angle in the Frenet coordinate system or global coordinate system, and thus obtaining the obstacle trajectory.

[0051] The aforementioned predicted trajectory of the autonomous vehicle refers to the movement path that the vehicle may follow within a future time period, derived from the vehicle's current motion state and driving intention model. Specifically, the predicted trajectory can be composed of continuous position points sampled at fixed time intervals along the longitudinal time axis in the Frenet coordinate system.

[0052] For example, multi-step rolling prediction can be performed based on the current vehicle state data combined with a dynamics model or behavior prediction model to obtain the vehicle's predicted trajectory. Specifically, based on the current vehicle state data, a two-degree-of-freedom vehicle dynamics model or kinematics model can be used to extrapolate trajectory points in future time periods at fixed time steps (e.g., 0.1 seconds) to obtain the vehicle's predicted trajectory.

[0053] Step S12: Analyze and process the vehicle status data, multiple candidate reference lines, obstacle trajectory and vehicle predicted trajectory to obtain analysis results. The analysis results are used to determine the reference line type corresponding to the multiple candidate reference lines. The reference line type includes safe reference lines and high-risk reference lines. The collision risk value corresponding to the safe reference line is less than or equal to the preset risk threshold, and the collision risk value corresponding to the high-risk reference line is greater than the preset risk threshold.

[0054] The aforementioned reference line type refers to the collision risk level classification assigned to each candidate reference line based on the interaction between the obstacle trajectory and the vehicle's predicted trajectory in the spatiotemporal dimension.

[0055] The aforementioned safety reference line refers to a candidate reference line that, under the current traffic environment, is determined to have no significant collision threat with potential dynamic obstacles and meets the preset safety margin after cross-analysis of obstacle trajectories and predicted vehicle trajectories in the spatiotemporal dimensions.

[0056] Specifically, if the collision risk value corresponding to the aforementioned safety reference line is less than or equal to the preset risk threshold, it indicates that when driving on the reference line, the vehicle will not have a lateral crossing conflict or a high probability of time collision with surrounding obstacles within the predicted time window, thus having sufficient safety redundancy to support subsequent path planning and control execution.

[0057] The aforementioned high-risk reference line refers to a candidate reference line that, under the current traffic environment, is determined to pose a significant collision threat to potential dynamic obstacles and exceeds the preset safety margin through cross-analysis of obstacle trajectories and predicted vehicle trajectories in the spatiotemporal dimensions.

[0058] Step S13: Determine the initial recommended lane based on the vehicle's status data, predicted trajectory, and safety reference line;

[0059] The aforementioned initial recommended lane refers to the candidate safety reference line that has the highest similarity to the predicted trajectory of the vehicle in terms of spatial path and the smallest lateral deviation.

[0060] Step S14: Based on the vehicle position information and multiple candidate reference lines, determine the initial locking target. The initial locking target is determined based on multiple lateral projection values ​​between the vehicle position information and multiple candidate reference lines. The lateral projection values ​​are used to represent the geometric offset of the candidate reference lines relative to the vehicle position information in the lateral direction.

[0061] The initial target lock mentioned above refers to the baseline reference line initially selected by the vehicle in an intersection scenario for subsequent path tracking and stability control. Specifically, the initial target lock can be determined based on the vehicle's position information and multiple lateral projection values ​​between multiple candidate reference lines.

[0062] For example, the initial locking target can be the candidate reference line corresponding to the minimum lateral projection value to ensure that the vehicle maintains a centered and stable driving posture when entering the intersection.

[0063] The aforementioned lateral projection values ​​refer to the vertical distance in the lateral direction of the centerline of each candidate reference line relative to the longitudinal axis of the vehicle, within the Frenet coordinate system. Specifically, these lateral projection values ​​represent the geometrical relative position of the candidate reference line and the vehicle's current position in lateral space, and their magnitude directly characterizes the degree of deviation between the reference line and the vehicle's current lane.

[0064] Step S15: Based on multiple candidate reference lines, the attribute information corresponding to the multiple candidate reference lines, the reference line type, the vehicle position information, the initial recommended lane, and the initial locked target, determine the target recommended lane.

[0065] The aforementioned target recommended lane refers to the lane that is ultimately selected as the unique target driving path output for the current path planning, based on a comprehensive analysis of the distribution of safe reference lines and high-risk reference lines, the topological attributes of candidate reference lines, the lateral offset relationship of the vehicle's current position, the initial recommended lane, and the initial locked target.

[0066] Figure 2 This is a schematic diagram of a path planning method according to one embodiment of this application, such as... Figure 2 As shown, during path planning, map data, vehicle status data, obstacle status data, obstacle trajectories, and the vehicle's predicted trajectory can be used as input data. Subsequently, initial screening and sorting of reference lines—geometric optimization—and obstacle risk assessment—predictive optimization are performed, followed by reference line locking and updating, ultimately outputting the optimal reference line information.

[0067] Specifically, the result of the initial screening and sorting of the reference lines and the geometric optimization is the initial target lock, and the result of the obstacle risk assessment and the predictive optimization is the initial recommended lane.

[0068] Based on steps S11 to S15 above, by acquiring vehicle state data, multiple candidate reference lines, attribute information corresponding to the multiple candidate reference lines, obstacle trajectories, and predicted vehicle trajectories, and analyzing and processing the vehicle state data, multiple candidate reference lines, obstacle trajectories, and predicted vehicle trajectories to obtain analysis results, and then determining the initial recommended lane based on the vehicle state data, predicted vehicle trajectories, and safety reference lines, and determining the initial locking target based on the vehicle position information and multiple candidate reference lines, and finally determining the target recommended lane based on the multiple candidate reference lines, their corresponding attribute information, reference line types, vehicle position information, initial recommended lane, and initial locking target, the goal of accurately selecting safe and feasible reference lines, dynamically avoiding trajectory conflict risks, intelligently matching the vehicle's future driving intentions, and stably locking the optimal path in complex dynamic intersection scenarios is achieved. This improves the accuracy of path planning for autonomous vehicles in intersection areas and solves the technical problem of low path planning accuracy in related technologies.

[0069] Optionally, the vehicle status data also includes: the vehicle's real-time speed and acceleration. In step S12, the vehicle status data, multiple candidate reference lines, obstacle trajectories, and the predicted vehicle trajectory are analyzed and processed to obtain the following analysis results:

[0070] Step S121: Based on multiple candidate reference lines and obstacle trajectories, construct an obstacle trajectory envelope, wherein the obstacle trajectory envelope is used to represent the spatial coverage area corresponding to the obstacle trajectory within a preset time period;

[0071] Step S122: Analyze and process each candidate reference line and obstacle trajectory envelope among multiple candidate reference lines to obtain the first lateral coordinate value and the second lateral coordinate value of the obstacle trajectory envelope relative to each candidate reference line, wherein the first lateral coordinate value is less than the second lateral coordinate value.

[0072] Step S123: Analyze and process the obstacle trajectory and the predicted trajectory of the vehicle to obtain a first time and a second time. The first time represents the preset time within a preset period for the obstacle to travel from its current position to the predicted collision point, and the second time represents the preset time within a preset period for the vehicle to travel from its current position to the predicted collision point. The first time is determined based on the current speed and acceleration of the obstacle, and the second time is determined based on the real-time speed and acceleration of the vehicle. The predicted collision point is the spatiotemporal conflict point corresponding to the obstacle trajectory and the predicted trajectory of the vehicle.

[0073] Step S124: Analyze and process the first horizontal coordinate value, the second horizontal coordinate value, the first time, and the second time to obtain the analysis results.

[0074] The aforementioned obstacle trajectory envelope refers to the spatiotemporal coverage area derived from the kinematic model within a preset time window, based on the current velocity, acceleration, and direction of motion of the dynamic obstacle. It is used to characterize all spatial locations that the obstacle may occupy in the future.

[0075] Specifically, the aforementioned obstacle trajectory envelope can be a quadrilateral or hexagonal spatial envelope formed by expanding the obstacle's predicted trajectory in the horizontal and vertical dimensions in combination with its geometric dimensions. Its boundary is jointly enclosed by the minimum and maximum horizontal and vertical coordinates of the obstacle at each moment between the start and end points of the predicted trajectory.

[0076] The aforementioned first lateral coordinate value refers to the minimum boundary value covered by the obstacle trajectory envelope in the Frenet coordinate system along the lateral direction (perpendicular to the tangent of the reference line), that is, the minimum lateral coordinate value among all points within the envelope. Specifically, the aforementioned first lateral coordinate value represents the position of the obstacle closest to the left edge of the reference line in the lateral direction, and is used to determine whether the obstacle approaches or crosses the reference line from the left.

[0077] The aforementioned second lateral coordinate value refers to the maximum boundary value covered by the obstacle trajectory envelope in the Frenet coordinate system along the lateral direction, that is, the maximum value among the lateral coordinates of all points within the envelope. Specifically, the aforementioned second lateral coordinate value represents the position of the obstacle closest to the right edge of the reference line in the lateral direction, and is used to determine whether the obstacle approaches or crosses the reference line from the right.

[0078] The aforementioned "first time" refers to the time required for an obstacle, based on its current speed, acceleration, and trajectory, to reach the point in time and space where it potentially clashes with the vehicle's predicted trajectory.

[0079] The aforementioned second time refers to the time required for the vehicle, based on its current speed, acceleration, and predicted trajectory, from the current moment until a potential conflict occurs between it and the obstacle's trajectory in space and time.

[0080] For example, the aforementioned first time can be calculated using the following formula:

[0081]

[0082] in, To be the first, The current speed corresponding to the obstacle. The acceleration corresponding to the obstacle. The longitudinal distance from the obstacle to the predicted collision point ( ).

[0083] For example, the aforementioned second time can be calculated using the following formula:

[0084]

[0085] in, For the second time, For the real-time speed of the vehicle, For the vehicle's acceleration, The longitudinal distance from the vehicle to the predicted collision point ( ).

[0086] Based on steps S121 to S124 above, an obstacle trajectory envelope is constructed based on multiple candidate reference lines and obstacle trajectories. Each candidate reference line and obstacle trajectory envelope is analyzed and processed to obtain the first lateral coordinate value and the second lateral coordinate value of the obstacle trajectory envelope relative to each candidate reference line. Then, the obstacle trajectory and the predicted trajectory of the vehicle are analyzed and processed to obtain the first time and the second time. Finally, the first lateral coordinate value, the second lateral coordinate value, the first time and the second time are analyzed and processed to obtain the analysis results, which can improve the accuracy of identifying potential collision risks.

[0087] Optionally, in step S124, the first horizontal coordinate value, the second horizontal coordinate value, the first time, and the second time are analyzed and processed to obtain the analysis results, including:

[0088] Step S1241: In response to the first lateral coordinate value being less than the first preset threshold and the second lateral coordinate value being greater than the second preset threshold, it is determined that the obstacle is in a lateral crossing state, wherein the first preset threshold is less than the second preset threshold.

[0089] Step S1242: In response to the obstacle being in a lateral crossing state and the difference between the first time and the second time being less than a preset time threshold, the candidate reference line is determined to be a high-risk reference line.

[0090] Step S1243: In response to the obstacle not being in a lateral crossing state, and / or the difference between the first time and the second time being greater than or equal to a preset time threshold, the candidate reference line is determined as a safe reference line.

[0091] For example, the difference between the first time and the second time can be expressed as:

[0092]

[0093] in, This represents the difference between the first and second time points.

[0094] For example, the first horizontal coordinate value mentioned above can be expressed as The aforementioned second horizontal coordinate value can be expressed as The first preset threshold can be -1.0m, the second preset threshold can be 1.0m, and the preset time threshold can be 2.0s. When <-1.0m, and When the distance is greater than 1.0m, it can be determined that the obstacle is in a lateral crossing state. Furthermore, the obstacle is in a lateral crossing state, and... If the time is less than 2.0 seconds, the candidate reference line can be identified as a high-risk reference line. Conversely, if the obstacle is not in a lateral crossing state, and / or, If the time is ≥2.0s, the candidate reference line can be determined as a safe reference line.

[0095] Figure 3 This is a schematic diagram of another path planning method according to one embodiment of this application, such as... Figure 3 As shown, during path planning, the conflict records are first cleared, i.e., the reference line conflict marker list (recording the reference line risk status and ID) and the risk reference line counter are initialized. Then, relevant obstacles are filtered (target type, non-virtual and non-static, located in the region of interest, speed) to form an obstacle list, which is then scanned line by line. When a relevant obstacle is in the region of interest, a lateral crossing is determined based on the spatial trajectory intersection; if a lateral crossing occurs, a temporal conflict check is performed. Subsequently, the conflict list is updated when a temporal conflict occurs. Conversely, when a relevant obstacle is not in the region of interest or has not laterally crossed, the reference line scanning is repeated.

[0096] Based on steps S1241 to S1243 above, if the first lateral coordinate value is less than the first preset threshold and the second lateral coordinate value is greater than the second preset threshold, it is determined that the obstacle is in a lateral crossing state. If the difference between the first and second times is less than a preset time threshold, the candidate reference line is determined to be a high-risk reference line. This accurately identifies potential collision risks that do not currently overlap spatially but will eventually conflict with the vehicle's predicted trajectory. This avoids "hidden collision threats" missed due to relying solely on static position judgment, improving the accuracy and safety of path decision-making at intersections and preventing vehicles from mistakenly entering lanes soon to be occupied by dynamic obstacles. If the obstacle is not in a lateral crossing state, and / or the difference between the first and second times is greater than or equal to a preset time threshold, the candidate reference line is determined to be a safe reference line. This eliminates interfering targets that are only spatially adjacent but have no actual collision trend, avoiding excessive filtering of reference lines due to misjudgment and reducing driving selection space. This ensures that enough feasible paths are retained under safe conditions to maintain the continuity and flexibility of path planning.

[0097] Optionally, the vehicle status data also includes: the vehicle's driving direction. In step S13, based on the vehicle status data, the predicted vehicle trajectory, and the safety reference line, the initial recommended lane is determined, including:

[0098] Step S131: Perform equidistant sampling processing on the safety reference line according to the vehicle's driving direction to obtain an equidistant sampling sequence. The equidistant sampling sequence includes a first sampling point and a second sampling point. The distance between the second sampling point and the starting point of the safety reference line is greater than the distance between the first sampling point and the starting point of the safety reference line.

[0099] Step S132: In response to the distance between the first sampling point and the second sampling point being less than a preset distance threshold, the second sampling point is deleted to obtain the center point sequence corresponding to the safety reference line.

[0100] Step S133: Determine the initial recommended lane based on the center point sequence and the predicted trajectory of the vehicle.

[0101] For example, the predicted trajectory of the vehicle over 5 seconds (50 points at 0.1-second intervals) can be extracted, and the trajectory points can be transformed from the Earth coordinate system to the vehicle coordinate system (Forward-Lateral-Up Coordinate System, FLU). The transformation formula is as follows:

[0102]

[0103]

[0104] in,( () represents the current location of the vehicle. The heading angle of the vehicle (in radians). This represents the horizontal coordinate of the current trajectory point in the global coordinate system. This represents the vertical coordinate of the current trajectory point in the global coordinate system. Let be the lateral coordinate of the vehicle's body center in the global coordinate system. This represents the longitudinal coordinate of the vehicle's body center in the global coordinate system.

[0105] Furthermore, the safety reference lines can be traversed, and each safety reference line can be sampled at equal intervals along the center line at 1m arc lengths to obtain an equidistant sampling sequence.

[0106] For example, the first and second sampling points mentioned above can be adjacent points in an equidistant sampling sequence. The distance between the first and second sampling points can be the arc length distance between them. When the arc length distance is less than a preset distance threshold, it indicates that there is redundant sampling or local curvature flatness at the reference line center point in this area. In this case, the second sampling point can be deleted, and only the first sampling point can be retained as the effective center point, thereby constructing a sparse but geometrically complete and non-redundant center point sequence. Furthermore, the initial recommended lane can be determined based on the center point sequence and the predicted trajectory of the vehicle.

[0107] Based on steps S131 to S133 above, the safety reference line is sampled at equal intervals according to the vehicle's driving direction to obtain an equal interval sampling sequence. When the distance between the first sampling point and the second sampling point is less than a preset distance threshold, the second sampling point is deleted to obtain the center point sequence corresponding to the safety reference line. Then, based on the center point sequence and the vehicle's predicted trajectory, the initial recommended lane is determined. This can avoid computational redundancy and interpolation deviation caused by dense sampling points, and ensure that the center point sequence truly reflects the lane's geometric features, thereby more accurately evaluating the lateral matching degree between the vehicle's predicted trajectory and the candidate reference line.

[0108] Optionally, in step S133, determining the initial recommended lane based on the center point sequence and the vehicle's predicted trajectory includes:

[0109] Step S1331: Analyze and process multiple predicted trajectory points in the center point sequence and the predicted trajectory of the vehicle to obtain multiple first center trajectory points and multiple second center trajectory points. The distance between the first center trajectory point and the predicted trajectory point is less than a preset distance threshold. The first center trajectory point and the second center trajectory point are adjacent trajectory points in the center point sequence.

[0110] Step S1332: Perform linear interpolation on multiple first center trajectory points and multiple second center trajectory points to obtain multiple interpolated center points;

[0111] Step S1333: Based on multiple interpolation center points and multiple predicted trajectory points, determine multiple target distances, where the target distance is the horizontal and vertical distance between the interpolation center point and the corresponding predicted trajectory point;

[0112] Step S1334: Based on the average value corresponding to multiple target distances, determine the matching degree between the center point sequence and the predicted trajectory of the vehicle, wherein the matching degree is inversely proportional to the average value;

[0113] Step S1335: Determine the initial recommended lane based on the matching degree between the center point sequence and the predicted trajectory of the vehicle, wherein the matching degree between the center point sequence corresponding to the initial recommended lane and the predicted trajectory of the vehicle is less than a preset matching degree threshold.

[0114] The aforementioned multiple first center trajectory points and multiple second center trajectory points refer to pairs of adjacent center points in the center point sequence, determined by point-by-point comparison based on the spatial distribution of the vehicle's predicted trajectory points. In each pair, the first center trajectory point is the discrete sampling point in the center point sequence that is closest to and located in front of or near the vehicle's predicted trajectory point along the longitudinal direction, while the second center trajectory point is the next center point that follows it. The two are continuously adjacent on the reference line centerline, and their positions constitute a small line segment interval containing the vehicle's predicted trajectory point.

[0115] For example, for each predicted trajectory point, the two adjacent points with the closest x-coordinates on the center point sequence can be searched as the first center trajectory point and the second center trajectory point.

[0116] The aforementioned multiple interpolation center points refer to the centerline coordinates of the predicted trajectory point, calculated using a linear interpolation method based on multiple first center trajectory points and multiple second center trajectory points, and located at the lateral projection position of the predicted trajectory point.

[0117] The aforementioned target distances refer to the horizontal and vertical distances between each interpolation center point and its corresponding predicted trajectory point in the Frenet coordinate system. In other words, in a local coordinate system with the tangent direction of the reference line as the vertical axis and the normal direction as the horizontal axis, it is the shortest horizontal offset obtained by drawing a perpendicular line from the predicted trajectory point to the center line where the interpolation center point is located.

[0118] For example, the average value corresponding to the above multiple target distances can be calculated using the following formula:

[0119]

[0120] in, This is the average value corresponding to the distances to multiple targets. For the target distance, N This represents the total number of trajectory points in the predicted trajectory of the vehicle. For example, when N=50, it means 50 predicted trajectory points sampled at 0.1-second intervals within the prediction time window (5 seconds).

[0121] Specifically, The degree of matching between the center point sequence and the predicted trajectory of the vehicle is inversely proportional to that, that is... The smaller the value, the higher the matching degree between the center point sequence and the predicted trajectory of the vehicle. Therefore, multiple values ​​can be selected. The minimum value in the value range is used as the safety reference line for the initial recommended lane.

[0122] Based on steps S1331 to S1335 above, the center point sequence and multiple predicted trajectory points in the vehicle's predicted trajectory are analyzed and processed to obtain multiple first center trajectory points and multiple second center trajectory points. Linear interpolation is then performed on the multiple first center trajectory points and multiple second center trajectory points to obtain multiple interpolated center points. Based on the multiple interpolated center points and multiple predicted trajectory points, multiple target distances are determined. Based on the average value of the multiple target distances, the matching degree between the center point sequence and the vehicle's predicted trajectory is determined. Finally, based on the matching degree between the center point sequence and the vehicle's predicted trajectory, the initial recommended lane is determined, which can filter out the lane that best matches the actual driving trend of the vehicle.

[0123] Optionally, in step S14, based on the vehicle's position information and multiple candidate reference lines, the initial target to be locked is determined, including:

[0124] Step S141: Based on the vehicle's position information and multiple candidate reference lines, determine the set of lateral projection values, wherein the set of lateral projection values ​​includes multiple lateral projection values;

[0125] Step S142: Sort multiple candidate reference lines in ascending order based on multiple lateral projection values ​​to obtain a recommended reference line sequence;

[0126] Step S143: Based on the recommended reference line sequence, determine the initial locking target, wherein the initial locking target is the candidate reference line corresponding to the preset sequence identifier in the recommended reference line sequence, and the preset sequence identifier is the first value.

[0127] The aforementioned set of lateral projection values ​​refers to the sequence of vertical distances from the center point of each candidate reference line to the vehicle's centerline along the lateral direction, within the Frenet coordinate system. Specifically, each value in the aforementioned set of lateral projection values ​​corresponds to a candidate reference line, reflecting the geometric offset of that reference line in lateral space relative to the vehicle's current position. The sign indicates the direction, and the absolute value indicates the distance.

[0128] For example, the lateral projection value can be calculated by projection based on the spatial relationship between the vehicle coordinate system and the centerline of the reference line in the current frame. When the vehicle is at an intersection, multiple candidate reference lines can be sorted according to their lateral features. Specifically, multiple lateral projection values ​​can be extracted from the labels of multiple candidate reference lines to establish a "lateral projection value - lane ID" mapping.

[0129] The aforementioned recommended reference line sequence refers to an ordered set of candidate reference lines formed by arranging the absolute values ​​of each value in the set of horizontal projection values ​​in ascending order. The position of each reference line in the sequence is determined by the absolute value of its corresponding horizontal projection value. The smaller the value, the higher the position of the reference line. The first item in the sequence is the reference line with the smallest horizontal offset, and the subsequent items are reference lines with increasing offsets.

[0130] For example, the first value mentioned above can be determined based on the first sequence identifier corresponding to the recommended reference line sequence. When the first sequence identifier corresponding to the recommended reference line sequence is 1, the first value is 1, and the initial target is the candidate reference line corresponding to the sequence identifier 1 in the recommended reference line sequence.

[0131] It should be noted that the values ​​in the embodiments of this application are merely examples and do not constitute a fixed setting for this application.

[0132] Figure 4 This is a schematic diagram of another path planning method according to one embodiment of this application, such as... Figure 4 As shown, when the vehicle's current lane ID is valid, multiple candidate reference lines in the current road environment can be identified, and the lateral offset between the vehicle and these candidate reference lines can be extracted. Furthermore, the lateral offsets can be sorted in ascending order, and a recommended reference line list can be output when the number of candidate reference lines is greater than one. Conversely, when the number of candidate reference lines is less than or equal to one, or when the vehicle's current lane ID is invalid, the vehicle exits the non-intersection mode and resets the reference line locking status.

[0133] Specifically, a valid current lane ID for a vehicle means that the vehicle's location can be successfully matched with a legal lane entity defined in the high-precision map. That is, the lane ID extracted from the reference line information is not empty or invalid (such as -1 or null), and the ID has a corresponding valid lane attribute in the map topology library.

[0134] Specifically, when the number of multiple candidate reference lines is less than or equal to one, or the vehicle's current lane ID is invalid, it can be determined that the vehicle has left the intersection area with a clear lane structure and entered a non-intersection scenario (such as a straight road, after a ramp exit, or an area with missing map data). Since intersection decisions rely on complex judgments such as the lateral sorting of multiple selectable reference lines, obstacle interaction analysis, and topology matching, if there are insufficient candidate reference lines or the lane ID is invalid, a valid candidate set cannot be constructed. Continuing to use the intersection strategy may lead to incorrect virtual line selection, incorrect locking, or no reference lines available, thereby causing path planning failure or control oscillations. Therefore, to ensure vehicle driving safety, it is necessary to actively exit the intersection decision mode, reset all state variables related to intersection locking (including but not limited to locking status flags, locking point coordinates, locking reference line IDs, and historical jump records), and re-initialize the path planning decision process only after re-entering a valid intersection scenario and acquiring sufficient candidate reference lines, thereby avoiding invalid or erroneous decision outputs that affect vehicle stability and driving safety.

[0135] Based on steps S141 to S143 above, a set of lateral projection values ​​is determined based on the vehicle's position information and multiple candidate reference lines. The candidate reference lines are then sorted in ascending order based on the multiple lateral projection values ​​to obtain a recommended reference line sequence. Based on the recommended reference line sequence, an initial locking target is determined. This allows the selection of the reference line with the smallest geometric offset and closest to the vehicle's current centering driving trend as the initial locking target. This reduces the vehicle's steering amplitude and lateral acceleration fluctuations at the beginning of the intersection, improving the smoothness and stability of path selection.

[0136] Figure 5 This is a schematic diagram of another path planning method according to one embodiment of this application, such as... Figure 5 As shown, the predicted trajectory of the vehicle is obtained, and after coordinate transformation of the trajectory points, candidate lanes are scanned one by one. Then, the lane center points are extracted, deduplication and smoothing are performed, and the average lateral distance between the vehicle and the candidate lanes is calculated, thus recording the lane-distance pairs. Further, when no candidate lane exists, the lane with the smallest distance is selected as the target recommended lane. Conversely, when candidate lanes still exist, the scan of each candidate lane and subsequent processes are repeated.

[0137] Optionally, the attribute information corresponding to the multiple candidate reference lines includes: the number of candidate reference lines and the locking point corresponding to the candidate reference lines. In step S15, based on the multiple candidate reference lines, the attribute information corresponding to the multiple candidate reference lines, the reference line type, the vehicle position information, the initial recommended lane, and the initial locking target, the target recommended lane is determined as follows:

[0138] Step S151: In response to the fact that the number of high-risk reference lines is less than the number of candidate reference lines in the current frame, and the real-time speed of the vehicle is greater than or equal to a preset speed threshold, the target recommended lane is determined as the initial recommended lane; or,

[0139] Step S152: In response to the fact that the number of high-risk reference lines is less than the number of candidate reference lines in the current frame, and the real-time speed of the vehicle is less than a preset speed threshold, the target recommended lane is determined as the lane corresponding to the initially locked target; or,

[0140] Step S153: In response to the number of high-risk reference lines being greater than or equal to the number of candidate reference lines in the current frame, and the vehicle's real-time speed being greater than or equal to a preset speed threshold, the target recommended lane is determined as the initial recommended lane; or,

[0141] Step S154: In response to the fact that the number of high-risk reference lines is greater than or equal to the number of candidate reference lines in the current frame, and the real-time speed of the vehicle is less than a preset speed threshold, the target recommended lane is determined based on the locking point corresponding to the candidate reference line and the vehicle's position information. The locking point corresponding to the candidate reference line is the starting point of the first non-virtual segment among the candidate reference lines, and the locking point is determined based on the boundary attributes corresponding to the candidate reference line.

[0142] The aforementioned lock point refers to the starting position of the first non-virtual boundary segment in the candidate reference line. Specifically, the lock point can be determined by the left and right boundary attributes of the candidate reference line. When both the left and right boundaries of the candidate reference line are marked as virtual, this segment is not used as a valid locking basis. The process continues to scan segment by segment along the longitudinal direction of the candidate reference line until a segment appears for the first time where at least one of the left and right boundaries is marked as a physical lane line. The starting point of this segment is then defined as the lock point.

[0143] For example, a locking status flag can be set for multiple candidate reference lines. Specifically, the locking status flag can be defined as three states: unlocked (locking flag is false, locking point, locking reference line ID, and lateral projection value are all default values), locked (locking flag is true, valid locking parameters are recorded), and pending update (locking flag is true, but update trigger conditions are met). The unlocked state corresponds to a locking flag of false. At this time, the locking point, locking reference line ID, and lateral projection value are all kept at the default initial values, indicating that a stable reference line reference has not yet been established. The locked state corresponds to a locking flag of true, and a valid locking reference line ID, the corresponding locking point spatial coordinates, and the lateral projection value have been successfully recorded, indicating that the current reference line is in a stable tracking state. The pending update state is a transitional locking state. Its locking flag is still true, but the environmental change or safety risk trigger conditions have been met, indicating that the current locked reference line no longer meets the optimal or safety requirements and needs to be updated in the next decision cycle. However, the switch has not yet been completed to ensure the continuity of decision logic and inter-frame consistency.

[0144] For example, the aforementioned locking reference line can be an optimal and safe reference line selected by the autonomous vehicle when passing through an intersection, based on the current driving state, obstacle risk assessment, vehicle predicted trajectory matching, and lane topology. This reference line serves as a long-term driving path benchmark, and remains unchanged unless the environment changes significantly or safety risks are triggered. This avoids frequent path switching due to perceived vibrations, brief disturbances, or minor deviations, thereby ensuring the stability, smoothness, and ride comfort of the vehicle. A smooth switch is only performed when preset environmental change conditions (such as a sudden change in the number of lanes), escalation of safety risks (such as the original locking line being deemed high-risk), or a significant deviation from the predicted intent are met, provided that the lateral jump threshold is satisfied. This achieves a balance between dynamic adaptation of decision-making and continuity of behavior.

[0145] For example, in the unlocked state, the lane corresponding to the initially locked target can be selected as the target recommended lane, and its state can be updated to locked. At the same time, its corresponding attribute information is recorded, and the lock point is calculated. Furthermore, the positioning information corresponding to the calculated lock point can be used as the spatial reference for subsequent determination of whether to switch reference lines (i.e., lane switching).

[0146] For example, in the locked state, environmental changes can be continuously monitored, and reference line switching can be performed when the number of reference lines acquired in the current frame changes compared to the previous frame, or when the currently locked reference line is a high-risk reference line and a safe reference line exists at the same time.

[0147] For example, based on the number of high-risk reference lines, a safe mode (the number of high-risk reference lines is less than the number of candidate reference lines in the current frame) or a downgrade mode (the number of high-risk reference lines is greater than or equal to the number of candidate reference lines in the current frame) can be selected. Then, the target recommended lane can be determined according to the selected mode, and a new locking point can be found on the reference line corresponding to the target recommended lane.

[0148] For example, when the number of high-risk reference lines is less than the number of candidate reference lines in the current frame, it is a safe mode where only safe reference lines are considered. In this safe mode, when the vehicle's real-time speed is greater than or equal to a preset speed threshold, the initially recommended lane is selected as the target recommended lane. Conversely, in this safe mode, when the vehicle's real-time speed is less than or equal to the preset speed threshold, the lane corresponding to the initially locked target is selected as the target recommended lane.

[0149] For example, when the number of high-risk reference lines is greater than or equal to the number of candidate reference lines in the current frame, a downgrade mode is adopted, considering all candidate reference lines. In this downgrade mode, when the vehicle's real-time speed is greater than or equal to a preset speed threshold, the initial recommended lane can be selected as the target recommended lane. Conversely, in this downgrade mode, when the vehicle's real-time speed is less than or equal to the preset speed threshold, the target recommended lane needs to be determined based on the locking points corresponding to the candidate reference lines and the vehicle's position information.

[0150] Figure 6 This is a schematic diagram of another path planning method according to one embodiment of this application, such as... Figure 6 As shown, when multiple reference lines are unlocked and a condition trigger / environmental change occurs, the reference line switching process is initiated. If the current scenario is not a full-collision scenario, the initially recommended lane is prioritized as the target recommended lane from among the safe reference lines. Conversely, if the current scenario is a full-collision scenario, a degraded mode is entered, ignoring the collision status, and the lane corresponding to the initially locked target is selected as the target recommended lane. When multiple candidate reference lines are not unlocked, the initial locking strategy is adopted. When no condition trigger / environmental change occurs, the locked state is maintained.

[0151] Specifically, the aforementioned full-conflict scenario refers to a scenario where the number of high-risk reference lines is greater than or equal to the number of candidate reference lines in the current frame.

[0152] Based on steps S151 to S154 above, by comprehensively considering the proportion of high-risk reference lines, the real-time speed status of the vehicle, and the spatial semantic consistency of the locking point, intelligent hierarchical decision-making for reference line selection can be achieved in complex dynamic intersection environments, thereby improving the accuracy of path planning and thus enhancing the driving experience.

[0153] Optionally, in step S154, based on the locking point corresponding to the candidate reference line and the vehicle's position information, the target recommended lane is determined as follows:

[0154] Step S1541: Establish a local coordinate system based on the location and orientation information corresponding to the locking point;

[0155] Step S1542: Based on the vehicle's position information and the lateral deviation value of the interpolation center point in the local coordinate system, determine the target recommended lane, wherein the lateral deviation value corresponding to the target recommended lane is less than a preset deviation threshold.

[0156] For example, the origin can be taken as the locking point, and the direction of the locking point can be... θA local coordinate system is established for the direction. The current position of the vehicle and the points of the candidate lanes are projected into the local coordinate system, and the lateral deviation of the vehicle's position information and the interpolation center point in the local coordinate system is calculated. Then, the lane corresponding to the candidate reference line with the smallest lateral deviation value is selected as the target recommended lane.

[0157] For example, in complex intersection scenarios, vehicle perception or maps may have local errors, or temporary disturbances from obstacles may cause multiple reference lines to "appear to be selectable." Therefore, a lateral jump threshold can be set to suppress frequent switching of reference lines. Specifically, the aforementioned lateral jump threshold can be expressed as:

[0158]

[0159] when When switching reference lines is allowed, otherwise the original reference lines will be retained.

[0160] Based on the above steps S1541 to S1542, it can be ensured that the reference line switching only occurs under the premise of satisfying physical stability and trajectory continuity, thereby improving the path tracking accuracy and driving smoothness of the vehicle in complex intersection scenarios and avoiding the risk of sudden changes in control commands caused by inconsistency of coordinate systems or reference drift.

[0161] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0162] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0163] This application also provides a path planning device for implementing the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0164] Figure 7 This is a structural block diagram of a path planning device according to one embodiment of this application, such as... Figure 7 As shown, the device includes:

[0165] The acquisition module 701 is used to acquire vehicle status data, multiple candidate reference lines, attribute information corresponding to the multiple candidate reference lines, obstacle trajectory and vehicle predicted trajectory. The vehicle status data includes vehicle position information, and the candidate reference lines are used to represent the vehicle's optional driving path within the intersection area.

[0166] Analysis module 702 is used to analyze and process the vehicle status data, multiple candidate reference lines, obstacle trajectory and vehicle predicted trajectory to obtain analysis results. The analysis results are used to determine the reference line type corresponding to multiple candidate reference lines. The reference line type includes safe reference lines and high-risk reference lines. The collision risk value corresponding to the safe reference line is less than or equal to the preset risk threshold, and the collision risk value corresponding to the high-risk reference line is greater than the preset risk threshold.

[0167] The first determining module 703 is used to determine the initial recommended lane based on the vehicle's status data, the vehicle's predicted trajectory, and the safety reference line;

[0168] The second determining module 704 is used to determine an initial locking target based on the vehicle position information and multiple candidate reference lines. The initial locking target is determined based on multiple lateral projection values ​​between the vehicle position information and multiple candidate reference lines. The lateral projection values ​​are used to represent the geometric offset of the candidate reference lines relative to the vehicle position information in the lateral direction.

[0169] The third determination module 705 is used to determine the target recommended lane based on multiple candidate reference lines, the attribute information corresponding to the multiple candidate reference lines, the reference line type, the vehicle position information, the initial recommended lane, and the initial locked target.

[0170] Optionally, the vehicle status data also includes: the vehicle's real-time speed and acceleration. The analysis module 702 is further configured to: construct an obstacle trajectory envelope based on multiple candidate reference lines and obstacle trajectories, wherein the obstacle trajectory envelope represents the spatial coverage area corresponding to the obstacle trajectory within a preset time; analyze and process each candidate reference line and obstacle trajectory envelope among the multiple candidate reference lines to obtain a first lateral coordinate value and a second lateral coordinate value of the obstacle trajectory envelope relative to each candidate reference line, wherein the first lateral coordinate value is less than the second lateral coordinate value; analyze and process the obstacle trajectory and the vehicle's predicted trajectory to obtain a first time and a second time, wherein the first time represents the preset time within a preset time for the obstacle to travel from its current position to the predicted collision point, and the second time represents the preset time within a preset time for the vehicle to travel from its current position to the predicted collision point, the first time being determined based on the obstacle's current speed and acceleration, the second time being determined based on the vehicle's real-time speed and acceleration, and the predicted collision point being the spatiotemporal conflict point corresponding to the obstacle trajectory and the vehicle's predicted trajectory; analyze and process the first lateral coordinate value, the second lateral coordinate value, the first time, and the second time to obtain the analysis result.

[0171] Optionally, the analysis module 702 is further configured to: determine that the obstacle is in a lateral crossing state in response to a first lateral coordinate value being less than a first preset threshold and a second lateral coordinate value being greater than a second preset threshold, wherein the first preset threshold is less than the second preset threshold; determine that the candidate reference line is a high-risk reference line in response to the obstacle being in a lateral crossing state and the difference between the first time and the second time being less than a preset time threshold; and determine that the candidate reference line is a safe reference line in response to the obstacle not being in a lateral crossing state, and / or the difference between the first time and the second time being greater than or equal to a preset time threshold.

[0172] Optionally, the vehicle status data further includes: the vehicle's driving direction. The first determining module 703 is further configured to: perform equidistant sampling processing on the safety reference line according to the vehicle's driving direction to obtain an equidistant sampling sequence, wherein the equidistant sampling sequence includes a first sampling point and a second sampling point, and the distance between the second sampling point and the starting point of the safety reference line is greater than the distance between the first sampling point and the starting point of the safety reference line; in response to the distance between the first sampling point and the second sampling point being less than a preset distance threshold, delete the second sampling point to obtain a center point sequence corresponding to the safety reference line; and determine the initial recommended lane based on the center point sequence and the vehicle's predicted trajectory.

[0173] Optionally, the first determining module 703 is further configured to: analyze and process multiple predicted trajectory points in the center point sequence and the vehicle's predicted trajectory to obtain multiple first center trajectory points and multiple second center trajectory points, wherein the distance between the first center trajectory points and the predicted trajectory points is less than a preset distance threshold, and the first center trajectory points and the second center trajectory points are adjacent trajectory points in the center point sequence; perform linear interpolation processing on the multiple first center trajectory points and the multiple second center trajectory points to obtain multiple interpolated center points; determine multiple target distances based on the multiple interpolated center points and the multiple predicted trajectory points, wherein the target distance is the lateral and vertical distance between the interpolated center points and the corresponding predicted trajectory points; determine the matching degree between the center point sequence and the vehicle's predicted trajectory based on the average value of the multiple target distances, wherein the matching degree is inversely proportional to the average value; and determine an initial recommended lane based on the matching degree between the center point sequence and the vehicle's predicted trajectory, wherein the matching degree between the center point sequence corresponding to the initial recommended lane and the vehicle's predicted trajectory is less than a preset matching degree threshold.

[0174] Optionally, the second determining module 704 is further configured to: determine a set of lateral projection values ​​based on the vehicle's position information and multiple candidate reference lines, wherein the set of lateral projection values ​​includes multiple lateral projection values; sort the multiple candidate reference lines in ascending order based on the multiple lateral projection values ​​to obtain a recommended reference line sequence; and determine an initial locking target based on the recommended reference line sequence, wherein the initial locking target is a candidate reference line corresponding to a preset sequence identifier in the recommended reference line sequence, and the preset sequence identifier is a first value.

[0175] Optionally, the attribute information corresponding to multiple candidate reference lines includes: the number of candidate reference lines and the locking point corresponding to the candidate reference lines. The third determining module 705 is further configured to: determine the target recommended lane as the initial recommended lane in response to the number of high-risk reference lines being less than the number of candidate reference lines in the current frame, and the real-time speed of the vehicle being greater than or equal to a preset speed threshold; or, determine the target recommended lane as the lane corresponding to the initial locking target in response to the number of high-risk reference lines being less than the number of candidate reference lines in the current frame, and the real-time speed of the vehicle being less than a preset speed threshold; or, determine the target recommended lane as the initial recommended lane in response to the number of high-risk reference lines being greater than or equal to the number of candidate reference lines in the current frame, and the real-time speed of the vehicle being greater than or equal to a preset speed threshold; or, determine the target recommended lane based on the locking point corresponding to the candidate reference line and the vehicle's position information in response to the number of high-risk reference lines being greater than or equal to the number of candidate reference lines in the current frame, and the real-time speed of the vehicle being less than a preset speed threshold, wherein the locking point corresponding to the candidate reference line is the starting point of the first non-virtual segment among the candidate reference lines, and the locking point is determined based on the boundary attributes corresponding to the candidate reference line.

[0176] Optionally, the third determining module 705 is further configured to: establish a local coordinate system based on the location information and orientation information corresponding to the locking point; and determine the target recommended lane based on the vehicle's location information and the lateral deviation value of the interpolation center point in the local coordinate system, wherein the lateral deviation value corresponding to the target recommended lane is less than a preset deviation threshold.

[0177] It should be noted that the above modules can be implemented by software or hardware. For the latter, they can be implemented in the following ways, but are not limited to: all the above modules are located in the same processor; or, the above modules are located in different processors in any combination.

[0178] Embodiments of this application also provide a vehicle, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods described in various embodiments of this application when it runs.

[0179] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program:

[0180] Step S11: Obtain vehicle status data, multiple candidate reference lines, attribute information corresponding to the multiple candidate reference lines, obstacle trajectory, and vehicle predicted trajectory. The vehicle status data includes vehicle position information, and the candidate reference lines are used to represent the vehicle's optional driving paths within the intersection area.

[0181] Step S12: Analyze and process the vehicle status data, multiple candidate reference lines, obstacle trajectory and vehicle predicted trajectory to obtain analysis results. The analysis results are used to determine the reference line type corresponding to the multiple candidate reference lines. The reference line type includes safe reference lines and high-risk reference lines. The collision risk value corresponding to the safe reference line is less than or equal to the preset risk threshold, and the collision risk value corresponding to the high-risk reference line is greater than the preset risk threshold.

[0182] Step S13: Based on the vehicle status data, the vehicle predicted trajectory, and the safety reference line, determine the initial recommended lane, wherein the matching degree between the initial recommended lane and the vehicle predicted trajectory is less than a preset matching degree threshold.

[0183] Step S14: Based on the vehicle position information and multiple candidate reference lines, determine the initial locking target. The initial locking target is determined based on multiple lateral projection values ​​between the vehicle position information and multiple candidate reference lines. The lateral projection values ​​are used to represent the geometric offset of the candidate reference lines relative to the vehicle position information in the lateral direction.

[0184] Step S15: Based on multiple candidate reference lines, the attribute information corresponding to the multiple candidate reference lines, the reference line type, the vehicle position information, the initial recommended lane, and the initial locked target, determine the target recommended lane.

[0185] Embodiments of this application also provide a computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0186] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0187] Step S11: Obtain vehicle status data, multiple candidate reference lines, attribute information corresponding to the multiple candidate reference lines, obstacle trajectory, and vehicle predicted trajectory. The vehicle status data includes vehicle position information, and the candidate reference lines are used to represent the vehicle's optional driving paths within the intersection area.

[0188] Step S12: Analyze and process the vehicle status data, multiple candidate reference lines, obstacle trajectory and vehicle predicted trajectory to obtain analysis results. The analysis results are used to determine the reference line type corresponding to the multiple candidate reference lines. The reference line type includes safe reference lines and high-risk reference lines. The collision risk value corresponding to the safe reference line is less than or equal to the preset risk threshold, and the collision risk value corresponding to the high-risk reference line is greater than the preset risk threshold.

[0189] Step S13: Based on the vehicle status data, the vehicle predicted trajectory, and the safety reference line, determine the initial recommended lane, wherein the matching degree between the initial recommended lane and the vehicle predicted trajectory is less than a preset matching degree threshold.

[0190] Step S14: Based on the vehicle position information and multiple candidate reference lines, determine the initial locking target. The initial locking target is determined based on multiple lateral projection values ​​between the vehicle position information and multiple candidate reference lines. The lateral projection values ​​are used to represent the geometric offset of the candidate reference lines relative to the vehicle position information in the lateral direction.

[0191] Step S15: Based on multiple candidate reference lines, the attribute information corresponding to the multiple candidate reference lines, the reference line type, the vehicle position information, the initial recommended lane, and the initial locked target, determine the target recommended lane.

[0192] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0193] Embodiments of this application also provide a computer program product, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0194] Optionally, in this embodiment, the above-mentioned computer program product can be configured as a computer program that performs the following steps:

[0195] Step S11: Obtain vehicle status data, multiple candidate reference lines, attribute information corresponding to the multiple candidate reference lines, obstacle trajectory, and vehicle predicted trajectory. The vehicle status data includes vehicle position information, and the candidate reference lines are used to represent the vehicle's optional driving paths within the intersection area.

[0196] Step S12: Analyze and process the vehicle status data, multiple candidate reference lines, obstacle trajectory and vehicle predicted trajectory to obtain analysis results. The analysis results are used to determine the reference line type corresponding to the multiple candidate reference lines. The reference line type includes safe reference lines and high-risk reference lines. The collision risk value corresponding to the safe reference line is less than or equal to the preset risk threshold, and the collision risk value corresponding to the high-risk reference line is greater than the preset risk threshold.

[0197] Step S13: Based on the vehicle status data, the vehicle predicted trajectory, and the safety reference line, determine the initial recommended lane, wherein the matching degree between the initial recommended lane and the vehicle predicted trajectory is less than a preset matching degree threshold.

[0198] Step S14: Based on the vehicle position information and multiple candidate reference lines, determine the initial locking target. The initial locking target is determined based on multiple lateral projection values ​​between the vehicle position information and multiple candidate reference lines. The lateral projection values ​​are used to represent the geometric offset of the candidate reference lines relative to the vehicle position information in the lateral direction.

[0199] Step S15: Based on multiple candidate reference lines, the attribute information corresponding to the multiple candidate reference lines, the reference line type, the vehicle position information, the initial recommended lane, and the initial locked target, determine the target recommended lane.

[0200] Embodiments of this application also provide a computer program product, including a non-volatile computer-readable storage medium for storing a computer program that, when executed by a processor, implements the methods in various embodiments of this application.

[0201] Embodiments of this application also provide a computer program that, when executed by a processor, implements the methods described in the various embodiments of this application.

[0202] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0203] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0204] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0205] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0206] If the integrated unit is implemented as 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 technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0207] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A path planning method, characterized in that, include: The system acquires vehicle status data, multiple candidate reference lines, attribute information corresponding to the multiple candidate reference lines, obstacle trajectories, and predicted vehicle trajectories. The vehicle status data includes vehicle position information, and the candidate reference lines are used to represent the vehicle's selectable driving paths within the intersection area. The vehicle status data, the multiple candidate reference lines, the obstacle trajectory, and the predicted trajectory of the vehicle are analyzed and processed to obtain analysis results. The analysis results are used to determine the reference line type corresponding to the multiple candidate reference lines. The reference line type includes safe reference lines and high-risk reference lines. The collision risk value corresponding to the safe reference line is less than or equal to a preset risk threshold, and the collision risk value corresponding to the high-risk reference line is greater than the preset risk threshold. Based on the vehicle status data, the predicted trajectory of the vehicle, and the safety reference line, an initial recommended lane is determined; Based on the vehicle position information and the plurality of candidate reference lines, an initial locking target is determined, wherein the initial locking target is determined based on a plurality of lateral projection values ​​between the vehicle position information and the plurality of candidate reference lines, and the lateral projection values ​​are used to represent the geometric offset of the candidate reference lines relative to the vehicle position information in the lateral direction; Based on the multiple candidate reference lines, the attribute information corresponding to the multiple candidate reference lines, the reference line type, the vehicle position information, the initial recommended lane, and the initial locked target, a target recommended lane is determined.

2. The method according to claim 1, characterized in that, The vehicle status data also includes: real-time vehicle speed and vehicle acceleration. The analysis and processing of the vehicle status data, the multiple candidate reference lines, the obstacle trajectory, and the predicted vehicle trajectory yields the following results: Based on the multiple candidate reference lines and the obstacle trajectory, an obstacle trajectory envelope is constructed, wherein the obstacle trajectory envelope is used to represent the spatial coverage area corresponding to the obstacle trajectory within a preset time period; Each candidate reference line and the obstacle trajectory envelope among the plurality of candidate reference lines are analyzed and processed to obtain a first lateral coordinate value and a second lateral coordinate value of the obstacle trajectory envelope relative to each candidate reference line, wherein the first lateral coordinate value is less than the second lateral coordinate value; The obstacle trajectory and the predicted trajectory of the vehicle are analyzed and processed to obtain a first time and a second time. The first time represents the preset time within which the obstacle travels from its current position to the predicted collision point within a preset time. The second time represents the preset time within the preset time within which the vehicle travels from its current position to the predicted collision point. The first time is determined based on the current speed and acceleration of the obstacle, and the second time is determined based on the real-time speed and acceleration of the vehicle. The predicted collision point is the spatiotemporal conflict point corresponding to the obstacle trajectory and the predicted trajectory of the vehicle. The first horizontal coordinate value, the second horizontal coordinate value, the first time, and the second time are analyzed and processed to obtain the analysis result.

3. The method according to claim 2, characterized in that, The analysis and processing of the first horizontal coordinate value, the second horizontal coordinate value, the first time, and the second time to obtain the analysis results include: In response to the first lateral coordinate value being less than a first preset threshold and the second lateral coordinate value being greater than a second preset threshold, it is determined that the obstacle is in a lateral crossing state, wherein the first preset threshold is less than the second preset threshold; In response to the obstacle being in a lateral crossing state, and the difference between the first time and the second time being less than a preset time threshold, the candidate reference line is determined to be a high-risk reference line; In response to the obstacle not being in the lateral crossing state, and / or the difference between the first time and the second time being greater than or equal to a preset time threshold, the candidate reference line is determined as the safety reference line.

4. The method according to claim 1, characterized in that, The vehicle status data also includes: the vehicle's driving direction. Determining the initial recommended lane based on the vehicle status data, the predicted vehicle trajectory, and the safety reference line includes: The safety reference line is sampled at equal intervals according to the vehicle's driving direction to obtain an equal interval sampling sequence. The equal interval sampling sequence includes a first sampling point and a second sampling point. The distance between the second sampling point and the starting point of the safety reference line is greater than the distance between the first sampling point and the starting point of the safety reference line. In response to the fact that the distance between the first sampling point and the second sampling point is less than a preset distance threshold, the second sampling point is deleted to obtain the center point sequence corresponding to the safety reference line; The initial recommended lane is determined based on the center point sequence and the predicted trajectory of the vehicle.

5. The method according to claim 4, characterized in that, The process of determining the initial recommended lane based on the center point sequence and the vehicle's predicted trajectory includes: The center point sequence and multiple predicted trajectory points in the predicted trajectory of the vehicle are analyzed and processed to obtain multiple first center trajectory points and multiple second center trajectory points. The distance between the first center trajectory point and the predicted trajectory point is less than a preset distance threshold. The first center trajectory point and the second center trajectory point are adjacent trajectory points in the center point sequence. Linear interpolation is performed on the plurality of first center trajectory points and the plurality of second center trajectory points to obtain a plurality of interpolated center points; Based on the multiple interpolation center points and the multiple predicted trajectory points, multiple target distances are determined, wherein the target distance is the horizontal and vertical distance between the interpolation center point and the corresponding predicted trajectory point; Based on the average value corresponding to the multiple target distances, the matching degree between the center point sequence and the predicted trajectory of the vehicle is determined, wherein the matching degree is inversely proportional to the average value; Based on the matching degree between the center point sequence and the predicted trajectory of the vehicle, an initial recommended lane is determined, wherein the matching degree between the center point sequence corresponding to the initial recommended lane and the predicted trajectory of the vehicle is less than a preset matching degree threshold.

6. The method according to claim 5, characterized in that, The process of determining the initial target lock based on the vehicle's position information and multiple candidate reference lines includes: Based on the vehicle's position information and multiple candidate reference lines, a set of lateral projection values ​​is determined, wherein the set of lateral projection values ​​includes the multiple lateral projection values; Based on the multiple lateral projection values, the multiple candidate reference lines are sorted in ascending order to obtain a recommended reference line sequence; Based on the recommended reference line sequence, the initial locking target is determined, wherein the initial locking target is a candidate reference line corresponding to a preset sequence identifier in the recommended reference line sequence, and the preset sequence identifier is a first value.

7. The method according to claim 6, characterized in that, The attribute information corresponding to the multiple candidate reference lines includes: the number of candidate reference lines and the lock point corresponding to the candidate reference lines. The vehicle status data also includes: the real-time speed of the vehicle. Determining the target recommended lane based on the multiple candidate reference lines, the attribute information corresponding to the multiple candidate reference lines, the reference line type, the vehicle position information, the initial recommended lane, and the initial lock target includes: In response to the fact that the number of high-risk reference lines is less than the number of candidate reference lines in the current frame, and the real-time speed of the vehicle is greater than or equal to a preset speed threshold, the target recommended lane is determined as the initial recommended lane; or, In response to the fact that the number of high-risk reference lines is less than the number of candidate reference lines in the current frame, and the real-time speed of the vehicle is less than the preset speed threshold, the target recommended lane is determined to be the lane corresponding to the initially locked target; or, In response to the fact that the number of high-risk reference lines is greater than or equal to the number of candidate reference lines in the current frame, and the real-time speed of the vehicle is greater than or equal to the preset speed threshold, the target recommended lane is determined as the initial recommended lane; or, In response to the fact that the number of high-risk reference lines is greater than or equal to the number of candidate reference lines in the current frame, and the real-time speed of the vehicle is less than the preset speed threshold, the target recommended lane is determined based on the locking point corresponding to the candidate reference line and the vehicle's position information. The locking point corresponding to the candidate reference line is the starting point of the first non-virtual segment among the candidate reference lines, and the locking point is determined based on the boundary attributes corresponding to the candidate reference line.

8. The method according to claim 7, characterized in that, The step of determining the target recommended lane based on the locking point corresponding to the candidate reference line and the vehicle's position information includes: A local coordinate system is established based on the location and orientation information corresponding to the locking point; Based on the vehicle's location information and the lateral deviation value of the interpolation center point in the local coordinate system, the target recommended lane is determined, wherein the lateral deviation value corresponding to the target recommended lane is less than a preset deviation threshold.

9. A vehicle, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device on which the storage medium is located to perform the method according to any one of claims 1 to 8.

Images