An autonomous driving path planning method, device, vehicle, and storage medium

By selecting an appropriate planner combination strategy based on the map data source, the problems of low path planning efficiency and path infeasibility in the existing technology are solved, and efficient and accurate path planning is achieved under different map data.

CN117302257BActive Publication Date: 2026-11-17FAW JIEFANG AUTOMOTIVE CO
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202311243142.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-09-25
Publication Date
2026-11-17
Estimated Expiration
2043-09-25

AI Technical Summary

Technical Problem

In the existing technology, maps from multiple map data sources are inefficient in path planning and may lead to infeasible paths, especially for road maps that have avoided obstacles and obstacle exemption maps. The uniform use of existing planners leads to problems of low efficiency or infeasible paths.

Method used

Based on the map data source, select an appropriate planner combination strategy, including task planner, global modification layer planner, limited area parking planner, dynamic obstacle trajectory prediction planner, kinematic planner, speed planner, and starting obstacle checker. By combining these planners, determine the planner combination strategy that matches the target map data and generate an accurate target driving path.

Benefits of technology

It improves the efficiency and accuracy of route planning, ensures the adaptability of route planning under different map data sources, avoids situations where the route is not feasible, and achieves accurate determination of the target driving route.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117302257B_ABST
    Figure CN117302257B_ABST
Patent Text Reader

Abstract

This invention discloses an autonomous driving path planning method, device, vehicle, and storage medium. The method includes: acquiring target map data corresponding to the current vehicle; determining a target planner combination strategy that matches the target map data from a set of preset path planners, based on the data source corresponding to the target map data; wherein the data source includes a first obstacle exemption map generated based on satellite imagery, a second obstacle exemption map acquired based on a navigation system, a conversion map obtained from a recorded path by the actual vehicle, and a map generated based on preset waypoints; determining a target driving path corresponding to the current vehicle based on the target map data and the target planner combination strategy; and controlling the current vehicle to drive according to the target driving path. The technical solution of this invention can accurately determine the target driving path, improving the efficiency of target driving path determination.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and in particular to an autonomous driving path planning method, device, vehicle, and storage medium. Background Technology

[0002] Autonomous driving is a cutting-edge technology that relies on computers and artificial intelligence to drive itself without a driver. As the application scenarios for autonomous vehicles become more widespread and map data sources become increasingly abundant, the ability to match different planners based on various map data sources has become a crucial factor affecting the accuracy of path planning. These various map data sources can include obstacle exemption road maps generated from satellite imagery, obstacle exemption maps obtained from navigation map systems, maps derived from paths recorded by actual vehicles, and maps with preset waypoint sequences.

[0003] In existing technologies, the planner used for maps with multiple data sources is consistent. Specifically, for maps with any data source, existing technologies employ a unified planner, such as a task planner and a global modification layer planner, for path planning.

[0004] However, for road maps where obstacles have already been avoided (such as maps derived from real vehicle paths), the task path in the map has already undergone obstacle avoidance processing during path planning. If a global modification layer planner is then used to perform obstacle avoidance processing on the task path again, the path planning efficiency will be low. Secondly, for obstacle-exempt road maps, if a task planner is used for path planning, there may be situations where the task path overlaps with obstacles, rendering the task path infeasible. Summary of the Invention

[0005] This invention provides an autonomous driving path planning method, device, vehicle, and storage medium, which can accurately determine the target driving path and improve the efficiency of target driving path determination.

[0006] In a first aspect, embodiments of the present invention provide an autonomous driving path planning method, the method comprising:

[0007] Obtain the target map data corresponding to the current vehicle, and determine the target planner combination strategy that matches the target map data from multiple preset path planners based on the data source corresponding to the target map data.

[0008] The data sources include a first obstacle exemption map generated based on satellite imagery, a second obstacle exemption map obtained based on a navigation system, a conversion map obtained based on a real vehicle recorded path, and a map generated based on preset waypoints.

[0009] Based on the target map data and the target planner combination strategy, determine the target driving path corresponding to the current vehicle;

[0010] Control the current vehicle to drive along the target driving path.

[0011] Optionally, based on the data source corresponding to the target map data, a target planner combination strategy matching the target map data is determined from a set of preset path planners, including:

[0012] If the data source corresponding to the target map data is a first obstacle exemption map, a second obstacle exemption map, or a map generated based on preset waypoints, then the preset task planner, global modification layer planner, limited area parking planner, dynamic obstacle trajectory prediction planner, kinematics planner, velocity planner, and starting obstacle checker are combined to obtain a target planner combination strategy that matches the target map data.

[0013] Optionally, based on the data source corresponding to the target map data, a target planner combination strategy matching the target map data is determined from a set of preset path planners, including:

[0014] If the target map data is a converted map obtained from a real vehicle recorded path, then a target planner combination strategy matching the target map data is determined from multiple preset path planners according to the preset obstacle avoidance requirements.

[0015] Optionally, based on preset obstacle avoidance requirements, a target planner combination strategy matching the target map data is determined from a set of preset path planners, including:

[0016] If the preset obstacle avoidance requirement is obstacle avoidance, then the preset task planner, limited area parking planner, dynamic obstacle trajectory prediction planner, kinematic planner, velocity planner and starting obstacle checker are combined to obtain a target planner combination strategy that matches the target map data.

[0017] Optionally, based on the data source corresponding to the target map data, a target planner combination strategy matching the target map data is determined from a set of preset path planners, including:

[0018] If the preset obstacle avoidance requirement is obstacle encounter parking, then the preset task planner, limited area parking planner, dynamic obstacle trajectory prediction planner, obstacle encounter parking planner, speed planner and starting obstacle checker are combined to obtain a target planner combination strategy that matches the target map data.

[0019] Optionally, based on the target map data and the target planner combination strategy, determine the target driving path corresponding to the current vehicle; control the current vehicle to drive according to the target driving path, including:

[0020] If the data source corresponding to the target map data is a first obstacle exemption map, a second obstacle exemption map, or a map generated based on preset waypoints, then the current vehicle position, preset vehicle start point, end point, and multiple task points are obtained based on the target map data.

[0021] A task planner is used to perform global path planning based on preset vehicle start and end points and multiple task points to obtain the initial task path.

[0022] A global modification layer planner is used to extract the target waypoint from the initial task path and replan the path between the current vehicle position and the target waypoint to obtain a medium-distance path;

[0023] The driving speed of dynamic obstacles is obtained through environmental sensing devices, and a dynamic obstacle trajectory prediction planner is used to simulate the dynamic obstacle driving at a constant speed in a constant direction. The simulation results are used as the trajectory prediction results of the dynamic obstacle.

[0024] A kinematics planner is used to determine the target driving path based on mid-distance path and trajectory prediction results;

[0025] A speed planner is used to determine the speed parameters of the current vehicle as it travels along the target driving path;

[0026] The starting obstacle detector is used to check for obstacles around the vehicle. When it is determined that there are no obstacles around the vehicle, the current vehicle is controlled to drive according to the target driving path and speed parameters.

[0027] Optionally, the method further includes:

[0028] A limited-area parking planner is used to determine the target parking path based on the current vehicle position and the preset parking range, so that the current vehicle can drive according to the target parking path.

[0029] Secondly, embodiments of the present invention also provide an autonomous driving path planning device, comprising:

[0030] The combined strategy determination module is used to acquire the target map data corresponding to the current vehicle, and determine the target planner combined strategy that matches the target map data from a set of preset path planners, based on the data source corresponding to the target map data. The data sources include a first obstacle exemption map generated based on satellite imagery, a second obstacle exemption map obtained based on the navigation system, a conversion map obtained based on the actual vehicle recorded path, and a map generated based on preset waypoints.

[0031] The target driving path determination module is used to determine the target driving path corresponding to the current vehicle based on the target map data and the target planner combination strategy.

[0032] The vehicle autonomous driving module is used to control the current vehicle to drive according to the target driving path.

[0033] Thirdly, embodiments of the present invention also provide a vehicle, the vehicle comprising:

[0034] At least one processor; and

[0035] A memory that is communicatively connected to at least one processor; wherein,

[0036] The memory stores a computer program that can be executed by at least one processor, such that the at least one processor is able to execute the autonomous driving path planning method provided in any embodiment of the present invention.

[0037] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing computer instructions that, when executed by a processor, implement the autonomous driving path planning method provided in any embodiment of the present invention.

[0038] The technical solution provided by this invention obtains target map data corresponding to the current vehicle, and determines a target planner combination strategy that matches the target map data from a set of preset path planners based on the data source corresponding to the target map data. The data sources include a first obstacle exemption map generated based on satellite imagery, a second obstacle exemption map obtained based on a navigation system, a conversion map obtained from a recorded path by the actual vehicle, and a map generated based on preset waypoints. Based on the target map data and the target planner combination strategy, the target driving path corresponding to the current vehicle is determined. This technical means of controlling the current vehicle to drive according to the target driving path can accurately determine the target driving path, improving the efficiency of target driving path determination.

[0039] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a flowchart of an autonomous driving path planning method provided in Embodiment 1 of the present invention;

[0042] Figure 2 This is a flowchart of another autonomous driving path planning method provided in Embodiment 2 of the present invention;

[0043] Figure 3 This is a flowchart of another autonomous driving path planning method provided in Embodiment 3 of the present invention;

[0044] Figure 4 This is a decision-making and planning architecture diagram applicable to unstructured roads provided according to an embodiment of the present invention;

[0045] Figure 5 This is a schematic diagram of the structure of an autonomous driving path planning device according to Embodiment 4 of the present invention;

[0046] Figure 6 This is a structural schematic diagram of a vehicle provided in Embodiment 5 of the present invention. Detailed Implementation

[0047] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0048] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention 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 the invention 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 a 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.

[0049] Example 1

[0050] Figure 1 This is a flowchart of an autonomous driving path planning method according to Embodiment 1 of the present invention. This embodiment is applicable to the situation where a vehicle plans an autonomous driving path in an unstructured road scenario. The method can be executed by an autonomous driving path planning device, which can be implemented in hardware and / or software and can be configured in a vehicle.

[0051] like Figure 1 As shown in this embodiment, an autonomous driving path planning method is disclosed, which includes:

[0052] S110. Obtain the target map data corresponding to the current vehicle. Based on the data source corresponding to the target map data, determine the target planner combination strategy that matches the target map data from among multiple preset path planners.

[0053] The data sources include a first obstacle exemption map generated based on satellite imagery, a second obstacle exemption map obtained based on a navigation system, a conversion map obtained based on a real vehicle recorded path, and a map generated based on preset waypoints.

[0054] In this embodiment, the target map data can be map data requested by the user. The path planner can be used to control the vehicle for autonomous driving. The target planner combination strategy can be a strategy in which multiple path planners cooperate to achieve autonomous driving. The first obstacle exemption map, the second obstacle exemption map, and the map generated based on preset waypoints do not model static obstacles. The transformed map obtained based on the actual vehicle recorded path establishes a model of static obstacles, that is, static obstacles have been avoided. Among them, the accuracy of the second obstacle exemption map obtained based on the navigation system is at the meter level. The preset multiple path planners may include a task planner, a global modification layer planner, a limited area parking planner, a dynamic obstacle trajectory prediction planner, a kinematics planner, an obstacle encounter parking planner, a speed planner, and a starting obstacle checker.

[0055] In one specific implementation, a target planner combination strategy can be determined based on the map data source corresponding to the target map data and preset user needs. These preset user needs may include navigation, obstacle avoidance, and parking.

[0056] For example, if the map data source corresponding to the target map data is a first obstacle exemption map generated based on satellite imagery, a path planner with static obstacle avoidance function can be selected from a set of preset path planners to be combined with other planners to obtain a target planner combination strategy.

[0057] As another example, if the source of the map data corresponding to the target map data is a converted map obtained based on the actual vehicle recorded path, then there is no need to select a path planner with static obstacle avoidance function and generate a target planner combination strategy.

[0058] S120. Based on the target map data and the target planner combination strategy, determine the target driving path corresponding to the current vehicle.

[0059] In this embodiment, the target driving path can be a path that has already avoided static obstacles.

[0060] In this step, specifically, the target driving path can be determined through the cooperation of the planners in the target planner combination strategy. For example, assuming the target planner combination strategy includes a task planner and a global modification layer planner, the task planner can first be used to plan the global path. Then, the global modification layer planner can be used to modify the global path, and the modified global path can be used as the target driving path.

[0061] S130, Control the current vehicle to drive according to the target driving path.

[0062] In practical applications, the target driving path can serve as input information for autonomous driving control, allowing the vehicle's autonomous driving decision-making system to combine the current vehicle position with the information to achieve autonomous driving.

[0063] The advantage of this setup is that, compared to existing technologies that use a unified path planner for maps from any data source, the technical solution in this embodiment improves the adaptability of the target planner combination strategy to each data source map by determining a target planner combination strategy that matches the target map data from among multiple preset path planners. This avoids including redundant path planners in the target planner combination strategy and improves the efficiency of determining the target driving path. Secondly, for obstacle exemption road maps, by determining a target planner combination strategy that matches the target map data, the lack of a path planner with obstacle avoidance functionality in the target planner combination strategy, which could render the target driving path infeasible, can achieve accurate determination of the target driving path.

[0064] The technical solution of this embodiment obtains the target map data corresponding to the current vehicle, and determines a target planner combination strategy that matches the target map data from a set of preset path planners based on the data source corresponding to the target map data. The data sources include a first obstacle exemption map generated based on satellite imagery, a second obstacle exemption map obtained based on a navigation system, a converted map obtained from a recorded path by the actual vehicle, and a map generated based on preset waypoints. Based on the target map data and the target planner combination strategy, the target driving path corresponding to the current vehicle is determined. This technical means of controlling the current vehicle to drive according to the target driving path solves the problem that existing technologies use a unified planner for maps from any data source, resulting in low path planning efficiency. It can accurately determine the target driving path and improve the efficiency of target driving path determination.

[0065] Example 2

[0066] Figure 2 This is a flowchart of another autonomous driving path planning method provided by Embodiment 2 of the present invention. This embodiment is a further optimization and extension based on the above embodiments and can be combined with various optional technical solutions in the above embodiments.

[0067] like Figure 2 As shown in this embodiment, another autonomous driving path planning method is disclosed, which includes:

[0068] S210. Obtain the target map data corresponding to the current vehicle.

[0069] S220. If the data source corresponding to the target map data is a first obstacle exemption map, a second obstacle exemption map, or a map generated based on preset waypoints, then the preset task planner, global modification layer planner, limited area parking planner, dynamic obstacle trajectory prediction planner, kinematics planner, velocity planner, and starting obstacle checker are combined to obtain a target planner combination strategy that matches the target map data.

[0070] In this embodiment, the task planner can be used to plan a task path from a preset vehicle start point to a preset end point. The global modification layer planner can be used to replan a path with a preset planning distance. The limited area parking planner can be used to plan a path between the current vehicle position and a preset parking range. The dynamic obstacle trajectory prediction planner can be used to predict the trajectory of dynamic obstacles. The kinematics planner can be used to plan a path with a preset movement distance. The preset movement distance can be less than the preset planning distance.

[0071] In one specific implementation, for each obstacle exemption map, a task planner can be selected to plan the global path from the vehicle's starting point to its destination. Then, a global modification layer planner is selected to achieve static obstacle avoidance. Since there are dynamic obstacles around the vehicle, a dynamic obstacle trajectory prediction planner can be selected to predict the dynamic obstacle trajectory. A kinematics planner is selected to replan the path with a preset movement distance. Next, based on the user's parking needs, a limited-area parking planner is selected to achieve limited-area parking. Finally, a speed planner can be selected to set the vehicle's speed for each path. A starting obstacle checker is selected to check the vehicle's surrounding environment when the vehicle starts.

[0072] The advantage of this setup is that, since the obstacle exemption map does not model static obstacles, a medium-distance path bypassing static obstacles is obtained by replanning the path with a preset correction distance using a global modification layer planner, thus improving the accuracy of the target driving path determination. Secondly, the kinematic planner further improves the accuracy of path planning by replanning paths with smaller distances. Finally, the starting obstacle checker checks the vehicle's surrounding environment when the vehicle starts, ensuring driving safety.

[0073] S230. If the data source corresponding to the target map data is a converted map obtained based on the actual vehicle recorded path, then according to the preset obstacle avoidance requirements, determine the target planner combination strategy that matches the target map data from among the preset multiple path planners.

[0074] In this embodiment, the preset obstacle avoidance requirements may include obstacle avoidance and obstacle parking.

[0075] In an optional embodiment of the present invention, a target planner combination strategy matching the target map data is determined from a plurality of preset path planners according to a preset obstacle avoidance requirement. This includes: if the preset obstacle avoidance requirement is obstacle avoidance, then the preset task planner, limited area parking planner, dynamic obstacle trajectory prediction planner, kinematic planner, speed planner and starting obstacle checker are combined to obtain a target planner combination strategy matching the target map data.

[0076] Specifically, for the converted map based on the recorded path of the real vehicle, the initial task path determined by the task planner has already avoided static obstacles, so there is no need to select the global modification layer planner, which improves the efficiency of path planning.

[0077] For example, if there are oncoming vehicles in the same lane and the road is too narrow for both vehicles to pass at the same time, a speed planner can be used to slow down the current vehicle before the two vehicles meet, thus avoiding a collision and improving driving safety.

[0078] In another optional embodiment of the present invention, based on the data source corresponding to the target map data, a target planner combination strategy matching the target map data is determined from a plurality of preset path planners, including: if the preset obstacle avoidance requirement is obstacle encounter parking, then the preset task planner, limited area parking planner, dynamic obstacle trajectory prediction planner, obstacle encounter parking planner, speed planner and starting obstacle checker are combined to obtain a target planner combination strategy matching the target map data.

[0079] In this embodiment, optionally, for any map corresponding to any data source, if the preset obstacle avoidance requirement is obstacle encounter and parking, the obstacle encounter and parking planner can be selected from a preset number of path planners to be combined with other planners to obtain a target planner combination strategy.

[0080] In one specific implementation, if a preset obstacle exists in front of the current vehicle, an obstacle-detection parking planner can be used to interrupt the planned path at the location of the preset obstacle (i.e., ignore the planned path from the preset obstacle to the destination), and stop the vehicle at a preset safe distance from the preset obstacle, at which point the current vehicle's speed is 0. The preset obstacle can be determined according to user needs. For example, the preset obstacle can be a straight-arm barrier or a no-entry sign.

[0081] S240. Based on the target map data and the target planner combination strategy, determine the target driving path corresponding to the current vehicle.

[0082] S250: Control the current vehicle to drive according to the target driving path.

[0083] The technical solution of this embodiment obtains the target map data corresponding to the current vehicle. If the data source of the target map data is a first obstacle exemption map, a second obstacle exemption map, or a map generated based on preset waypoints, then the preset task planner, global modification layer planner, limited area parking planner, dynamic obstacle trajectory prediction planner, kinematic planner, speed planner, and starting obstacle checker are combined to obtain a target planner combination strategy that matches the target map data. If the data source of the target map data is a converted map obtained based on a real vehicle recorded path, then the target planner combination strategy that matches the target map data is determined from multiple preset path planners according to preset obstacle avoidance requirements. Based on the target map data and the target planner combination strategy, the target driving path corresponding to the current vehicle is determined. This technical means of controlling the current vehicle to drive according to the target driving path solves the problem that the existing technology uses a unified planner for maps with any data source, resulting in low path planning efficiency. It can make full use of various data sources to accurately determine the target driving path, thus improving the efficiency of target driving path determination.

[0084] Example 3

[0085] Figure 3 This is a flowchart of another autonomous driving path planning method provided by Embodiment 3 of the present invention. This embodiment is a further optimization and extension based on the above embodiments and can be combined with various optional technical solutions in the above embodiments.

[0086] like Figure 3 As shown, another autonomous driving path planning method disclosed in this embodiment includes:

[0087] S310. Obtain the target map data corresponding to the current vehicle. Based on the data source corresponding to the target map data, determine the target planner combination strategy that matches the target map data from among multiple preset path planners.

[0088] S320. If the data source corresponding to the target map data is a first obstacle exemption map, a second obstacle exemption map, or a map generated based on preset waypoints, then the current vehicle position, preset vehicle start point, end point, and multiple task points are obtained according to the target map data.

[0089] S330 uses a task planner to perform global path planning based on preset vehicle start and end points and multiple task points to obtain the initial task path.

[0090] In this embodiment, the initial task path may include a preset vehicle start point, end point, and multiple task points.

[0091] In one specific implementation, a task planner can be used to perform global path planning using a global planning algorithm to obtain the initial task path. The aforementioned global planning algorithm can include heuristic time-space search (A-star, A*) and dynamic programming (DP) algorithms.

[0092] For example, assuming there are preset starting points, a first task point, a second task point, and an ending point, and the distance between the first task point and the starting point is less than the distance between the second task point and the starting point, a global planning algorithm can be used first to plan the path from the starting point to the first task point. Then, the path between the first and second task points can be planned. Finally, the path from the second task point to the ending point can be planned.

[0093] The advantage of this setup is that, compared to existing technologies that only plan the path between the preset vehicle start and end points, the technical solution of this embodiment improves the accuracy of path planning by planning the paths from the start point to the task point, between multiple task points, and from the task point to the end point separately.

[0094] S340. Using a global modification layer planner, the target waypoint is extracted from the initial task path, and the path between the current vehicle position and the target waypoint is replanned to obtain a medium-distance path.

[0095] In this step, specifically, target waypoints can be extracted from the initial task path based on a preset planned distance (e.g., 75 meters) using an equidistant or equally spaced indexing method. Then, a grid-based arbitrary angle pathfinding algorithm, Basic Theta* or A* post-processing algorithm, can be used to replan the path from the current vehicle position to the target waypoint. Optionally, when the current vehicle reaches the target waypoint, the aforementioned target waypoint can be used as the new current vehicle position, and a new target waypoint can be extracted from the initial task path. This process is repeated to achieve obstacle avoidance processing on the initial task path.

[0096] The advantage of this setup is that, for maps that do not model obstacles within the road, the target waypoint set may collide with obstacles, making the path infeasible. Therefore, by replanning the path from the current vehicle position to the target waypoint, the accuracy of the target driving path is improved.

[0097] S350: The driving speed of the dynamic obstacle is obtained through the environmental perception device, and the dynamic obstacle trajectory prediction planner is used to simulate the dynamic obstacle driving in a constant direction at a driving speed. The simulation result is used as the trajectory prediction result of the dynamic obstacle.

[0098] Specifically, in this step, the trajectory prediction result of the dynamic obstacle within the preset time range can be determined by assuming that the dynamic obstacle travels in a constant direction at a constant speed or constant acceleration within the preset time range.

[0099] The S360 uses a kinematic planner to determine the target driving path based on mid-distance path and trajectory prediction results.

[0100] In one specific implementation, the vehicle's projection point on the mid-distance path is determined based on its current position. Then, a target point is determined on the mid-distance path based on a preset longitudinal distance and the projection point, and this target point is used as the planning endpoint. The trajectory prediction results of dynamic obstacles and static obstacles are projected onto a grid map. Next, the lateral sampling range is determined based on the distance between the projection point and the planning endpoint, and the vehicle's maximum front wheel angle. The formula for the lateral sampling range is (projection point - planning endpoint) × tan(maximum front wheel angle). Multiple sampling points are determined based on the lateral sampling range and the preset longitudinal distance. Finally, a cubic polynomial is fitted based on the sampling points. The cubic polynomial is solved using four known quantities. These four known quantities can be the longitudinal distance of the projection point of the current vehicle pose on the mid-distance path, the tangent of the difference between the vehicle's heading angle and the heading angle at the projection point, the lateral offset of the sampling point, and the vehicle's driving direction being parallel to the path.

[0101] S370 uses a speed planner to determine the speed parameters of the current vehicle as it travels along the target driving path.

[0102] In this embodiment, speed parameters may include vehicle straight-line speed, vehicle turning speed, vehicle stopping speed, etc. Specifically, a speed planner can be used to impose speed limits on the target driving path based on the speed parameters. Then, a quadratic programming (QP) method can be used to optimize the current vehicle speed to make the target driving path smoother while satisfying vehicle dynamics constraints. The aforementioned vehicle dynamics constraints may include the vehicle's maximum speed, maximum acceleration, and maximum jerk, etc.

[0103] S380: Uses a starting obstacle detector to check for obstacles around the vehicle. When it is determined that there are no obstacles around the vehicle, it controls the current vehicle to drive according to the target driving path and speed parameters.

[0104] In an optional embodiment of the invention, the method further includes: using a limited area parking planner to determine a target parking path based on the current vehicle position and a preset parking range, so that the current vehicle drives according to the target parking path.

[0105] In this step, specifically step 1, we can first assume that there are no obstacles within the preset parking area. Then, using the current vehicle position as the center, starting with a preset minimum radius, the radius is gradually expanded according to a preset step size, without exceeding the preset maximum radius, until there are no obstacles within a circle of a defined radius. At this point, there are obstacles in the (N+1)th circle, and no obstacles in the Nth circle. Therefore, the Nth circle can be used as the starting exploration circle, with the current vehicle position as the center and the current vehicle direction as the direction. The preset minimum and maximum radii can be set according to user needs. For example, the preset minimum radius can be set to the radius of the current vehicle's circumcircle, and the preset maximum radius can be set to 5 times the radius of the current vehicle's circumcircle. The preset step size can be set according to the grid map resolution and preset speed requirements.

[0106] Step 2 involves using a preset parking point within the pre-defined parking area as the center, and gradually expanding the radius starting from a preset minimum radius. Based on this, and assuming there are no obstacles within the preset parking area, the radius of the endpoint exploration circle can be the maximum radius within the preset parking area.

[0107] Step 3: Determine the target lateral sampling range based on the vehicle's orientation angle and the preset orientation angle neighborhood (e.g., 40-60 degrees to the left and right of the orientation angle). Then, based on the target lateral sampling range, determine reference sampling points on the circumference of the starting exploration circle.

[0108] Step 4: Evaluate each reference sampling point using an A*-like heuristic algorithm. The cost of each reference sampling point can be obtained by adding the searched distance and the heuristic distance. The searched distance is the sum of the radii of all exploration circles before the current exploration circle, and the heuristic distance can be the Euclidean distance from the center of the current exploration circle to the reference sampling point. The reference sampling point with the lowest replacement value is selected as the target sampling point.

[0109] Step 5: Generate a new exploration circle centered on the target sampling point. The direction of the new exploration circle is the same as the direction of the line connecting the center of the previous exploration circle to the target sampling point.

[0110] Step 6: Based on the new exploration circle, determine new target sampling points until the overlap between the new exploration circle and the endpoint exploration circle is greater than a preset overlap threshold, thus completing the exploration. The preset overlap threshold can be determined based on user needs and historical experience. For example, it can be set to 0.6. Optionally, the overlap can be determined based on the overlapping area of ​​the two circles. For example, if the overlapping area of ​​the two circles is 60%, then the overlap can be considered to be 0.6.

[0111] Step 7: Traverse back from the endpoint exploration circle to the starting exploration circle. All the exploration circles form the exploration corridor; the line connecting the centers of all the exploration circles from the starting exploration circle to the endpoint exploration circle forms the initial reference path.

[0112] Step 8: Cancel obstacle release within the preset parking area, i.e., reconsider obstacles within the preset parking area. Generate an exploration path based on kinematic constraints.

[0113] The above exploration path is generated as follows: First, an offline trajectory cluster is generated offline based on the vehicle model and preset parameters. This method allows for matching the current vehicle pose at any location, quickly generating offline trajectories, and saving online computing resources. Then, the maximum front wheel steering angle of the vehicle is equally divided according to the preset number of trajectories corresponding to the offline trajectory cluster. The 360-degree yaw angle is divided according to the preset yaw angle, which can be set as 'a'. Afterward, given the preset yaw angle, each offline trajectory corresponds to one front wheel steering angle. Except for straight-ahead trajectories, all other turning offline trajectories stop after turning through a preset yaw angle. Straight-ahead trajectories stop after passing a preset fixed length. Finally, the current vehicle starting point and yaw angle are obtained, and an offline trajectory cluster is generated at that position. Each offline trajectory in the offline trajectory cluster is evaluated.

[0114] Evaluation metrics include: the heuristic distance (e.g., Euclidean distance) from the offline trajectory endpoint to the target point, the distance from the offline trajectory endpoint to the nearest center of the circle on the initial reference path, the radius of the exploration circle created with the offline trajectory endpoint, whether the offline trajectory direction is consistent with the initial reference path (positive infinity if it is, 0 if it is), and the collision cost (positive infinity if a collision occurs, 0 otherwise). The optimal offline trajectory is selected from the current offline trajectory cluster. The trajectory cluster is expanded further, using the optimal trajectory endpoint as the new current point. When the optimal trajectory endpoint is within the preset parking area, the expansion of the offline trajectory cluster stops, and the exploration path is obtained.

[0115] After stopping the expansion of the offline trajectory cluster, a backtracking method is used to obtain the target parking path located within the exploration corridor and satisfying the vehicle kinematic constraints.

[0116] The advantage of this setup is that, since limited-area parking refers to the current vehicle autonomously parking within a preset parking range (without parking spaces, non-fixed-point parking), it is suitable for autonomous driving in unstructured road scenarios. Therefore, the technical solution of this embodiment adds a limited-area parking planner to the target planner combination strategy, improving the efficiency of parking in unstructured road scenarios. Secondly, the technical solution of this embodiment first generates an exploration corridor based on the generation of an exploration circle, and then performs a heuristic search within the exploration corridor that satisfies the Ackerman steering model, obtaining a path located within the exploration corridor that satisfies the vehicle's kinematic constraints. This improves the search efficiency of the exploration path, effectively avoiding collisions and quickly reaching the target area, with even more obvious advantages in scenarios with many obstacles and complex environments.

[0117] Figure 4 This is a decision planning architecture diagram applicable to unstructured roads provided according to an embodiment of the present invention.

[0118] like Figure 4 As shown, this embodiment provides an architecture including a task planner, a global modification layer planner, a limited-area parking planner, a dynamic obstacle trajectory prediction planner, a kinematics planner, an obstacle-encounter parking planner, a speed planner, and a starting obstacle checker. For any data source map, the task planner, speed planner, and starting obstacle checker must be selected. If the map data source is an obstacle exemption map, the global modification layer planner, limited-area parking planner, dynamic obstacle trajectory prediction planner, and kinematics planner must be further selected. If the map data source is a converted map based on a recorded vehicle path and obstacle avoidance is required, the limited-area parking planner, dynamic obstacle trajectory prediction planner, and kinematics planner must be further selected. If the map data source is a converted map based on a recorded vehicle path and obstacle avoidance is not required, the limited-area parking planner, dynamic obstacle trajectory prediction planner, and obstacle-encounter parking planner must be further selected.

[0119] The advantage of this setup is that by adapting to various data source maps in unstructured road scenarios at the architecture layer, different target planner combination strategies can be determined, thereby improving the accuracy and efficiency of path planning while fully utilizing the data source maps. Secondly, the planner combination strategies determined at the architecture layer have a wider range of applicability and a simpler implementation process.

[0120] The technical solution of this embodiment acquires target map data and determines a target planner combination strategy based on the data source corresponding to the target map data. If the data source corresponding to the target map data is a first obstacle exemption map, a second obstacle exemption map, or a map generated based on preset waypoints, then the current vehicle position, preset vehicle start point, end point, and multiple task points are obtained based on the target map data. A task planner is used to perform global path planning based on the preset vehicle start point, end point, and multiple task points to obtain an initial task path. A global modification layer planner is used to extract target waypoints from the initial task path and replan the path between the current vehicle position and the target waypoints to obtain a medium-distance path. The driving speed of dynamic obstacles is acquired through an environmental perception device, and... A dynamic obstacle trajectory prediction planner is employed to simulate dynamic obstacles traveling at a constant speed in a constant direction, and the simulation results are used as the trajectory prediction results for the dynamic obstacles. A kinematic planner is used to determine the target driving path based on the mid-distance path and the trajectory prediction results. A speed planner is used to determine the speed parameters of the current vehicle as it travels along the target driving path. A starting obstacle checker is used to check the obstruction of obstacles around the vehicle. When it is determined that there are no obstacles around the vehicle, the current vehicle is controlled to drive according to the target driving path and speed parameters. This technical means solves the problem that existing technologies use a uniform planner for maps with arbitrary data sources, resulting in low path planning efficiency. It can accurately determine the target driving path and improve the efficiency of target driving path determination.

[0121] Example 4

[0122] Figure 5 This is a schematic diagram of an autonomous driving path planning device according to Embodiment 4 of the present invention. This embodiment is applicable to the situation where a vehicle plans an autonomous driving path in an unstructured road scenario. The autonomous driving path planning device can be implemented in hardware and / or software and can be configured in a vehicle.

[0123] like Figure 5 As shown, the autonomous driving path planning device disclosed in this embodiment includes:

[0124] The combination strategy determination module 51 is used to obtain the target map data corresponding to the current vehicle, and determine the target planner combination strategy that matches the target map data from a number of preset path planners according to the data source corresponding to the target map data; wherein, the data source includes a first obstacle exemption map generated based on satellite imagery, a second obstacle exemption map obtained based on the navigation system, a conversion map obtained based on the actual vehicle recorded path, and a map generated based on preset waypoints.

[0125] The target driving path determination module 52 is used to determine the target driving path corresponding to the current vehicle based on the target map data and the target planner combination strategy.

[0126] The vehicle autonomous driving module 53 is used to control the current vehicle to drive according to the target driving path.

[0127] The technical solution in this embodiment solves the problem that existing technologies use a unified planner for maps with arbitrary data sources, resulting in low path planning efficiency, by combining the strategy determination module, the target driving path determination module, and the vehicle autonomous driving module. This allows for accurate determination of the target driving path and improves the efficiency of target driving path determination.

[0128] Optionally, the combination strategy determination module 51 includes:

[0129] The first combination strategy determination unit is used to combine the preset task planner, global modification layer planner, limited area parking planner, dynamic obstacle trajectory prediction planner, kinematic planner, speed planner and starting obstacle checker to obtain a target planner combination strategy that matches the target map data if the data source corresponding to the target map data is a first obstacle exemption map, a second obstacle exemption map or a map generated based on preset waypoints.

[0130] The second combination strategy determination unit is used to determine the target planner combination strategy that matches the target map data from multiple preset path planners according to preset obstacle avoidance requirements if the data source corresponding to the target map data is a converted map obtained based on the actual vehicle recorded path.

[0131] The obstacle avoidance strategy determination unit is used to combine the preset task planner, limited area parking planner, dynamic obstacle trajectory prediction planner, kinematic planner, velocity planner and starting obstacle checker if the preset obstacle avoidance requirement is obstacle avoidance, to obtain a target planner combination strategy that matches the target map data.

[0132] The obstacle encounter parking strategy determination unit is used to combine the preset task planner, limited area parking planner, dynamic obstacle trajectory prediction planner, obstacle encounter parking planner, speed planner and starting obstacle checker if the preset obstacle avoidance requirement is obstacle encounter parking, to obtain a target planner combination strategy that matches the target map data.

[0133] Optionally, the target driving path determination module 52 includes:

[0134] The location acquisition unit is used to acquire the current vehicle location, preset vehicle start point, end point, and multiple task points based on the target map data if the data source corresponding to the target map data is a first obstacle exemption map, a second obstacle exemption map, or a map generated based on preset waypoints.

[0135] The initial task path determination unit is used to perform global path planning based on preset vehicle start point, end point and multiple task points using a task planner to obtain the initial task path.

[0136] The mid-distance path determination unit is used to extract the target waypoint from the initial task path using the global modification layer planner, and replan the path between the current vehicle position and the target waypoint to obtain the mid-distance path;

[0137] The trajectory prediction result determination unit is used to obtain the driving speed of the dynamic obstacle through the environmental sensing device, and to use the dynamic obstacle trajectory prediction planner to simulate the dynamic obstacle driving at a constant speed in a constant direction, and use the simulation result as the trajectory prediction result of the dynamic obstacle.

[0138] The target driving path determination unit is used to determine the target driving path based on the mid-distance path and trajectory prediction results using a kinematic planner.

[0139] The speed parameter determination unit is used to determine the speed parameters of the current vehicle as it travels along the target driving path using a speed planner.

[0140] The vehicle automatic driving unit is used to check the obstruction of obstacles around the vehicle using a starting obstacle checker, and when it is determined that there are no obstacles around the vehicle, it controls the current vehicle to drive according to the target driving path and speed parameters.

[0141] The target parking path determination unit is used to determine the target parking path based on the current vehicle position and the preset parking range using a limited area parking planner, so that the current vehicle drives according to the target parking path.

[0142] The autonomous driving path planning device provided in this embodiment of the invention can execute the autonomous driving path planning method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method execution. Content not described in detail in this embodiment can be referred to the description in any method embodiment of this application.

[0143] Example 5

[0144] Figure 6 A schematic diagram of the structure of a vehicle 10 that can be used to implement an embodiment of the present invention is shown.

[0145] like Figure 6As shown, vehicle 10 includes at least one processor 11 and a memory, such as read-only memory (ROM) 12 and random access memory (RAM) 13, communicatively connected to at least one processor 11. The memory stores computer programs executable by at least one processor. Processor 11 can perform various appropriate actions and processes based on the computer program stored in ROM 12 or loaded from storage unit 18 into RAM 13. RAM 13 can also store various programs and data required for the operation of vehicle 10. Processor 11, ROM 12, and RAM 13 are interconnected via bus 14. Input / output (I / O) interface 15 is also connected to bus 14.

[0146] Multiple components in vehicle 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows vehicle 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0147] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as autonomous driving path planning methods.

[0148] In some embodiments, the autonomous driving path planning method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on vehicle 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the autonomous driving path planning method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the autonomous driving path planning method by any other suitable means (e.g., by means of firmware).

[0149] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0150] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0151] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0152] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0153] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0154] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0155] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0156] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. An automated driving path planning method, characterized by, The method includes: Obtain the target map data corresponding to the current vehicle, and determine the target planner combination strategy that matches the target map data from a set of preset path planners, based on the data source corresponding to the target map data. The data sources include a first obstacle exemption map generated based on satellite imagery, a second obstacle exemption map obtained based on a navigation system, a conversion map obtained based on a real vehicle recorded path, and a map generated based on preset waypoints; Based on the target map data and the target planner combination strategy, the target driving path corresponding to the current vehicle is determined; Control the current vehicle to drive according to the target driving path; Specifically, based on the data source corresponding to the target map data, a target planner combination strategy matching the target map data is determined from a set of preset path planners, including: If the data source corresponding to the target map data is a first obstacle exemption map, a second obstacle exemption map, or a map generated based on preset waypoints, then the preset task planner, global modification layer planner, limited area parking planner, dynamic obstacle trajectory prediction planner, kinematics planner, velocity planner, and starting obstacle checker are combined to obtain a target planner combination strategy that matches the target map data. If the data source corresponding to the target map data is a converted map obtained based on the actual vehicle recorded path, then according to the preset obstacle avoidance requirements, a target planner combination strategy that matches the target map data is determined from multiple preset path planners. The step of determining a target planner combination strategy that matches the target map data from a preset set of multiple path planners based on preset obstacle avoidance requirements includes: If the preset obstacle avoidance requirement is obstacle avoidance, then the preset task planner, limited area parking planner, dynamic obstacle trajectory prediction planner, kinematic planner, velocity planner and starting obstacle checker are combined to obtain a target planner combination strategy that matches the target map data. If the preset obstacle avoidance requirement is obstacle encounter parking, then the preset task planner, limited area parking planner, dynamic obstacle trajectory prediction planner, obstacle encounter parking planner, speed planner and starting obstacle checker are combined to obtain a target planner combination strategy that matches the target map data.

2. The method of claim 1, wherein, Based on the target map data and the target planner combination strategy, the target driving path corresponding to the current vehicle is determined; Controlling the current vehicle to drive according to the target driving path includes: If the data source corresponding to the target map data is a first obstacle exemption map, a second obstacle exemption map, or a map generated based on preset waypoints, then the current vehicle position, preset vehicle start point, end point, and multiple task points are obtained according to the target map data. A task planner is used to perform global path planning based on preset vehicle start and end points and multiple task points to obtain the initial task path. A global modification layer planner is used to extract the target waypoint from the initial task path and replan the path between the current vehicle position and the target waypoint to obtain a medium-distance path; The driving speed of the dynamic obstacle is obtained through environmental sensing equipment, and the dynamic obstacle trajectory prediction planner is used to simulate the dynamic obstacle traveling in a constant direction at the driving speed. The simulation result is used as the trajectory prediction result of the dynamic obstacle. A kinematic planner is used to determine the target driving path based on the mid-distance path and trajectory prediction results; A speed planner is used to determine the speed parameters of the current vehicle as it travels along the target driving path; An obstacle detection device is used to check for obstacles around the vehicle. When it is determined that there are no obstacles around the vehicle, the current vehicle is controlled to drive according to the target driving path and speed parameters.

3. The method of claim 2, wherein, The method further includes: A limited-area parking planner is used to determine a target parking path based on the current vehicle position and a preset parking range, so that the current vehicle drives according to the target parking path.

4. An automatic driving route planning device characterized by comprising: The apparatus is used to perform the autonomous driving path planning method according to any one of claims 1-3, the apparatus comprising: The combination strategy determination module is used to obtain the target map data corresponding to the current vehicle, and determine the target planner combination strategy that matches the target map data from a set of preset path planners, based on the data source corresponding to the target map data. The data sources include a first obstacle exemption map generated based on satellite imagery, a second obstacle exemption map obtained based on a navigation system, a conversion map obtained based on a real vehicle recorded path, and a map generated based on preset waypoints; The target driving path determination module is used to determine the target driving path corresponding to the current vehicle based on the target map data and the target planner combination strategy. The vehicle autonomous driving module is used to control the current vehicle to drive according to the target driving path.

5. A vehicle characterized by comprising: The vehicles include: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the autonomous driving path planning method according to any one of claims 1-3.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the autonomous driving path planning method according to any one of claims 1-3.

Citation Information

Patent Citations

  • Apparatus and method for generating traveling path of autonomous vehicle

    CN107300914A

  • Path planning method, electronic equipment and mobile device

    CN114184203A