Intelligent driving vehicle, obstacle avoidance path planning method and device thereof and medium

Abstract

The invention discloses an intelligent driving vehicle, an obstacle avoidance path planning method and device thereof and a medium, and relates to the technical field of path planning. The obstacle avoidance path planning method comprises the steps that a current obstacle sensed on a global path is responded, a first obstacle avoidance path is planned, and the expected end point of the first obstacle avoidance path is located in a first obstacle avoidance safety space relative to the current obstacle; and when the terminal point of the first obstacle avoidance path is reached, obstacle sensing is carried out again, and a second obstacle avoidance path is planned according to a sensing result. The embodiment of the invention provides a step-by-step obstacle avoidance planning strategy, the first step is used for planning the first obstacle avoidance path according to the current obstacle information, and the second step is used for planning at the end point of the first obstacle avoidance path with a wider sensing range, so that the problem of easy planning failure caused by an obstacle sensing blind area is solved, the computing power consumption is low, and the planning efficiency is improved. And landing use in a plurality of application scenes is facilitated.

Description

Technical Field

[0001] The present application relates to the field of path planning technology, and in particular to an intelligent driving vehicle and an obstacle avoidance path planning method, device, and medium thereof. Background Art

[0002] Obstacle avoidance planning is an essential function for intelligent vehicles, such as smart cars, intelligent construction machinery, and mobile robots. This function ensures that intelligent vehicles can automatically generate paths to avoid obstacles during autonomous operation.

[0003] Existing intelligent driving vehicle obstacle avoidance path planning solutions mainly include:

[0004] Based on the currently perceived environmental information, a one-time local path planning is performed to determine an obstacle avoidance path. However, this solution, which relies on a local map for path planning, is limited by map size and blind spots in obstacle perception, and often fails to capture complete obstacle information. Expanding the local map consumes more computing power, impacting the overall stability and efficiency of the intelligent driving vehicle, while also increasing hardware requirements. This limitation limits the applicability of this one-time local planning approach, making it unsuitable for many engineering scenarios. For example, in the mining environment in which intelligent driving mining trucks operate, obstacles such as fallen rocks, earth piles, and other vehicles are common. There may even be long obstacles such as temporary construction equipment and other mining trucks. When such a long obstacle is in front of the mining truck, the overall situation of the obstacle and the state of the environment behind it are unknown. For example, it is impossible to determine whether there are other vehicles traveling in the opposite direction behind the obstacle, resulting in blind spots in perception. This results in incomplete obstacle information, making one-time planning more difficult and reducing the success rate of planning.

[0005] Therefore, a new obstacle avoidance path planning scheme for intelligent driving vehicles is needed. Summary of the Invention

[0006] The purpose of the embodiments of the present application is to provide an intelligent driving vehicle and its obstacle avoidance path planning method, device and medium, so as to at least partially solve the above technical problems.

[0007] To achieve the above-mentioned objectives, the first aspect of the present application provides a method for planning an obstacle avoidance path for an intelligent driving vehicle, comprising: in response to sensing a current obstacle on a global path, planning a first obstacle avoidance path, wherein the expected end point of the first obstacle avoidance path is located within a first obstacle avoidance safety space relative to the current obstacle; and upon reaching the expected end point of the first obstacle avoidance path, re-performing obstacle perception and planning a second obstacle avoidance path based on the perception results.

[0008] In an embodiment of the present application, the process of planning the first obstacle avoidance path also includes determining the first obstacle avoidance safety space through the following steps: determining the global path point on the global path that is closest to the current obstacle; starting from the global path point, extending multiple path points with different headings along a first preset direction until the extended path points extend to the road boundary or another obstacle; and constructing the first obstacle avoidance safety space relative to the current obstacle based on the global path point and the multiple path points extended therefrom.

[0009] In an embodiment of the present application, planning the first obstacle avoidance path includes: starting from the global path point in the first obstacle avoidance safety space, determining multiple alternative end points of the first obstacle avoidance path along a direction perpendicular to the heading of the vehicle; performing path planning based on the starting point of the first obstacle avoidance path and each determined alternative end point to obtain corresponding multiple obstacle avoidance paths, wherein the starting point of the first obstacle avoidance path is on the global path and the distance between the vehicle and the current obstacle satisfies a preset obstacle avoidance condition; and selecting a path with the minimum path cost from the multiple obstacle avoidance paths as the first obstacle avoidance path, wherein the alternative end point corresponding to the path with the minimum path cost is the expected end point of the first obstacle avoidance path.

[0010] In an embodiment of the present application, the re-performing obstacle perception includes: starting from the vehicle's current position, searching for a first obstacle point for the current obstacle along a direction perpendicular to the vehicle's heading; starting from the first obstacle point, traversing obstacle points along the heading of the current obstacle until the end point of the current obstacle is searched; and continuing to search for obstacle points a preset number of times starting from the end point of the current obstacle. If a new obstacle point is found, a perception result indicating the presence of a new obstacle is obtained.

[0011] In an embodiment of the present application, planning the second obstacle avoidance path includes constructing a second obstacle avoidance safety space through the following steps: determining a final end point, wherein if the new obstacle exists, the end point of the new obstacle is the final end point, otherwise the end point of the current obstacle is the final end point; and starting from the final end point, constructing a second obstacle avoidance safety space along a second preset direction.

[0012] In an embodiment of the present application, after constructing the second obstacle avoidance safety space, planning the second obstacle avoidance path further includes: determining a plurality of intermediate control points of the obstacle avoidance path along the second preset direction within the second obstacle avoidance safety space; performing path planning based on the starting point of the second obstacle path and the plurality of intermediate control points of the obstacle avoidance path to obtain a corresponding plurality of obstacle avoidance paths, wherein the expected end point of the first obstacle path is used as the starting point of the second obstacle path; and selecting a path with the minimum path cost from the plurality of obstacle avoidance paths as the second obstacle path.

[0013] In an embodiment of the present application, the second preset direction is a direction perpendicular to the heading of a reference path point, and the reference path point is a path point on the global path that is closest to the final end point.

[0014] In an embodiment of the present application, the first obstacle avoidance path and the second obstacle avoidance path are planned based on a grid map.

[0015] In an embodiment of the present application, after planning the second obstacle avoidance path, the obstacle avoidance path planning method further includes: performing collision risk detection on the second obstacle avoidance path; if there is a collision risk, detecting the collision point and determining the path point on the global path that is closest to the collision point; and constructing a new obstacle avoidance safety space based on a direction perpendicular to the heading of the determined path point to re-plan the obstacle avoidance path.

[0016] A second aspect of the present application provides an obstacle avoidance path planning device for an intelligent driving vehicle, comprising: a first path planning module for planning a first obstacle avoidance path in response to sensing a current obstacle on a global path, wherein the expected end point of the first obstacle avoidance path is located within a first obstacle avoidance safety space relative to the current obstacle; and a second path planning module for planning a second obstacle avoidance path based on the perception result of re-obstacle perception when the expected end point of the first obstacle avoidance path is reached.

[0017] A third aspect of the present application provides a driving control method for an intelligent driving vehicle, comprising: when the vehicle senses an obstacle on a global path, controlling the vehicle to change lanes from the global path to a first obstacle avoidance path; and when the vehicle reaches the expected end point of the first obstacle avoidance path, controlling the vehicle to change lanes from the first obstacle avoidance path to a second obstacle avoidance path, and returning to the global path via the second obstacle avoidance path. The first obstacle avoidance path and the second obstacle avoidance path are generated based on any of the above-described obstacle avoidance path planning methods.

[0018] In a fourth aspect, the present application provides a control device for an intelligent driving vehicle, comprising: a memory configured to store instructions; and a processor configured to call the instructions from the memory and to implement any of the above-mentioned obstacle avoidance path planning methods or driving control methods when executing the instructions.

[0019] A fifth aspect of the present application provides an intelligent driving vehicle, comprising any of the above-mentioned obstacle avoidance path planning devices or control devices.

[0020] In an embodiment of the present application, the intelligent driving vehicle is any one of an intelligent car, an intelligent engineering machinery, an intelligent agricultural machinery and a mobile robot.

[0021] In a sixth aspect, the present application provides a machine-readable storage medium having stored thereon instructions for enabling a machine to execute any of the above-mentioned obstacle avoidance path planning methods or driving control methods.

[0022] Through the above technical solution, the embodiment of the present application provides a step-by-step obstacle avoidance planning strategy. The first step is to plan a first obstacle avoidance path based on the current obstacle information, and the second step is to plan at the expected end point of the first obstacle avoidance path with a wider perception range, thereby solving the problem of easy planning failure due to blind spots in obstacle perception, and the computing power consumption is low, which is conducive to implementation in multiple application scenarios.

[0023] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the following detailed description, they are used to explain the embodiments of the present application but do not constitute a limitation on the embodiments of the present application. In the accompanying drawings:

[0025] Figure 1 1 is a flow chart of an obstacle avoidance path planning method for an intelligent driving vehicle according to the first embodiment of the present application;

[0026] Figure 2 This is a schematic diagram of the principle of step-by-step obstacle avoidance path planning based on a grid map in Example 1 of the present application;

[0027] Figure 3 1 is a flow chart of determining the first obstacle avoidance safe space in the first embodiment of the present application;

[0028] Figure 4 This is a schematic diagram of the process of planning the first obstacle circumvention path in Example 1 of the present application;

[0029] Figure 5This is a schematic diagram of the process of re-performing obstacle perception in Example 1 of the present application;

[0030] Figure 6 Schematic diagram of the process of planning the second obstacle circumvention path in Example 1 of the present application;

[0031] Figure 7 This is a schematic diagram of the process of collision handling for the second obstacle avoidance path in the first embodiment of the present application;

[0032] Figure 8 This is a flowchart of step-by-step obstacle avoidance planning for a mining truck in the example of the first embodiment of the present application;

[0033] Figure 9 This is a schematic diagram of the principle of determining obstacle avoidance conditions for a mining truck in the example of the first embodiment of the present application;

[0034] Figure 10 This is a schematic diagram of the principle of planning the first obstacle avoidance path for a mining truck in the example of the first embodiment of the present application.

[0035] Figure 11 This is a schematic diagram of the principle of planning the second obstacle avoidance path for a mining truck in the example of the first embodiment of the present application;

[0036] Figure 12 This is a schematic diagram of a process for determining the final end point of an obstacle for a mining truck in the example of the first embodiment of the present application;

[0037] Figure 13 This is a schematic diagram of the principle of performing collision detection on the second obstacle avoidance path for a mining truck in the example of the first embodiment of the present application;

[0038] Figure 14 2 is a schematic structural diagram of an obstacle avoidance path planning device for an intelligent driving vehicle according to a second embodiment of the present application;

[0039] Figure 15 is a flow chart of a driving control method for an intelligent driving vehicle according to a third embodiment of the present application; and

[0040] Figure 16 It is a structural diagram of the control device of the intelligent driving vehicle of Example 4 of the present application.

[0041] Description of Reference Numerals

[0042] 100 First path planning module 200 Second path planning module

[0043] 300 Obstacle Perception Module 400 Path Evaluation Module DETAILED DESCRIPTION

[0044] To make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the specific implementation methods described herein are only used to illustrate and explain the embodiments of the present application and are not used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0045] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application are in compliance with the relevant provisions of laws and regulations. In the embodiments of this application, certain software, components, models, and other existing solutions in the industry may be mentioned. These should be considered as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of this application, but it does not mean that the applicant has or will necessarily use such solutions.

[0046] It should be noted that if the embodiments of the present application involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, movement status, etc. between the various components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indications will also change accordingly.

[0047] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of such features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the fact that they can be implemented by ordinary technicians in this field. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0048] Before introducing the embodiments of the present application in detail, some of the terms involved are introduced here to facilitate a clearer understanding of the embodiments of the present application.

[0049] 1. Global path planning and local path planning: These are two key concepts in intelligent driving path planning. Global path planning refers to the process of finding an optimal path from a starting point to a destination for an intelligent driving vehicle within the entire environment. Local path planning, given a global path, helps the intelligent driving vehicle make real-time adjustments and avoid obstacles when encountering dynamic obstacles or changes in the local environment. In the embodiments of this application, local path planning is performed on the global path obtained by global path planning to obtain an obstacle avoidance path.

[0050] 2. Obstacle Avoidance Safe Space: This refers to the buffer zone or safe distance that ensures that when an intelligent driving vehicle encounters an obstacle during driving, the vehicle can slow down or avoid the obstacle in time to avoid collision.

[0051] 3. Grid Map: This is a map representation method that divides the environment into regular grid cells. The application scenarios of the embodiments of this application include construction sites such as mines. Compared to conventional roads, such scenes do not have clear lane line information. Therefore, the use of grid maps is more suitable for path planning in such scenarios.

[0052] Example 1

[0053] Figure 1 This is a flow chart of the obstacle avoidance path planning method for an intelligent driving vehicle according to the first embodiment of the present application. Figure 1 As shown, the obstacle avoidance path planning method includes the following steps S100 and S200.

[0054] Step S100 : In response to sensing a current obstacle on a global path, planning a first obstacle avoidance path, wherein an expected end point of the first obstacle avoidance path is located within a first obstacle avoidance safety space relative to the current obstacle.

[0055] Step S200: When the expected end point of the first obstacle circumventing path is reached, obstacle sensing is performed again, and a second obstacle circumventing path is planned based on the sensing result.

[0056] Through step S100 and step S200, it can be seen that when the vehicle is traveling on the global path, if an obstacle is detected, the embodiment of the present application does not perform a one-time obstacle avoidance path planning, but instead designs a step-by-step obstacle avoidance path planning method.

[0057] Specifically, Figure 2 This is a schematic diagram of the principle of step-by-step obstacle avoidance path planning based on a grid map in the first embodiment of the present application. Figure 2As shown, when the vehicle travels from the starting point Start_point along the global path L1 (the end point of the global path is denoted as Goal_point) to the parking point park_point where an obstacle is detected, the first stage obstacle avoidance path planning is performed in step S100 to obtain the first obstacle avoidance path L11 to assist the vehicle in safely bypassing the current obstacle (obstacle point Obstacle_point). When the vehicle reaches the expected end point Front_Mid_point of the first obstacle avoidance path L11, obstacle detection is performed again in step S200 and the second stage obstacle avoidance path planning is started to obtain the second obstacle avoidance path L12, so that the vehicle reaches the end point Goal_point of the global path via the second obstacle avoidance path L12. The expected endpoint of the first obstacle avoidance path L11, Front_Mid_point, lies between the obstacle point Obstacle_point and the path boundary point Boundary_point shown on the map (i.e., within the obstacle avoidance safety space relative to the obstacle). Given the parking point park_point and the expected endpoint Front_Mid_point, a polynomial curve planner can be used to generate the first obstacle avoidance path L11. For the second obstacle avoidance path L12, the middle control point Back_Mid_point of the obstacle avoidance path can be determined first. Given the starting point of the second obstacle avoidance path L12 (i.e., the expected endpoint Front_Mid_point of the first obstacle avoidance path L11) and the middle control point Back_Mid_point, a Bezier curve planner can be used to generate the second obstacle avoidance path L12.

[0058] Thus, compared to a one-time obstacle avoidance path planning to obtain a unique obstacle avoidance path, the embodiment of the present application adopts a step-by-step obstacle avoidance path planning strategy. The first step involves planning a first obstacle avoidance path L11 based solely on current obstacle information, reaching a safe obstacle avoidance space relative to the current obstacle. At the expected endpoint of this first obstacle avoidance path L11, the obstacle perception range is wider, reducing blind spots in all directions of the obstacle perception, such as the following perception blind spots: 1) the obstacle is too long to be perceived at its end point; 2) an oncoming vehicle temporarily appears behind the obstacle. The second step, based on the perception information obtained at the expected endpoint of the first obstacle avoidance path L11, further plans a second obstacle avoidance path L12, thereby resolving the issue of planning failures due to obstacle perception blind spots. Furthermore, compared to global planning, each step in the step-by-step obstacle avoidance path planning in the embodiment of the present application can adapt to smaller maps, eliminating map size limitations and avoiding the computational overhead associated with expanding local maps, thus reducing overall requirements for the intelligent driving system. In addition, compared with the high requirements of real-time dynamic planning on the response speed and stability of the intelligent driving system, the overall hardware and algorithm performance, the obstacle avoidance path planning method of the embodiment of the present application has the advantages of low cost and low computing power consumption, which helps it to be implemented in multiple application scenarios.

[0059] The first stage obstacle avoidance path planning corresponding to step S100 and the second stage obstacle avoidance path planning corresponding to step S200 are described in detail below.

[0060] 1. The first stage of obstacle avoidance path planning.

[0061] As described above, the first stage obstacle avoidance path planning of step S100 involves determining the first obstacle avoidance safe space and planning the first obstacle avoidance path. In a preferred embodiment, Figure 3 As shown, in the process of planning the first obstacle avoidance path, the first obstacle avoidance safety space is determined by performing the following steps S310-S330:

[0062] Step S310: Determine the global path point on the global path that is closest to the current obstacle.

[0063] Step S320 , starting from the global waypoint, extending a plurality of waypoints with different headings along a first predetermined direction until the extended waypoints reach a road boundary or another obstacle. The first predetermined direction may be, for example, a direction perpendicular to the heading of the global waypoint or a road width direction corresponding to the global waypoint.

[0064] Step S330: constructing the first obstacle avoidance safety space relative to the current obstacle based on the global path point and multiple path points extended therefrom.

[0065] It should be noted that steps S310-S330 are merely a preferred method for determining the obstacle avoidance safety space in this embodiment of the present application. In other examples, the obstacle avoidance safety space can be determined by expanding from other pathpoints beyond the aforementioned global pathpoints. Alternatively, an expansion polygon algorithm can be used to expand the obstacle outline outward by a safe distance to obtain the obstacle avoidance safety space. Furthermore, the intelligent driving vehicle can be any of intelligent automobiles, intelligent construction machinery, intelligent agricultural machinery, and mobile robots. The details of determining the first obstacle avoidance safety space in steps S310-S330 will be further described below using the example of a mining truck, and will not be further elaborated here.

[0066] Furthermore, based on the determined first obstacle avoidance safe space, in a more preferred embodiment, as Figure 4 As shown, the first obstacle avoidance path is planned through the following steps S410-S430:

[0067] Step S410 : In the first obstacle avoidance safety space, starting from the global path point, a plurality of candidate endpoints are determined along a direction perpendicular to the heading of the vehicle.

[0068] Step S420: Path planning is performed based on the starting point of the first obstacle circumventing path and each determined candidate end point to obtain corresponding multiple obstacle circumventing paths.

[0069] The starting point of the first obstacle avoidance path is on the global path and ensures that the distance between the vehicle and the current obstacle meets a preset obstacle avoidance condition.

[0070] Step S430: Select a path with the minimum path cost from the multiple obstacle circumventing paths to serve as the first obstacle circumventing path.

[0071] The candidate endpoint corresponding to the path with the minimum path cost is the expected endpoint of the first obstacle circumvention path.

[0072] It is easy to know that the above steps S410-S430 respectively involve path point selection, path generation and path evaluation, which will be specifically introduced below with reference to the example of a mining card, and will not be repeated here.

[0073] 2. The second stage of obstacle avoidance path planning.

[0074] As mentioned above, the second stage obstacle avoidance path planning in step S200 involves obstacle perception and second obstacle avoidance path planning. Figure 5 As shown, in a preferred embodiment, for step S200, re-performing obstacle perception includes the following steps S510-S530:

[0075] Step S510 : Starting from the current position of the vehicle, searching for the first obstacle point for the current obstacle along a direction perpendicular to the vehicle heading.

[0076] Step S520 : Starting from the first obstacle point, traverse obstacle points along the heading of the current obstacle until the end point of the current obstacle is found.

[0077] Step S530: Starting from the end point of the current obstacle, continue searching for obstacle points for a preset number of times. If a new obstacle point is found, a perception result indicating the presence of a new obstacle is obtained.

[0078] Based on the perception results, in a more preferred embodiment, as Figure 6 As shown, planning the second obstacle avoidance path may include two parts: obstacle avoidance space construction and path planning. The obstacle avoidance space construction corresponds to the following steps S610-S620, and the path planning corresponds to the following steps S630-S650:

[0079] Step S610: Determine the final end point. If a new obstacle exists, use the end point of the new obstacle as the final end point; otherwise, use the end point of the current obstacle as the final end point.

[0080] Step S620: Starting from the final end point, construct a second obstacle avoidance safe space along a second preset direction.

[0081] In this example, the second preset direction is perpendicular to the heading of a reference pathpoint, which is the pathpoint on the global path closest to the final endpoint. It should be understood that the selection of the second preset direction is not limited to this example. Pathpoints may also be extended in other directions to determine the second obstacle avoidance safe space, depending on the definition of the obstacle avoidance safe space. For example, the second preset direction may be the road width corresponding to the reference pathpoint.

[0082] In step S630, multiple intermediate control points along the second preset direction are determined within the second obstacle avoidance safety space. For example, starting from the final endpoint, multiple intermediate control points along the direction perpendicular to the heading of the reference path point are determined within the second obstacle avoidance safety space. In other examples, other path points within the second obstacle avoidance safety space may also be used as intermediate control points along the obstacle avoidance path.

[0083] Step S640: Path planning is performed based on the starting point of the second obstacle avoidance path and the plurality of intermediate control points of the obstacle avoidance paths to obtain a plurality of corresponding obstacle avoidance paths, wherein the expected end point of the first obstacle avoidance path is used as the starting point of the second obstacle avoidance path.

[0084] Step S650: Select a path with the minimum path cost from the multiple obstacle circumventing paths to serve as the second obstacle circumventing path.

[0085] Through the above steps S610-S650, a second obstacle-avoiding path is planned, so that the vehicle can bypass the new obstacle through the second obstacle-avoiding path and return to the global path. However, in the process of returning to the global path, continuous obstacles may appear, causing collisions. To avoid this problem, in a preferred embodiment, as shown in FIG. Figure 7 As shown, after planning the second obstacle avoidance path, the obstacle avoidance path planning method may further include the following steps S710-S730.

[0086] Step S710: performing collision risk detection on the second obstacle avoidance path.

[0087] Step S720 : If there is a collision risk, a collision point is detected, and a path point on the global path that is closest to the collision point is determined.

[0088] Step S730: construct a new obstacle avoidance safety space based on a direction perpendicular to the heading of the determined path point to re-plan the obstacle avoidance path.

[0089] That is, dynamic obstacle avoidance planning is implemented through steps S710-730 to avoid obstacle avoidance failure caused by new obstacles due to environmental changes.

[0090] Similarly, the following will also combine the example of mining cards to Figure 5 、 Figure 6 、 Figure 7 The steps involved are described in detail and will not be repeated here.

[0091] The following example uses a mining truck to illustrate the application of the aforementioned obstacle avoidance path planning method. This example describes the scenario of autonomous operation by intelligently driven mining trucks. The mining environment in which they operate may contain long obstacles such as temporary construction sites and other mining trucks. Furthermore, it is impossible to determine whether there are other vehicles traveling in the opposite direction behind the obstacles. This creates perception blind spots and incomplete obstacle information, making one-time obstacle avoidance path planning more difficult and resulting in a low success rate. Furthermore, current mining truck obstacle avoidance path planning involves solving the mining truck's obstacle avoidance path through methods such as searching, sampling, and establishing a repulsive potential field, and setting constraints to ensure the quality of the planned obstacle avoidance path. However, this planning method is also limited by the size of the map and blind spots in obstacle perception. It often fails to obtain complete obstacle information, resulting in uncertainty about the planned destination and, consequently, the inability to plan a safe and feasible obstacle avoidance path.

[0092] To solve the above problems in mining truck path planning, this example is based on Figure 2The step-by-step obstacle avoidance path planning principle shown uses a local grid map to plan the first obstacle avoidance path and the second obstacle avoidance path. That is, during the automatic operation of the intelligent driving mining truck, it continuously detects whether there are obstacles on the local path, and judges the next specific action of the mining truck based on the distance to the obstacle. When the detection distance is less than a certain threshold, the mining truck is safely parked by issuing a parking instruction, and then the obstacle avoidance path planning method of the embodiment of the present application is executed by analyzing whether the obstacle avoidance requirements are met to achieve step-by-step obstacle avoidance. Here, the specific step-by-step obstacle avoidance planning process is as follows: Figure 8 As shown, it mainly includes the following steps S810-S870:

[0093] Step S810: obstacle detection.

[0094] Specifically, obstacles on the global path of the mining truck are detected step by step, and the global coordinate information of the obstacles and their heading are calculated.

[0095] Step S820: Determine whether the obstacle avoidance condition is met. If so, execute step S3; otherwise, determine that the obstacle cannot be avoided.

[0096] Specifically, the distance between the obstacle and the vehicle is calculated, and the vehicle's minimum turning radius is used to determine whether the obstacle can be avoided at the current distance, that is, whether the preset obstacle avoidance conditions are met. Figure 9 The specific judgment process includes:

[0097] First, the turning radius R of the right front corner point B of the mining truck is calculated based on the minimum turning radius of the vehicle:

[0098]

[0099] Among them, R min is the minimum turning radius of the vehicle, W is the vehicle width, and L is the vehicle length.

[0100] Secondly, to ensure that the distance between the vehicle and the obstacle is sufficient to avoid the obstacle, the following conditions must be met:

[0101]

[0102] Where s is the straight-line distance between the obstacle corner point C and the center of the vehicle, that is, the distance between point C and point D in the figure, and y C 、y A They are the coordinate values of point C and the vehicle center point A respectively.

[0103] Finally, considering the safety of the obstacle avoidance process, it is necessary to increase the safety threshold to further ensure it. Therefore, the following formula is designed to determine whether the obstacle avoidance conditions are met:

[0104] Dis≥s+D T

[0105] Where Dis is the relative distance between the vehicle and the obstacle, D T The safety distance threshold is set.

[0106] Step S830: Execute the first stage obstacle avoidance path planning to obtain multiple obstacle avoidance paths.

[0107] Among them, the planning principle of the first obstacle avoidance path is as follows: Figure 10 As shown, combined Figure 3 and Figure 4 The specific planning process includes:

[0108] First, from the parking point Park_piont(x s ,y s ,θ s ), when determining the first-stage endpoint, Front_Mid_point, fixed sampling is performed based on the lateral space. Based on the global path position (coordinates and heading) of the obstacle point obtained in step S810, the global pathpoint closest to the obstacle is found. A certain number of pathpoints are extended backward from the global pathpoint to obtain pathpoints in multiple headings. These pathpoints are then extended perpendicularly to the heading direction to the road edge or the next obstacle point to obtain a lateral obstacle avoidance safety space.

[0109] Then, within the horizontal safety space, follow Figure 4 According to the teaching of the planner, sampling points (i.e., alternative end points) are set at fixed intervals perpendicular to the heading direction, and the path starting point and these sampling points are input into the planner to output multiple obstacle avoidance paths. Among them, the planner can use a polynomial curve planner, which solves the planned path in the vehicle coordinate system by calculating the boundary conditions of the starting point and the end point, and screens the maximum curvature of the path to make it meet the vehicle kinematic requirements. For the multiple obstacle avoidance paths output by the planner, safety and feasibility tests need to be performed to determine the optimal end point, which is recorded as Optimal_point(x e ,y e ,θ e Safety testing primarily involves collision detection, such as using a grid occupancy check method to check all points on the path to ensure there are no collisions along the entire path. Feasibility testing primarily involves path evaluation, which will be involved in the next step.

[0110] Step S840: evaluating the first segment of the path.

[0111] Specifically, path evaluation mainly includes path length and path smoothness evaluation. First, normalize the two evaluation indicators to ensure the consistency of dimensions. Then, set the weight value to evaluate the cost of the candidate path as a whole. The specific cost calculation is shown in the following formula:

[0112]

[0113] J=W L J L +W C J C

[0114] Among them, J L is the length penalty, J C is the curvature cost, J is the final cost; L is the path length, L max and L min is the maximum length and minimum length of all candidate paths; C is the maximum curvature of the path, C max and C min is the maximum and minimum curvature among all candidate paths; W L and W C are the length coefficient weight and curvature coefficient weight respectively.

[0115] Finally, the obstacle avoidance path with the smallest replacement value is selected for coordinate transformation, the path is converted to the global coordinate system, and the corresponding first obstacle avoidance path is output.

[0116] Step S850: Execute the second stage of obstacle avoidance path planning to obtain multiple obstacle avoidance paths.

[0117] Specifically, after the first obstacle avoidance path is completed, it is necessary to construct the second stage obstacle avoidance safety space, while taking into account the continuous obstacles that may exist behind the obstacle. Figure 5 and Figure 6 The planning principle of the second obstacle avoidance path is as follows Figure 11 As shown, it is necessary to first traverse the lateral boundaries of the obstacle, determine the final end point (Final_End_point), and then reconstruct the second obstacle avoidance safety space. Then, multiple candidate midpoints (i.e., midpoints in the obstacle avoidance path, denoted as Candi_Mid_point) are determined within the second obstacle avoidance safety space. Finally, the optimal midpoint (Back_Mid_point) is determined through path evaluation. The candidate path corresponding to this optimal midpoint (Back_Mid_point) is the second obstacle avoidance path.

[0118] Among them, such as Figure 12 As shown, determining the final end point may include the following steps:

[0119] Step S1210: Search for the first obstacle point.

[0120] Specifically, based on the vehicle's current position (the expected end point of the first obstacle avoidance path), the first obstacle point, i.e., the coordinates of the obstacle's lateral boundary surface, is searched in the direction perpendicular to the vehicle's heading. This step is equivalent to initializing the search, so the number of searches Count = 0.

[0121] Step S1220: traverse obstacle points along the obstacle heading direction.

[0122] Specifically, starting from the first obstacle point searched in the previous step, traverse in the heading direction of the obstacle and gradually search to its end, that is, the first obstacle end, which is recorded as Initial_End_point.

[0123] Step S1230 , determine whether the traversal has reached the end, if so, execute step S1240 , otherwise set Count=0 to return to step S1220 and traverse again.

[0124] Step S1240: Continue traversing obstacle points forward.

[0125] Considering the possibility of continuous obstacles, where there is a certain distance between two obstacles, a detection threshold N is set. Specifically, when the search reaches the end of the current obstacle, Count++ is used to continue searching N points forward. N is determined by the search distance. If a new obstacle is found within this distance, the search continues; if no other obstacles are found, the search stops. Ensure that there are no more obstacles within this distance. The end point found when this condition is met is the planned obstacle end point.

[0126] Step S1250: Determine whether Count>N is true. If so, output the end position of the obstacle, that is, the final end point.

[0127] It can be seen that if there is a new obstacle, the end point of the new obstacle is the final end point Final_End_point, otherwise the end point of the current obstacle is the final end point Final_End_point. The final end point determined by this traversal process ensures that Figure 10 The longer obstacles shown and Figure 11 New obstacles shown are fully traversed, helping to reduce obstacle perception blind spots.

[0128] After obtaining the final endpoint, its coordinates are labeled (x, y). The closest point to the final endpoint is found on the global path, and its heading angle is obtained. A lateral safe space, orthogonal to this heading angle, is constructed at the end of the obstacle, i.e., the second obstacle avoidance safe space. Then, sampling is performed along this perpendicular direction at the end of the constructed lateral safe space to determine the intermediate control point of the obstacle avoidance path. Based on the path starting point and the intermediate control points of the obstacle avoidance path, a Bezier curve planner is used for obstacle avoidance planning, resulting in multiple obstacle avoidance paths.

[0129] Here, the first-stage planning in step S830 uses a polynomial curve planner, while the second-stage planning in step S850 uses a Bezier curve planner. In other words, the two stages utilize different planners based on scenario characteristics, and each performs obstacle avoidance checks to ensure that the mining truck's actual operating environment is capable of avoiding obstacles. To ensure the feasibility and optimality of the obstacle avoidance path, both path safety and feasibility checks are added to screen the obstacle avoidance path. The safety checks in the second stage are similar to those in the first stage and will not be further elaborated here.

[0130] Step S860: evaluating the second path segment.

[0131] The specific evaluation method is the same as the first path evaluation shown in step S840 above, and will not be described in detail here.

[0132] Step S870: collision detection.

[0133] Figure 13 The principle of collision detection for the second obstacle avoidance path is shown, which mainly includes: performing collision detection on the second obstacle avoidance path, considering the dynamic planning process of obstacle avoidance, and detecting whether there is a collision on the obstacle avoidance path in real time during the obstacle avoidance process to avoid obstacle avoidance failure caused by the appearance of new obstacles (denoted as M in the figure) due to environmental changes. Figure 7 Teaching and reference Figure 13 , when the second obstacle avoidance path currently being executed is detected ( Figure 13 After a collision occurs, the position of the collision point is calculated, and the nearest point of the collision point on the global path is found to obtain its heading angle. Then, the above steps S810-S860 are repeated along the vertical direction to replan a new obstacle avoidance path ( Figure 13 In this way, this example can promptly discard old paths with collisions, allowing obstacle avoidance planning to respond to new environmental information and update the obstacle avoidance path, ensuring the safety and success rate of the obstacle avoidance process and ensuring that path planning always adapts to environmental changes.

[0134] The collision detection in step S870 essentially implements dynamic obstacle avoidance planning. This dynamic planning method triggers replanning based on collision detection of the real-time obstacle avoidance path. This replanning uses the original obstacle avoidance planning method, triggered by the possibility of a collision. This ensures that the planned obstacle avoidance path is always adaptable to environmental changes.

[0135] In summary, this example shows that the obstacle avoidance path planning method of the embodiment of the present application improves the automatic obstacle avoidance capability of the mining truck in the actual operating environment, and the obstacle avoidance success rate is higher. At the same time, the method of sampling candidate paths and adding path evaluation can screen out the optimal obstacle avoidance path, ensuring that the mining truck can better track and control, reduce the difficulty of path tracking, and thus improve control accuracy.

[0136] Example 2

[0137] Based on the same inventive concept as the obstacle avoidance path planning of the first embodiment, the second embodiment of the present application provides an obstacle avoidance path planning device for an intelligent driving vehicle, such as Figure 14 As shown, it includes: a first path planning module 100, which is used to plan a first obstacle avoidance path in response to sensing a current obstacle on the global path, wherein the expected end point of the first obstacle avoidance path is located within a first obstacle avoidance safety space relative to the current obstacle; and a second path planning module 200, which is used to plan a second obstacle avoidance path according to the perception result of re-obstacle perception when reaching the expected end point of the first obstacle avoidance path.

[0138] In a preferred embodiment, the first path planning module 100 adopts, for example, a polynomial curve planner, and the polynomial curve planner is configured to output multiple obstacle avoidance paths based on the starting point and multiple alternative end points of the first obstacle avoidance path, wherein the starting point of the first obstacle avoidance path is on the global path and makes the distance between the vehicle and the current obstacle meet a preset obstacle avoidance condition, and the alternative end points are several path points selected from the global path point in a direction perpendicular to the heading of the vehicle within the first obstacle avoidance safety space.

[0139] In a preferred embodiment, the second path planning module 200 adopts, for example, a Bezier curve planner, and the Bezier curve planner is configured to output corresponding multiple obstacle avoidance paths based on the starting point of the second obstacle avoidance path and multiple intermediate control points, wherein the starting point of the second obstacle avoidance path is the expected end point of the first obstacle avoidance path.

[0140] Regarding the selection of the candidate end point of the first obstacle circumventing path and the intermediate control point of the second obstacle circumventing path, reference may be made to the first embodiment, which will not be described in detail here.

[0141] In a more preferred embodiment, the obstacle avoidance path planning device may further include: an obstacle perception module 300 for performing obstacle perception in each path planning stage; and / or a path evaluation module 400 for performing path cost evaluation on the obstacle avoidance path outputted in each path planning stage.

[0142] For example, the obstacle sensing module 300 may be an obstacle detection device such as a laser radar, a camera, etc. The details of the path cost evaluation performed by the path evaluation module 400 may also refer to the first embodiment.

[0143] It should be noted that more implementation details and effects of the obstacle avoidance path planning device of Example 2 of the present application can be referred to Example 1, and will not be repeated here.

[0144] Example 3

[0145] Based on the first embodiment, the third embodiment of the present application provides a driving control method for an intelligent driving vehicle, such as Figure 15 As shown, it mainly includes the following steps S1510-S1520:

[0146] Step S1510: When a vehicle senses an obstacle on a global path, the vehicle is controlled to change lanes from the global path to a first obstacle avoidance path.

[0147] Step S1520: When the vehicle reaches the expected end point of the first obstacle bypassing path, the vehicle is controlled to change lanes from the first obstacle bypassing path to a second obstacle bypassing path, and then return to the global path through the second obstacle bypassing path.

[0148] The first obstacle avoidance path and the second obstacle avoidance path are generated based on the obstacle avoidance path planning method of the first embodiment or the obstacle avoidance path planning device of the second embodiment, which will not be described in detail here.

[0149] The driving control method of the third embodiment of the present application controls the vehicle to change lanes in response to detecting an obstacle on the global path to travel on the first obstacle avoidance path and the second obstacle avoidance path planned based on the above embodiment, so that the vehicle can not only use the first obstacle avoidance path to successfully avoid the current obstacle, but also use the second obstacle avoidance path to avoid obstacles in the perception blind spot, thereby ensuring the safe driving of the vehicle and assisting the vehicle to return to the global path in a timely manner.

[0150] The driving control method of embodiment three of the present application is executed by, for example, the autonomous driving domain controller of the vehicle, and the method of embodiment one or the module of embodiment two can be integrated into a functional module in the autonomous driving domain controller.

[0151] For more implementation details and effects of the driving control method of Example 3 of the present application, please refer to Example 1 or Example 2, and will not be repeated here.

[0152] Example 4

[0153] like Figure 16 As shown, embodiment 4 of the present application provides a control device for an intelligent driving vehicle, comprising: a memory configured to store instructions; and a processor configured to call the instructions from the memory and execute the obstacle avoidance path planning method described in embodiment 1 or the driving control method described in embodiment 3.

[0154] That is, the control device of the fourth embodiment of the present application can execute the obstacle avoidance path planning step of the first embodiment, and can also execute the vehicle driving control step of the third embodiment.

[0155] For more implementation details and effects of the control device of embodiment 4 of the present application, please refer to the other embodiments mentioned above and will not be repeated here.

[0156] Example 5

[0157] Embodiment 5 of the present application provides an intelligent driving vehicle, including the obstacle avoidance path planning device of embodiment 2 or the control device of embodiment 4.

[0158] The intelligent driving vehicle can be any of a smart car, smart construction machinery, smart agricultural machinery, and a mobile robot. For example, it can be a self-driving new energy vehicle among smart cars, a mining truck among smart construction machinery, a harvester among smart agricultural machinery, and a delivery robot or a sweeping robot among mobile robots.

[0159] For more implementation details and effects of the intelligent driving vehicle of Example 4 of the present application, please refer to the other embodiments mentioned above and will not be repeated here.

[0160] An embodiment of the present application also provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute any of the above-mentioned obstacle avoidance path planning methods or driving control methods.

[0161] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0162] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0163] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0164] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0165] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0166] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0167] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0168] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0169] The above are merely embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A method for planning an obstacle avoidance path for an intelligent driving vehicle, characterized in that: include: In response to sensing a current obstacle on the global path, planning a first obstacle avoidance path, wherein an expected end point of the first obstacle avoidance path is located within a first obstacle avoidance safety space relative to the current obstacle; and When the expected end point of the first obstacle circumventing path is reached, obstacle perception is performed again, and a second obstacle circumventing path is planned based on the perception result.

2. The obstacle avoidance path planning method according to claim 1, characterized in that: The process of planning the first obstacle avoidance path further includes determining the first obstacle avoidance safe space through the following steps: Determining a global path point on the global path that is closest to the current obstacle; Starting from the global waypoint, extending a plurality of waypoints at different headings along a first predetermined direction until the extended waypoints extend to a road boundary or another obstacle; and Based on the global path point and multiple path points extended therefrom, the first obstacle avoidance safety space relative to the current obstacle is constructed.

3. The obstacle avoidance path planning method according to claim 2, characterized in that: Planning the first obstacle avoidance path includes: In the first obstacle avoidance safety space, starting from the global path point, determining a plurality of candidate end points of the first obstacle avoidance path along a direction perpendicular to the heading of the vehicle; performing path planning based on the starting point of the first obstacle avoidance path and each determined alternative end point to obtain corresponding multiple obstacle avoidance paths, wherein the starting point of the first obstacle avoidance path is on the global path and the distance between the vehicle and the current obstacle satisfies a preset obstacle avoidance condition; and A path with the minimum path cost is selected from the multiple obstacle circumventing paths to serve as the first obstacle circumventing path, wherein the candidate endpoint corresponding to the path with the minimum path cost is the expected endpoint of the first obstacle circumventing path.

4. The obstacle avoidance path planning method according to claim 1, characterized in that: The re-performing obstacle perception includes: Starting from the current position of the vehicle, searching for the first obstacle point for the current obstacle along the vertical direction of the vehicle heading; Starting from the first obstacle point, traversing obstacle points along the heading of the current obstacle until the end point of the current obstacle is found; and Starting from the end point of the current obstacle, the obstacle point search is continued for a preset number of times. If a new obstacle point is found, a perception result indicating the existence of a new obstacle is obtained.

5. The obstacle avoidance path planning method according to claim 4, characterized in that: Planning the second obstacle avoidance path includes constructing a second obstacle avoidance safe space through the following steps: Determine a final end point, wherein if the new obstacle exists, the end point of the new obstacle is the final end point, otherwise the end point of the current obstacle is the final end point; as well as Starting from the final end point, a second obstacle avoidance safe space is constructed along a second preset direction.

6. The obstacle avoidance path planning method according to claim 5, characterized in that: After constructing the second obstacle avoidance safe space, planning the second obstacle avoidance path further includes: Determining, within the second obstacle avoidance safety space, a plurality of intermediate control points along the second preset direction along the obstacle avoidance path; performing path planning based on the starting point of the second obstacle avoidance path and the plurality of intermediate control points of the obstacle avoidance paths to obtain a corresponding plurality of obstacle avoidance paths, wherein the expected end point of the first obstacle avoidance path is used as the starting point of the second obstacle avoidance path; and A path with the minimum path cost is selected from the multiple obstacle circumventing paths to serve as the second obstacle circumventing path.

7. The obstacle avoidance path planning method according to claim 5, characterized in that: The second preset direction is a direction perpendicular to the heading of a reference path point, and the reference path point is a path point on the global path that is closest to the final end point.

8. The obstacle avoidance path planning method according to any one of claims 1 to 7, characterized in that: After planning the second obstacle avoidance path, the obstacle avoidance path planning method further includes: performing collision risk detection on the second obstacle avoidance path; In the event of a collision risk, detecting a collision point and determining a path point on the global path that is closest to the collision point; and A new obstacle avoidance safety space is constructed based on a direction perpendicular to the heading of the determined path point to re-plan the obstacle avoidance path.

9. A driving control method for an intelligent driving vehicle, characterized in that: include: When the vehicle senses an obstacle on the global path, controlling the vehicle to change lanes from the global path to a first obstacle avoidance path; as well as When the vehicle reaches an expected end point of the first obstacle bypassing path, controlling the vehicle to change lanes from the first obstacle bypassing path to a second obstacle bypassing path, and then returning to the global path via the second obstacle bypassing path; The first obstacle avoidance path and the second obstacle avoidance path are generated based on the obstacle avoidance path planning method according to any one of claims 1 to 8.

10. A control device for an intelligent driving vehicle, characterized in that: include: a memory configured to store instructions; as well as A processor is configured to call the instructions from the memory and to implement the obstacle avoidance path planning method described in any one of claims 1 to 8 or the driving control method described in claim 9 when executing the instructions.

11. An intelligent driving vehicle, characterized in that: Comprising the control device according to claim 10.

12. A machine-readable storage medium, characterized in that The machine-readable storage medium stores instructions for enabling a machine to execute the obstacle avoidance path planning method according to any one of claims 1 to 8 or the driving control method according to claim 9.

Images