Path planning methods, devices, vehicles and storage media
By using a Cartesian coordinate system to describe the path and obstacle boundaries in path planning, the problem of insufficient accuracy in the Frenet coordinate system is solved, and higher-precision path planning is achieved.
Patent Information
- Application Number
- CN202210515455.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-05-11
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2042-05-11
AI Technical Summary
Existing path planning methods based on the Frenet coordinate system have low accuracy due to their one-dimensionality, and cannot accurately identify trajectory points on the road, resulting in insufficient path planning accuracy.
The Cartesian coordinate system is used to describe the path boundary and obstacle boundary information. The target driving space is determined by the Cartesian coordinate system, and the planned path is determined based on this space, preset constraints and optimization objectives.
This improves the accuracy of path planning, ensuring that vehicles can uniquely identify trajectory points on the road, avoiding the problem of multiple trajectory points corresponding to a single coordinate, and thus enhancing the accuracy of path planning.
Smart Images

Figure CN117091615B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of autonomous driving technology, and more specifically, to a method for tracking vehicle driving trajectory, a control system, a vehicle, and a storage medium. Background Technology
[0002] The main function of path planning is to enable vehicles to find a collision-free and safe path from the starting point to the destination within a specified area, which plays a very important role in the field of autonomous driving technology.
[0003] The path planning method provided by the relevant technology is as follows: First, the vehicle and environment are mapped from the Cartesian coordinate system to the Frenet coordinate system. Then, the road boundary is determined based on the Frenet coordinate system. Using the linear dynamic programming method, the optimization objectives are the minimum trajectory slope, the minimum curvature, the comfortable riding experience, and the shortest distance to the path obtained by the linear dynamic programming. The position of the connection point of the piecewise fifth-order polynomial, the first derivative, the second derivative, and the third derivative are used as equality constraints. The constraints of the natural road boundary and the obstacle boundary are used as inequality constraints. A mathematical model for planning the path is established, and the optimal solution of the above mathematical model is obtained as the planned path.
[0004] However, in related technologies, the road boundaries determined based on the Frenet coordinate system are in the form of line segments (one-dimensional case), which cannot uniquely identify the trajectory points on the road, thus leading to low accuracy in path planning. Summary of the Invention
[0005] This application provides a route planning method, apparatus, vehicle, and storage medium.
[0006] In a first aspect, some embodiments of this application provide a path planning method, which includes: acquiring multiple sampling points; determining path boundary information corresponding to the multiple sampling points in a Cartesian coordinate system, wherein the path boundary information corresponding to the sampling points includes a first planar region determined based on the sampling points; determining a target obstacle; determining obstacle boundary information corresponding to the target obstacle in a Cartesian coordinate system, wherein the obstacle boundary information corresponding to the target obstacle includes a second planar region determined based on the target obstacle; determining a target driving space based on the path boundary information and obstacle boundary information corresponding to the multiple sampling points; and determining a planned path based on the target driving space, preset constraints, and preset optimization objectives.
[0007] Secondly, some embodiments of this application provide a path planning device, which includes: a sampling point acquisition module for acquiring multiple sampling points; a first determination module for determining path boundary information corresponding to the multiple sampling points in a Cartesian coordinate system, wherein the path boundary information corresponding to the sampling points includes a first planar region determined based on the sampling points; an obstacle determination module for determining a target obstacle; a second determination module for determining obstacle boundary information corresponding to the target obstacle in a Cartesian coordinate system, wherein the obstacle boundary information corresponding to the target obstacle includes a second planar region determined based on the target obstacle; a driving space determination module for determining a target driving space based on the path boundary information and obstacle boundary information corresponding to the multiple sampling points; and a path planning module for determining a planned path based on the target driving space, preset constraints, and preset optimization objectives.
[0008] Thirdly, some embodiments of this application also provide a vehicle, which includes one or more processors, a memory, and one or more application programs. The one or more application programs are stored in the memory and configured to be executed by the one or more processors, and are configured to perform the vehicle routing method described above.
[0009] Fourthly, embodiments of this application also provide a computer-readable storage medium storing computer program instructions. These computer program instructions can be invoked by a processor to execute the aforementioned path planning method.
[0010] Fifthly, embodiments of this application also provide a computer program product that, when executed, implements the path planning method described above.
[0011] This application provides a path planning method that uses a Cartesian coordinate system to describe the path boundary information determined based on each sampling point, and also uses a Cartesian coordinate system to describe the obstacle boundary information determined based on the target obstacle. Based on the aforementioned path boundary information and obstacle boundary information, a target driving space that is suitable for vehicle travel and will not cause a collision is determined. Finally, based on the target driving space, preset constraints, and preset optimization objectives, a planned path is determined. Since the Cartesian coordinate system is two-dimensional, any trajectory point on the road can be uniquely represented by a combination of horizontal and vertical coordinates. Compared to the Frenet coordinate system, which only uses a single coordinate value to describe road boundary information, the embodiments of this application can avoid the situation where one coordinate corresponds to multiple trajectory points, thus improving the accuracy of path planning. Attached Figure Description
[0012] To more clearly illustrate the technical solutions in the embodiments of this application, 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 this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0013] Figure 1 This paper presents a comparison diagram of trajectory points represented in Frenet coordinate system and Cartesian coordinate system, as provided in an embodiment of this application.
[0014] Figure 2 A schematic diagram of a vehicle provided in an embodiment of this application is shown.
[0015] Figure 3 A flowchart illustrating a path planning method provided in an embodiment of this application is shown.
[0016] Figure 4 This is a schematic diagram of obtaining sampling points provided in an embodiment of this application.
[0017] Figure 5 This is a schematic diagram of determining the target driving space provided in an embodiment of this application.
[0018] Figure 6 This is a schematic diagram illustrating the determination of the planned path provided in an embodiment of this application.
[0019] Figure 7 A flowchart illustrating another path planning method provided in an embodiment of this application is shown.
[0020] Figure 8 A block diagram of a path planning device provided in an embodiment of this application is shown.
[0021] Figure 9 A block diagram of a vehicle provided in an embodiment of this application is shown.
[0022] Figure 10 A block diagram of a computer-readable storage medium provided in an embodiment of this application is shown. Detailed Implementation
[0023] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.
[0024] To enable those skilled in the art to better understand the solutions of this application, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.
[0025] The inventors discovered that when determining road boundaries and performing subsequent path planning in the Frenet coordinate system, because the Frenet coordinate system is one-dimensional, there may be a situation where one coordinate corresponds to multiple trajectory points on the road. For example, see [link to relevant documentation]. Figure 1 In part (a) of the diagram, in the U-turn scenario, trajectory points A and B are both represented by coordinates (x) in the Frenet coordinate system. Since the trajectory points on the road cannot be accurately represented, the path planning methods provided by the existing technology have the problem of low accuracy.
[0026] To address the aforementioned problems, the inventors designed a novel path planning method. This method uses a Cartesian coordinate system to describe the path boundary information determined by each sampling point, and also uses a Cartesian coordinate system to describe the obstacle boundary information determined by the target obstacle. Based on these path and obstacle boundary information, a target driving space suitable for vehicle travel without collisions is determined. Finally, based on the target driving space, preset constraints, and preset optimization objectives, the planned path is determined. Because the Cartesian coordinate system is two-dimensional, any trajectory point on the road can be uniquely represented by a combination of its horizontal and vertical coordinates. For example, see [link to relevant documentation]. Figure 1 In part (b) of the diagram, in the U-turn scenario, trajectory point A is represented by (x, y1) in the Cartesian coordinate system, and trajectory point B is represented by (x, y2) in the Cartesian coordinate system. Compared to the Frenet coordinate system, which uses only one coordinate value to describe road boundary information, the embodiments of this application can avoid the situation where one coordinate corresponds to multiple trajectory points, thus improving the accuracy of path planning.
[0027] To facilitate a detailed explanation of the solution presented in this application, the application environment of the example in this application will be described below with reference to the accompanying drawings. Please refer to... Figure 2 , Figure 2 This example illustrates the application environment of the path planning method provided in this application embodiment. The method is applied to a vehicle 200, which refers to a means of transportation driven or towed by a power device for passenger transport or goods transport, including but not limited to cars, minibuses, buses, etc. The vehicle 200 includes a control system and a vehicle execution system.
[0028] In this embodiment, the control system 220 is used to implement path planning. Optionally, the steps of the control system 220 to implement path planning are as follows: the control system 220 determines the driving space available for the vehicle to travel in the Cartesian coordinate system, and obtains preset constraints and preset optimization objectives. Based on the above three elements, a mathematical model for solving the planned path is constructed, and then the planned path is determined by solving the above mathematical model.
[0029] Vehicle actuators are used to drive along a planned path. Vehicle actuators can be vehicle chassis actuators, including Electrical Power Steering (EPS), Electronic Stability Program (ESP), and Vehicle Control Unit (VCU), etc.
[0030] In some embodiments, the vehicle 200 further includes a positioning device for determining the real-time location of the vehicle 200. Optionally, the vehicle 200 obtains its current location via the positioning device as its starting point for travel. Optionally, the positioning device includes a Global Navigation Satellite System (GNSS). GNSS is a high-precision radio navigation positioning system based on artificial Earth satellites, capable of providing accurate real-time location information to the vehicle 200 anywhere in the world and in near-Earth space.
[0031] In some embodiments, the vehicle 200 further includes a touchscreen for allowing a user to input one or both of the preceding and following items of the travel destination or travel origin. In some embodiments, the touchscreen displays a user interface with route planning functionality, which includes an electronic map on which the user selects one or both of the preceding and following items of the travel destination or travel origin. In other embodiments, the user interface includes a start point input box and a destination input box, wherein the start point input box is used to input the travel origin, and the destination input box is used to input the travel destination.
[0032] like Figure 3 As shown, this application provides a flowchart of a path planning method. The method includes the following steps.
[0033] Step 301: Obtain multiple sampling points.
[0034] The vehicle samples along a reference path at a preset sampling distance, resulting in multiple sampling points. This reference path can be the lane centerline of a road, the vehicle's historical driving trajectory, or the historical driving trajectory of other vehicles; this embodiment does not limit the specific path.
[0035] The preset sampling distance is set in advance based on experiments or experience; for example, the preset sampling distance is 5 meters. The maximum sampling distance is set according to the accuracy requirements of the planned path and the vehicle's computing power; for example, the maximum sampling distance is 75 meters. The number of sampling points is determined based on the maximum sampling distance and the preset sampling distance. Specifically, the number of sampling points is the ratio between the maximum sampling distance and the preset sampling distance. Referring to the example above, the number of sampling points is 75 / 5 = 25.
[0036] The process for determining multiple sampling points for the vehicle is as follows: The vehicle determines the starting point of the reference route as the first sampling point. Then, based on the previous sampling point, it draws an arc with a radius equal to a preset sampling distance. The intersection of this arc and the reference path is determined as the next sampling point. It should be noted that if a polyline exists on the reference path, and the distance between the target sampling point and the endpoint of the polyline is less than the preset sampling distance, then the endpoint of the polyline is used as the sampling point following the target sampling point. Figure 4 The diagram illustrates the acquisition of sampling points provided in the embodiments of this application. After determining the first sampling point on the polyline path, a circle is drawn with a preset sampling distance; the intersection of the circle and the polyline path is the next sampling point. Figure 4 The black dots in the image are the sampling points.
[0037] In some embodiments, the vehicle acquires multiple sampling points after receiving a route planning instruction. Optionally, the user interface displayed on the vehicle's central control screen or in-vehicle navigation system includes an electronic map, a starting point input box, a destination input box, and a start navigation control. If a trigger instruction for the start navigation control is received after determining the starting point and destination, the step of acquiring multiple sampling points is executed. It should be noted that the method for determining the starting point includes at least one of the following: acquiring the starting point selected on the electronic map, acquiring the starting point entered in the starting point input box, or acquiring the vehicle's current location as the starting point using a positioning device. The method for determining the destination includes at least one of the following: acquiring the destination selected on the electronic map, or acquiring the destination entered in the destination input box.
[0038] When the route between the starting point and the destination is long, vehicles typically divide the route into multiple segments and then perform path planning for each segment separately, thus achieving more refined path planning. In this case, given that the planned path for the i-th segment is determined, multiple sampling points are obtained from the (i+1)-th segment to determine the planned path for the (i+1)-th segment, where i is a positive integer.
[0039] Step 302: Obtain the path boundary information corresponding to multiple sampling points in the Cartesian coordinate system.
[0040] The Cartesian coordinate system is a two-dimensional coordinate system. The path boundary information corresponding to the sampling point in the Cartesian coordinate system includes two dimensions: horizontal and vertical. The horizontal and vertical dimensions are perpendicular to each other, and the vertical dimension represents the road direction. This application does not limit the location of the origin of the Cartesian coordinate system; it only uses the location of the first sampling point as an example for illustration.
[0041] The path boundary information corresponding to the sampling point includes a first planar region determined based on the sampling point. This application embodiment does not limit the shape of the first planar region; it only uses a rectangular region as an example for illustration.
[0042] In the scenario where the first planar region is rectangular, the path boundary information corresponding to the sampling point may include the coordinates of the center point of the first planar region in Cartesian coordinates, the length of the first planar region, and its width. In other possible implementations, the path boundary information corresponding to the sampling point may also include the coordinates of at least two vertices of the first planar region in Cartesian coordinates. It should be noted that if the path boundary information corresponding to the sampling point includes the coordinates of two vertices of the first planar region in Cartesian coordinates, these two vertices are not adjacent. The process of determining the path boundary information corresponding to the sampling point will be described in the following embodiments.
[0043] In this embodiment, the path boundary information is described by two dimensions, horizontal and vertical. Any trajectory point on the road can be uniquely represented by a combination of horizontal and vertical coordinates. Compared with the Frenet coordinate system, which describes the road boundary information by only one coordinate value, this embodiment can avoid the situation where one coordinate corresponds to multiple trajectory points, thus improving the accuracy of path planning.
[0044] Step 303: Identify the target obstacle.
[0045] The vehicle uses obstacle detection devices to detect obstacles on a reference path, and then selects the target obstacle from the detected obstacles. Obstacle detection devices include, but are not limited to, lidar, surround-view cameras, etc.
[0046] In some embodiments, the vehicle acquires the location information of each obstacle. For each obstacle, it acquires the distance between the obstacle and each sampling point. If the minimum distance is less than or equal to a preset distance, the obstacle is identified as a target obstacle. In other embodiments, the vehicle determines obstacle boundary information based on the location information of each obstacle. Then, obstacles where there is an overlapping area between the obstacle boundary information and the path boundary information corresponding to each sampling point, and where the area of the overlapping area is greater than a preset area, are identified as target obstacles. The preset distance and preset area are set according to actual needs, and this application embodiment does not limit them.
[0047] Step 304: Determine the obstacle boundary information corresponding to the target obstacle in the Cartesian coordinate system.
[0048] It should be noted that, in this embodiment, the sampling point and the target obstacle are in the same coordinate system to simplify the subsequent calculation process. The coordinate system of the target obstacle includes two dimensions: horizontal and vertical. The horizontal and vertical dimensions are perpendicular to each other, with the vertical direction representing the road direction. The origin is the location of the first sampling point.
[0049] The boundary information corresponding to the target obstacle includes a second planar region determined based on the target obstacle. The shape of the second planar region is not limited in this embodiment; however, a rectangular region is used as an example for illustration.
[0050] When the second planar region is rectangular, the obstacle boundary information corresponding to the target obstacle may include the coordinates of the center point of the second planar region in Cartesian coordinates, the length of the second planar region, and its width. In other possible implementations, the obstacle boundary information corresponding to the target obstacle may also include the coordinates of at least two vertices of the second planar region in Cartesian coordinates. It should be noted that if the obstacle boundary information corresponding to the target obstacle includes the coordinates of two vertices of the first planar region in Cartesian coordinates, these two vertices are not adjacent. The process of determining the obstacle boundary information corresponding to the target obstacle will be described in the following embodiments.
[0051] Step 305: Determine the target driving space based on the path boundary information and obstacle boundary information corresponding to multiple sampling points.
[0052] The target driving space refers to the driving space that meets the vehicle's driving needs and will not collide with the target obstacle. The process of determining the target driving space will be described in the following examples.
[0053] Step 306: Determine the planned path based on the target driving space, preset constraints, and preset optimization objectives.
[0054] The preset constraints refer to the constraints that the planned path is expected to satisfy. The preset optimization objective refers to the goal that is expected to be achieved while driving on the planned path. The vehicle constructs a planned path model based on the target driving space, the preset constraints, and the preset optimization objective, and then solves the planned path model to obtain the planned path. Optionally, the vehicle solves the path planning model using the CVXOPT optimization library to obtain one or more optimal solutions, each optimal solution corresponding to a planned path.
[0055] It should be noted that, due to the extremely long time required to solve the path planning model, an additional thread is started to complete the solution of the path planning model in order to avoid the main thread from being blocked. This thread can be started after receiving the path planning instruction or after the path planning model has been built; this embodiment of the application does not limit this.
[0056] In summary, the technical solution provided in this application uses a Cartesian coordinate system to describe the path boundary information determined based on each sampling point, and also uses a Cartesian coordinate system to describe the obstacle boundary information determined based on the target obstacle. Based on the aforementioned path boundary information and obstacle boundary information, a target driving space that is suitable for vehicle travel and will not cause a collision is determined. Finally, based on the target driving space, preset constraints, and preset optimization objectives, a planned path is determined. Since the Cartesian coordinate system is two-dimensional, any trajectory point on the road can be uniquely represented by a combination of horizontal and vertical coordinates. This application embodiment can avoid the situation where one coordinate corresponds to multiple trajectory points, thus improving the accuracy of path planning.
[0057] The process of determining the path boundary information corresponding to the sampling points is described below. Based on Figure 3 In the optional embodiments provided by the examples, step 302 can be replaced by the following sub-steps.
[0058] Step 302a: For the i-th sampling point, determine the first projection point of the i-th sampling point on the first road boundary and the second projection point of the i-th sampling point on the second road boundary, where i is a positive integer.
[0059] When there is one lane in a road, the first road boundary and the second road boundary are the extensions of the two sides of the road. When there are multiple lanes in a road, the first road boundary is the extension of one side of the road, and the second road boundary is a lane line; the second road boundary is the extension of one side of the road, and the first road boundary is a lane line; or, both the first road boundary and the second road boundary are lane lines.
[0060] The first projection point of the i-th sampling point on the first road boundary refers to a point on the first road boundary where the line connecting the i-th sampling point and the first road boundary is perpendicular to the first road boundary. The second projection point of the i-th sampling point on the second road boundary refers to a point on the second road boundary where the line connecting the i-th sampling point and the second road boundary is perpendicular to the second road boundary.
[0061] Optionally, when i is 1, for the i-th sampling point, the vehicle starts searching from a set starting point on the first road boundary to obtain the first projection point of the first sampling point on the first road boundary, and starts searching from a set starting point on the second road boundary to obtain the second projection point of the first sampling point on the second road boundary. The set starting points of the first and second road boundaries are set based on experiments or experience, and this embodiment does not limit them.
[0062] Optionally, when i is greater than 1, for the i-th sampling point, the search begins at the first projection point of the (i-1)-th sampling point on the first road boundary, and a search is performed on the first road boundary to obtain the first projection point of the i-th sampling point on the first road boundary; similarly, the search begins at the second projection point of the (i-1)-th sampling point on the second road boundary, and a search is performed on the second road boundary to obtain the second projection point of the i-th sampling point on the second road boundary. This method avoids starting the search from a predetermined starting point on either the first or second road boundary, thus improving the search efficiency for the projection points (first and second projection points).
[0063] Step 302b: Determine the center point corresponding to the i-th sampling point based on the i-th sampling point, the first projection point, and the second projection point.
[0064] Optionally, the vehicle first determines the midpoint of the line connecting the first and second projection points, and then aligns this midpoint with the i-th sampling point. The aligned point is also the center point corresponding to the i-th sampling point. Aligning the midpoint with the i-th sampling point means making the line connecting the midpoint and the i-th sampling point parallel to a specified direction, which is perpendicular to the road direction. If the midpoint and the i-th sampling point are in the same rectangular coordinate system, and the vertical axis of this coordinate system is the road direction and the horizontal axis is perpendicular to the road direction, then aligning the midpoint with the i-th sampling point means setting the vertical coordinate of the midpoint to the vertical coordinate of the i-th sampling point.
[0065] Step 302c: Obtain the first constraint distance in the first direction and the second constraint distance in the second direction.
[0066] The first direction and the second direction are perpendicular to each other. Optionally, the first direction is the road direction, and the second direction is perpendicular to the road direction. The first constraint distance can be preset; for example, the first constraint distance is 5 meters. The second preset distance is set according to the actual width of the vehicle; for example, the second constraint distance is 1.3 times the vehicle width. For example, if the vehicle width is 1.5 meters, the second constraint distance is 1.3 * 1.5 = 1.95 meters.
[0067] Step 302d: Based on the center point corresponding to the i-th sampling point, the first constraint distance and the second constraint distance, determine the path boundary information corresponding to the i-th sampling point.
[0068] Optionally, the vehicle takes the center point corresponding to the i-th sampling point as the center point of the first planar region, the first constraint distance is the length direction of the first planar region, and the second constraint distance is the width direction of the second planar region, and finally determines the first planar region.
[0069] The process of determining the obstacle boundary information corresponding to the target obstacle is similar to that of determining the path boundary information corresponding to the sampling point. Based on... Figure 3 In the optional embodiments provided by the illustrated embodiment, step 304 can be replaced by the following sub-steps.
[0070] Step 304a: For the target obstacle, determine the third projection point of the target obstacle on the first road boundary and the fourth projection point of the target obstacle on the second road boundary.
[0071] The vehicle first determines the key points of the target obstacle, then uses the projection point of the key point on the first road boundary as the third projection point, and the projection point of the key point on the second road boundary as the fourth projection point. The key points of the target obstacle can be the center point or any vertex of the target obstacle; this embodiment does not limit this.
[0072] In some embodiments, for a target obstacle, the vehicle first determines the sampling point with the minimum distance to the target obstacle. Then, based on the sampling point, a first projection point on the first road boundary is searched in two directions for a third projection point, and based on the sampling point, a second projection point on the second road boundary is searched in two directions for a fourth projection point. The two directions refer to the road direction and the opposite direction. This method avoids starting the search from a predetermined starting point on the first or second road boundary, improving the search efficiency of the projection points (the third and fourth projection points).
[0073] Step 304b: Determine the center point of the target obstacle based on the key points, the third projection point, and the fourth projection point of the target obstacle.
[0074] Optionally, the vehicle first determines the midpoint of the line connecting the third and fourth projection points, and then aligns this midpoint with the key point of the target obstacle. The aligned point is the center point of the target obstacle. The alignment method can be found in the explanation of step 302b, which will not be repeated here.
[0075] Step 304c: Obtain the third constraint distance in the first direction and the fourth constraint distance in the second direction.
[0076] The third and fourth constraint distances can be preset. For example, the third constraint distance is 5 meters, and the fourth constraint distance is 1.0 times the vehicle width. For instance, if the vehicle width is 1.5 meters, the fourth constraint distance is 1.3 * 1.0 = 1.3 meters.
[0077] Step 304d: Based on the center point corresponding to the target obstacle, the third constraint distance and the fourth constraint distance, determine the obstacle boundary information corresponding to the target obstacle.
[0078] Optionally, the vehicle uses the center point corresponding to the target obstacle as the center point of the second planar region, the third constraint distance as the length direction of the second planar region, and the fourth constraint distance as the width direction of the second planar region to finally determine the second planar region.
[0079] The process of determining the path boundary information corresponding to the sampling points is described below. Based on Figure 3 In the optional embodiments provided by the illustrated embodiment, step 305 can be replaced by the following sub-steps.
[0080] Step 305a: Based on the path boundary information and obstacle boundary information corresponding to multiple sampling points, determine m drivable areas, where m is an integer greater than 2.
[0081] If there is no overlap between the path boundary information and obstacle boundary information corresponding to the sampling point, the path boundary information corresponding to the sampling point is determined as a drivable area.
[0082] If there is an overlapping area between the path boundary information and obstacle boundary information corresponding to the sampling point, and the area of the overlapping area is smaller than the preset area, the non-overlapping area between the path boundary information and obstacle boundary information corresponding to the sampling point is determined as the drivable area.
[0083] Step 305b: Based on m drivable areas and a pre-set reference line, determine n candidate drivable areas, where n is an integer greater than 1 and less than m.
[0084] After determining m drivable zones, the vehicle first establishes a reference line. Optionally, the vehicle can use the centerline of the first drivable zone as the reference line, or it can use the straight line that crosses the most drivable zones as the reference line.
[0085] After the reference line is determined, the vehicle selects n candidate driving areas from m drivable areas, with the vehicle prioritizing the drivable areas that are close to the reference line as candidate driving areas.
[0086] In this embodiment, the drivable space is formed into a two-dimensional array in the longitudinal and transverse directions. Based on the array, a path search is established with the planar area as the basic node unit. While ensuring the path length, the path is kept as close to the reference line as possible, which can avoid the situation where obstacles have to be bypassed in the large space when they are on the lane change line.
[0087] Step 305c: In the case where the candidate driving regions adjacent to each other in the target direction among the m candidate driving regions all satisfy the communication condition, the plurality of candidate driving regions are determined as the target driving space.
[0088] The target direction is the road direction. In some embodiments, the vehicle extends the candidate driving regions by a preset length in the target direction, and then performs connectivity determination. Specifically, the vehicle solves the inequality linear programming models corresponding to the 8 edges of two adjacent candidate driving regions. If the inequality linear programming models have solutions, it indicates that the two adjacent candidate driving regions satisfy the communication condition; if the inequality linear programming models have no solutions, it indicates that the two adjacent candidate driving regions do not satisfy the communication condition.
[0089] Reference Figure 5 , Figure 5 The rectangular boxes therein are drivable regions, wherein the solid-line rectangular boxes are candidate driving regions selected according to the reference line. The arrows between adjacent rectangular boxes in the road direction are used to indicate that the adjacent rectangular boxes satisfy the communication condition, and the plurality of solid-line rectangular boxes described above constitute the target driving space.
[0090] Reference Figure 6 , which shows a schematic diagram of path planning provided by an embodiment of the present application. After the target driving space is determined, one path point is determined in each candidate driving region in the target driving space, and the path points of each candidate driving region constitute a planned path. Figure 6 , each path point connected by a dashed arrow constitutes the planned path.
[0091] The preset constraint conditions are described below. In some embodiments, the preset constraint conditions include at least one of the following: a boundary constraint condition, a start point constraint condition, an end point constraint condition, and a continuity constraint condition.
[0092] In some embodiments, the vehicle determines the boundary constraint condition based on the path boundary information respectively corresponding to the plurality of sampling points.
[0093] Optionally, the vehicle establishes a Cartesian rectangular coordinate system with the first sampling point as the origin, the longitudinal axis as the road direction, and the transverse axis perpendicular to the road direction, and then establishes an inequality based on the two parallel edges (parallel to the road direction) of the path boundary information corresponding to the sampling point and the above Cartesian rectangular coordinate system. For example, if the abscissa of points on the first parallel edge of the two parallel edges is x1, and the abscissa of points on the second parallel edge of the two parallel edges is -x1, the determined inequality is -x1 < x < x1, which indicates that the abscissa of the trajectory point on the planned path should be greater than -x1 and less than x1.
[0094] Optionally, the vehicle can also establish piecewise inequalities based on the target driving space and the aforementioned Cartesian coordinate system. Specifically, the vehicle obtains two parallel edges (parallel to the road direction) for each candidate driving region in the target driving space, and establishes piecewise inequalities based on the parallel edges of each candidate driving region and the aforementioned Cartesian coordinate system. For example, the x-coordinate of the points on the first parallel edge of the first to fifth candidate regions is x2, the x-coordinate of the points on the second parallel edge of the first to fifth candidate regions is -x2, the y-coordinate of the points on the third parallel edge of the fifth candidate region (perpendicular to the road direction and far from the origin) is y1, the x-coordinate of the points on the first parallel edge of the sixth candidate region is x3, the x-coordinate of the points on the second parallel edge of the sixth candidate region is 0, and the y-coordinate of the points on the third parallel edge of the sixth candidate region (perpendicular to the road direction and far from the origin) is y2. Then the constructed piecewise inequalities are as follows:
[0095]
[0096] In some embodiments, the vehicle determines the starting point constraint in the preset constraints based on the vehicle's current position or the starting position in the historical driving trajectory.
[0097] The starting point constraints include constraints on the starting position, the vehicle's heading at the starting position, and the vehicle's curvature at the starting position. The starting position constraint is represented by the function value, the vehicle's heading constraint is represented by the first derivative of the function, and the vehicle's curvature constraint is represented by the second derivative of the function.
[0098] Optionally, if the vehicle does not have a historical driving trajectory or the deviation from the historical driving trajectory is large, the starting point constraint is determined based on the vehicle's current position. When the starting point constraint is determined based on the vehicle's current position, the unitized direction vector should be used as the constraint for the vehicle's heading at the starting position. In this case, the vehicle first uses the steering wheel angle of the vehicle model to calculate the current curvature, then constructs a parabola at a predetermined position with the current curvature and a first derivative of 1. The parabola is then rotated and translated so that its endpoints coincide with the vehicle's current position and direction. Finally, the second derivative of the function is calculated.
[0099] In some embodiments, the vehicle determines the endpoint constraint among preset constraints based on the road direction and the path direction of the planned path.
[0100] Optionally, the endpoint constraints include constraints on the endpoint position and constraints on the vehicle's heading at the endpoint position. The endpoint position constraints are represented by the function value of the function, and the vehicle's heading constraints at the endpoint position are represented by the first derivative of the function.
[0101] The constraints at the destination position must ensure that the function value is exactly the same as the destination of the path. The constraints on the vehicle's heading at the destination position must ensure that the first derivative of the function is a normalized forward direction vector, that is, the path direction of the planned path is the same as the road direction.
[0102] In some embodiments, the continuity constraint condition in the preset constraint conditions is determined based on the parameter information of the curve connection point of two adjacent curves.
[0103] Optionally, the vehicle determines multiple curves based on multiple sampling points. The continuity constraints that adjacent curves must satisfy include at least one of the following: the endpoint of the i-th curve coincides with the starting point of the (i+1)-th curve; the endpoint of the i-th curve and the starting point of the (i+1)-th curve have the same direction; and the endpoint of the i-th curve and the starting point of the (i+1)-th curve have the same curvature. In this embodiment, with a maximum sampling distance of 75 meters, the segmented curves can be set to 3, each covering 25 meters.
[0104] Specifically, the endpoint of the i-th curve coincides with the starting point of the (i+1)-th curve if and only if the function values of the i-th and (i+1)-th curves are the same at the curve connection point. The endpoint of the i-th curve and the starting point of the (i+1)-th curve have the same direction if and only if their first derivatives are the same at the curve connection point. The endpoint of the i-th curve and the starting point of the (i+1)-th curve have the same second derivatives at the curve connection point. Furthermore, to ensure continuous steering wheel changes at the curve connection point of adjacent curves, the following constraint can be set: the third derivatives of the i-th and (i+1)-th curves are the same at the curve connection point.
[0105] The preset constraints are described below. In some embodiments, the preset optimization objectives include at least one of the following: the planned path is close to a preset reference line, the planned path is close to the road direction, the steering wheel control angle of the vehicle when driving on the planned path is less than a preset angle, and the number of steering wheel controls when driving on the planned path is less than a preset number.
[0106] The planned path closely follows the road direction to mitigate the path correction problem caused by understeer when turning or oversteer when changing lanes. This manifests as a wavy path after turning or changing lanes, with the amplitude gradually decreasing as the straight-line distance increases.
[0107] Optionally, the vehicle measures whether the planned path closely follows the pre-set reference line by the square of the difference between the path point and the reference point. The smaller the square of the difference between the path point and the reference point, the closer the planned path is to the pre-set reference line; the larger the square of the difference between the path point and the reference point, the greater the deviation of the planned path from the pre-set reference line.
[0108] Optionally, the vehicle measures whether the planned path closely follows the road direction by the square of the product of a first vector representing the first derivative of a waypoint and a second vector. The second direction is a unit vector along the road direction. The smaller the square of the product of the first and second vectors, the closer the planned path is to the road direction. The larger the square of the product of the first and second vectors, the greater the deviation of the planned path from the road direction.
[0109] Optionally, the vehicle uses the squared-second derivative integral to measure whether the steering wheel control angle is less than a preset angle when traveling along the planned path. The steering wheel control angle when traveling along the planned path refers to the rotation angle of the steering wheel. Specifically, the squared-second derivative integral is positively correlated with the steering wheel control angle; the smaller the squared-second derivative integral, the smaller the steering wheel control angle. The larger the squared-second derivative integral, the larger the steering wheel control angle.
[0110] Optionally, the vehicle uses the third-order squared integral to measure whether the number of steering wheel controls while traveling along the planned path is less than a preset number. The number of steering wheel controls while traveling along the planned path refers to the number of times the steering wheel is turned. Specifically, the third-order squared integral is positively correlated with the number of steering wheel controls; the smaller the third-order squared integral, the fewer the number of steering wheel controls. The larger the third-order squared integral, the more the number of steering wheel controls.
[0111] Optionally, the vehicle performs a weighted sum of the four objectives to determine the final optimization objective. It should be noted that the weighting coefficients for the four optimization objectives differ across different scenarios. In one example, in a U-turn scenario, the optimization objective of planning a path close to a pre-set reference line has a weight of 1; the optimization objective of planning a path close to the road direction has a weight of 100; the optimization objective of the vehicle's steering wheel control angle being less than a preset angle while traveling along the planned path has a weight of 1; and the optimization objective of the vehicle's steering wheel control number being less than a preset number while traveling along the planned path has a weight of 1. In a road-following scenario, the optimization objective of planning a path close to a pre-set reference line has a weight of 5; the optimization objective of planning a path close to the road direction has a weight of 100; the optimization objective of the vehicle's steering wheel control angle being less than a preset angle while traveling along the planned path has a weight of 1; and the optimization objective of the vehicle's steering wheel control number being less than a preset number while traveling along the planned path has a weight of 1.
[0112] Please see Figure 7 The diagram illustrates a flowchart of a path planning method provided in one embodiment of this application. The method includes:
[0113] Step 701: Obtain multiple sampling points.
[0114] Step 702: In the Cartesian coordinate system, determine the path boundary information corresponding to multiple sampling points respectively. The path boundary information corresponding to the sampling points includes the first planar region determined based on the sampling points.
[0115] Step 703: Identify the target obstacle.
[0116] Step 704: Determine the obstacle boundary information corresponding to the target obstacle in Cartesian coordinates. The obstacle boundary information corresponding to the target obstacle includes a second planar region determined based on the target obstacle.
[0117] Step 705: Determine the target driving space based on the path boundary information and obstacle boundary information corresponding to multiple sampling points.
[0118] Step 706: Based on at least one of the preset reference line, preset curvature requirements, and preset road direction requirements, and the target driving space, search for the endpoint of the planned path.
[0119] The preset curvature requirement means that the curvature of the planned path at the endpoint is less than the preset radius. The preset radius refers to the turning radius of the vehicle, which can be set according to the actual working conditions of the vehicle. For example, the preset radius is 6 meters.
[0120] The preset road direction requirement means that the planned path closely follows the road direction. Optionally, the vehicle measures whether the planned path closely follows the road direction by the absolute value of the dot product of the local path vector and the rightward unit vector of the bounding box. The smaller the absolute value of the dot product of the local path vector and the rightward unit vector of the bounding box, the closer the planned path is to the road direction. In addition, the cosine of the angle between the local path vector and the rightward unit vector of the bounding box needs to be less than or equal to a preset value. The preset value is set based on experiments or experience. For example, the preset value is 0.6.
[0121] It should be noted that if the distance between the searched path endpoint and the nearest sampling point is less than a preset distance, the nearest sampling point will be used as the endpoint of the planned path; conversely, if the distance between the searched path start point and the nearest sampling point is greater than a preset distance, the searched path endpoint will be used as the endpoint of the planned path. The preset distance is set based on experiments or experience, and this embodiment does not limit it. For example, the preset distance is 1 meter.
[0122] If the destination is inside or directly behind an obstacle, the planned path obtained in this case may not meet the boundary conditions of the destination or may have an excessive curvature at the destination location. Such a path is not traversable. Therefore, it is necessary to search for the destination in advance to avoid the above situations.
[0123] Step 707: Determine the planned path based on the target driving space, preset constraints, preset optimization objectives, and the endpoint of the planned path.
[0124] The vehicle constructs a planning path model based on the target driving space, preset constraints, preset optimization objectives, and the endpoint of the planned path. Then, the planning path model is solved to obtain the planned path.
[0125] In summary, the technical solution provided by the embodiments of this application can flexibly adjust the endpoint of the planned path by searching for at least one of the following: a pre-set reference line, a pre-set curvature requirement, and a pre-set road direction requirement, as well as the target driving space, thereby avoiding the occurrence of unsolvable problems.
[0126] Please see Figure 8 The present application also provides a path planning device, which includes:
[0127] The sampling point acquisition module 810 is used to acquire multiple sampling points.
[0128] The first determining module 820 is used to determine the path boundary information corresponding to multiple sampling points in the Cartesian coordinate system. The path boundary information corresponding to the sampling points includes a first planar region determined based on the sampling points.
[0129] The obstacle determination module 830 is used to determine the target obstacle.
[0130] The second determining module 840 is used to determine the obstacle boundary information corresponding to the target obstacle in Cartesian coordinates. The obstacle boundary information corresponding to the target obstacle includes a second planar region determined based on the target obstacle.
[0131] The driving space determination module 850 is used to determine the target driving space based on the path boundary information and obstacle boundary information corresponding to multiple sampling points.
[0132] The path planning module 860 is used to determine the planned path based on the target driving space, preset constraints, and preset optimization objectives.
[0133] In summary, the technical solution provided in this application uses a Cartesian coordinate system to describe the path boundary information determined based on each sampling point, and also uses a Cartesian coordinate system to describe the obstacle boundary information determined based on the target obstacle. Based on the aforementioned path boundary information and obstacle boundary information, a target driving space that is suitable for vehicle travel and will not cause a collision is determined. Finally, based on the target driving space, preset constraints, and preset optimization objectives, a planned path is determined. Since the Cartesian coordinate system is two-dimensional, any trajectory point on the road can be uniquely represented by a combination of horizontal and vertical coordinates. This application embodiment can avoid the situation where one coordinate corresponds to multiple trajectory points, thus improving the accuracy of path planning.
[0134] In some embodiments, the first determining module 820 is configured to, for the i-th sampling point, determine a first projection point of the i-th sampling point on the first road boundary and a second projection point of the i-th sampling point on the second road boundary, where i is a positive integer; determine the center point corresponding to the i-th sampling point based on the i-th sampling point, the first projection point and the second projection point; obtain the constraint distance in the first direction and the constraint distance in the second direction, wherein the first direction and the second direction are perpendicular to each other; and determine the path boundary information corresponding to the i-th sampling point based on the center point corresponding to the i-th sampling point, the constraint distance in the first direction and the constraint distance in the second direction.
[0135] In some embodiments, when i is greater than 1, the first determining module 820 is specifically used for the i-th sampling point: starting from the first projection point of the (i-1)-th sampling point on the first road boundary, searching on the first road boundary to obtain the first projection point of the i-th sampling point on the first road boundary; starting from the second projection point of the (i-1)-th sampling point on the second road boundary, searching on the second road boundary to obtain the second projection point of the i-th sampling point on the second road boundary.
[0136] In some embodiments, the device further includes: a destination search module (not shown in the figure). The destination search module is used to search for the destination of the planned path based on at least one of a preset reference line, a preset curvature requirement, and a preset road direction requirement, and a target driving space; wherein the preset curvature requirement means that the curvature of the planned path at the destination is less than a preset radius, and the preset road direction requirement means that the planned path closely follows the road direction. A path planning module 860 is used to determine the planned path based on the target driving space, preset constraints, preset optimization objectives, and the destination of the planned path.
[0137] In some embodiments, the driving space determination module 850 is used to determine m drivable areas based on the path boundary information and obstacle boundary information corresponding to multiple sampling points, where m is an integer greater than 2; based on the m drivable areas and a pre-set reference line, determine n candidate driving areas, where n is an integer greater than 1 and less than m; and if the m candidate driving areas are adjacent to each other in the target direction and all meet the connectivity condition, determine the multiple candidate driving areas as the target driving space, where the target direction is the driving direction of the vehicle.
[0138] In some embodiments, the device further includes a constraint determination module (not shown in the figure). The constraint determination module is configured to: determine boundary constraints among preset constraints based on path boundary information corresponding to multiple sampling points; and / or, determine start-point constraints among preset constraints based on the vehicle's current position or the start position in its historical driving trajectory; and / or, determine end-point constraints among preset constraints based on the road direction and the path direction of the planned path; and / or, determine continuity constraints among preset constraints based on parameter information of the curve connection points of two adjacent curves.
[0139] In some embodiments, the device further includes an optimization target determination module (not shown in the figure). The optimization target determination module is used to: determine a preset optimization target, the preset optimization target including at least one of the following: the planned path is close to a preset reference line, the planned path is close to the road direction, the steering wheel control angle of the vehicle when driving on the planned path is less than a preset angle, and the number of steering wheel controls when driving on the planned path is less than a preset number.
[0140] Please see Figure 9 The illustration shows that an embodiment of this application also provides a vehicle 900, which includes one or more processors 910, a memory 920, and one or more application programs. The one or more application programs are stored in the memory and configured to be executed by the one or more processors, and are configured to perform the parking path determination method described above.
[0141] The processor 910 may include one or more processing cores. The processor 910 connects to various parts of the entire battery management system using various interfaces and lines, and performs various functions and processes data of the battery management system by running or executing instructions, programs, code sets, or instruction sets stored in the memory 920, and by calling data stored in the memory 920. Optionally, the processor 910 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 910 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the displayed content; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 910 and may be implemented separately through a communication chip.
[0142] The memory 920 may include random access memory (RAM) or read-only memory (ROM). The memory 920 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 920 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as touch functionality, sound playback functionality, image playback functionality, etc.), and instructions for implementing the various method embodiments described below. The data storage area may also store data created during the use of the electronic device (such as phonebook data, audio and video data, chat log data, etc.).
[0143] Please see Figure 10 The present application also provides a computer-readable storage medium 1000, which stores computer program instructions 1010 that can be invoked by a processor to execute the methods described in the above embodiments.
[0144] Computer-readable storage media can be electronic storage devices such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Optionally, computer-readable storage media includes non-transitory computer-readable storage medium. Computer-readable storage medium 1000 has storage space for computer program instructions 1010 that perform any of the method steps described above. These computer program instructions 1010 can be read from or written to one or more computer program products.
[0145] The above are merely preferred embodiments of this application and are not intended to limit this application in any way. Although this application has disclosed preferred embodiments as above, it is not intended to limit this application. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the technical solution of this application. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of this application without departing from the scope of the technical solution of this application shall still fall within the scope of the technical solution of this application.
Claims
1. A path planning method, characterized in that, The method includes: Obtain multiple sampling points; In a Cartesian coordinate system, path boundary information corresponding to multiple sampling points is determined, and the path boundary information corresponding to the sampling points includes a first planar region determined based on the sampling points. Identify the target obstacle; In Cartesian coordinates, the obstacle boundary information corresponding to the target obstacle is determined, and the obstacle boundary information corresponding to the target obstacle includes a second planar region determined based on the target obstacle; Based on the path boundary information and obstacle boundary information corresponding to multiple sampling points, the target driving space is determined. Based on the target driving space, preset constraints, and preset optimization objectives, a planned path is determined; wherein, determining the path boundary information corresponding to multiple sampling points in a Cartesian coordinate system includes: For the i-th sampling point, determine the first projection point of the i-th sampling point on the first road boundary and the second projection point of the i-th sampling point on the second road boundary, where i is a positive integer; The center point corresponding to the i-th sampling point is determined based on the i-th sampling point, the first projection point, and the second projection point; Obtain the constraint distance in the first direction and the constraint distance in the second direction, wherein the first direction and the second direction are perpendicular to each other; Based on the center point corresponding to the i-th sampling point, the constraint distance in the first direction, and the constraint distance in the second direction, the path boundary information corresponding to the i-th sampling point is determined.
2. The method according to claim 1, characterized in that, When i is greater than 1, determining the first projection point of the i-th sampling point on the first road boundary and the second projection point of the i-th sampling point on the second road boundary includes: For the i-th sampling point: Starting from the first projection point of the (i-1)th sampling point on the first road boundary, search on the first road boundary to obtain the first projection point of the i-th sampling point on the first road boundary; Starting from the second projection point of the (i-1)th sampling point on the second road boundary, a search is performed on the second road boundary to obtain the second projection point of the i-th sampling point on the second road boundary.
3. The method according to claim 1, characterized in that, Before determining the planned path based on the target driving space, preset constraints, and preset optimization objectives, the process also includes: Based on at least one of a pre-set reference line, a pre-set curvature requirement, and a pre-set road direction requirement, and the target driving space, the endpoint of the planned path is searched; wherein, the pre-set curvature requirement means that the curvature of the planned path at the endpoint is less than a pre-set radius, and the pre-set road direction requirement means that the planned path is close to the road direction; The step of determining the planned path based on the target driving space, preset constraints, and preset optimization objectives includes: The planned path is determined based on the target driving space, the preset constraints, the preset optimization target, and the endpoint of the planned path.
4. The method according to claim 1, characterized in that, The determination of the target driving space based on the path boundary information and obstacle boundary information corresponding to multiple sampling points includes: Based on the path boundary information and obstacle boundary information corresponding to multiple sampling points, m drivable areas are determined, where m is an integer greater than 2; Based on the m drivable areas and the pre-set reference lines, n candidate drivable areas are determined, where n is an integer greater than 1 and less than m; If all the candidate driving areas adjacent to each other in the target direction meet the connectivity condition, then the multiple candidate driving areas are determined as the target driving space, where the target direction is the driving direction of the vehicle.
5. The method according to any one of claims 1 to 4, characterized in that, Before determining the planned path based on the target driving space, preset constraints, and preset optimization objectives, the process also includes: Based on the path boundary information corresponding to the multiple sampling points, the boundary constraints in the preset constraints are determined; or / and, Based on the vehicle's current position or its starting position in its historical driving trajectory, determine the starting point constraint condition in the preset constraint conditions; or / and, Based on the road direction and the path direction of the planned path, determine the endpoint constraint condition in the preset constraint conditions; or / and, Multiple curves are determined based on multiple sampling points. The continuity constraint conditions satisfied by two adjacent curves among the multiple curves include at least one of the following: the curve endpoint of the i-th curve coincides with the curve start point of the (i+1)-th curve; the curve endpoint of the i-th curve and the curve start point of the (i+1)-th curve have the same direction; and the curve endpoint of the i-th curve and the curve start point of the (i+1)-th curve have the same curvature.
6. The method according to any one of claims 1 to 4, characterized in that, Before determining the planned path based on the target driving space, preset constraints, and preset optimization objectives, the process also includes: The preset optimization objective is determined, and the preset optimization objective includes at least one of the following: the planned path is close to the preset reference line, the planned path is close to the road direction, the steering wheel control angle of the vehicle when driving on the planned path is less than a preset angle, and the number of steering wheel controls of the vehicle when driving on the planned path is less than a preset number.
7. A path planning device, characterized in that, The device includes: The sampling point acquisition module is used to acquire multiple sampling points; The first determining module is used to determine the path boundary information corresponding to the plurality of sampling points in the Cartesian coordinate system, wherein the path boundary information corresponding to the sampling points includes a first planar region determined based on the sampling points; The obstacle determination module is used to determine the target obstacle; The second determining module is used to determine the obstacle boundary information corresponding to the target obstacle in Cartesian coordinates, wherein the obstacle boundary information corresponding to the target obstacle includes a second planar region determined based on the target obstacle; The driving space determination module is used to determine the target driving space based on the path boundary information and obstacle boundary information corresponding to multiple sampling points respectively; The path planning module is used to determine the planned path based on the target driving space, preset constraints, and preset optimization objectives. The first determining module is configured to: for the i-th sampling point, determine a first projection point of the i-th sampling point on the first road boundary and a second projection point of the i-th sampling point on the second road boundary, where i is a positive integer; determine the center point corresponding to the i-th sampling point based on the i-th sampling point, the first projection point, and the second projection point; obtain the constraint distance in a first direction and the constraint distance in a second direction, wherein the first direction and the second direction are perpendicular to each other; and determine the path boundary information corresponding to the i-th sampling point based on the center point corresponding to the i-th sampling point, the constraint distance in the first direction, and the constraint distance in the second direction.
8. A vehicle, characterized in that, The vehicles include: One or more processors; Memory; One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more applications being configured to perform the method as described in any one of claims 1-6.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions that can be invoked by a processor to perform the method as described in any one of claims 1-6.
Citation Information
Patent Citations
Trajectory optimization method and device
CN111830979A
