Vehicle path planning method and device, storage medium, equipment and vehicle

Abstract

The invention discloses a vehicle path planning method and device, a storage medium, equipment and a vehicle, and the method comprises the steps: dividing an unstructured road envelope based on vehicle front environment perception into a plurality of Voronoi regions, the unstructured road envelope being a polygonal envelope enclosed by irregular road boundaries; according to the Voronoi area where the path planning point is located, the target distance between the path planning point and a target boundary line is determined, and the target boundary line is an unstructured road boundary closest to the path planning point in the unstructured road envelope; constructing a path planning model according to the road boundary constraint, the vehicle kinematics constraint, the vehicle control quantity constraint and the vehicle initial state; and calling the target distance to solve the path planning model, and determining an optimal solution of the path planning model as a target planning path. Therefore, according to the embodiment of the invention, the high-quality planning path suitable for the unstructured road can be efficiently planned without depending on a navigation map.

Description

Technical Field

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

[0002] Regional valet parking systems utilize vehicle perception information to plan feasible routes and autonomously guide vehicles to the vicinity of target parking spaces. Narrow-lane passage assistance systems are an effective means of improving the safety and efficiency of inexperienced drivers navigating narrow roads. Both regional valet parking and narrow-lane passage scenarios involve unstructured roads (i.e., irregular roads). However, because unstructured roads change in real time as vehicles move, it is difficult to plan high-quality routes even with navigation maps. Therefore, how to plan routes suitable for unstructured roads without relying on navigation maps is a key problem to be solved. Summary of the Invention

[0003] This application provides a vehicle route planning method, apparatus, storage medium, device, and vehicle, which can efficiently plan high-quality routes suitable for unstructured roads without relying on navigation maps.

[0004] The specific technical solution is as follows:

[0005] In a first aspect, embodiments of this application provide a vehicle routing method, the method comprising:

[0006] The unstructured road envelope based on the vehicle's forward environment perception is divided into multiple Voronoi regions, wherein the unstructured road envelope is a polygonal envelope enclosed by irregular road boundaries;

[0007] Based on the Voronoi region where the path planning point is located, determine the target distance from the path planning point to the target boundary line, wherein the target boundary line is the unstructured road boundary in the unstructured road envelope that is closest to the path planning point;

[0008] A path planning model is constructed based on road boundary constraints, vehicle kinematic constraints, vehicle control constraints, and the initial state of the vehicle.

[0009] The target distance is used to solve the path planning model, and the optimal solution of the path planning model is determined as the target planning path.

[0010] As can be seen from the above scheme, the embodiments of this application can not only obtain high-quality target planning paths by comprehensively considering road boundary constraints, vehicle kinematic constraints and vehicle control constraints without relying on navigation maps, but also quickly obtain the unstructured road boundary closest to the path planning point by matching the path planning point with multiple Voronoi regions divided by the unstructured road envelope during the solution process, thereby improving the solution efficiency of the path planning model and thus improving the efficiency of path planning.

[0011] In one possible implementation, the unstructured road envelope based on vehicle-front environment perception is divided into multiple Voronoi regions, including:

[0012] Based on the concavity or convexity of the vertices, the unstructured road envelope is divided into multiple minimum units, wherein each minimum unit includes a convex edge or multiple consecutive concave edges.

[0013] The smallest unit is taken as the leaf node of the target tree, the unstructured road envelope is taken as the root node of the target tree, and starting from the leaf node, the adjacent nodes of the current level are merged as child nodes to obtain the parent node corresponding to the previous level, until the root node is reached, and the target tree is completed.

[0014] In the target tree, starting from the leaf node, the central axis of each node in the current level is first determined, and then the sibling nodes in the current level and their central axes are merged to generate the central axis of the corresponding parent node in the previous level, until the target central axis of the root node is generated.

[0015] The unstructured road envelope is divided into multiple Voronoi regions based on the target centerline and the Voronoi edges determined by the auxiliary lines used in generating the target centerline.

[0016] As described above, this embodiment first divides the unstructured road envelope into multiple smallest units as leaf nodes, uses the unstructured road envelope as the root node, and generates internal nodes based on the leaf nodes. Then, a target tree is created based on the leaf nodes, root node, and internal nodes. Finally, the central axis is calculated level by level from the leaf nodes to the root node, ultimately generating the target central axis based on the nodes. Based on the Voronoi edges of the target central axis, the unstructured road envelope is divided into multiple Voronoi regions. This divide-and-conquer Voronoi region partitioning method can improve processing efficiency and accuracy.

[0017] In one possible implementation, the sibling nodes and their central axes in the current level are merged to generate the central axis of the corresponding parent node in the next higher level, including:

[0018] The elements contained in each sibling node in the current level and the central axis originating from each vertex are cached respectively as the cache information corresponding to each sibling node. The elements include convex edges, concave edges and concave points.

[0019] Clear the central axis starting from the common vertex of the sibling nodes in the cache. For each sibling node, starting from the common vertex, traverse the central axis starting from each vertex of the sibling node in the order of the vertices of the sibling node. Regenerate and cache the central axis starting from the common vertex after merging the sibling nodes, and update the central axis starting from the non-common vertex.

[0020] The regenerated central axis originating from the common vertex, the updated central axis originating from the non-common vertex, and the elements contained in all sibling nodes in the current level are used as cache information for the corresponding parent node in the previous level. The central axis originating from the common vertex in the cache information of the parent node is the central axis of the parent node.

[0021] As can be seen from the above scheme, the embodiments of this application can accelerate the generation efficiency of the centerline of the unstructured road envelope by caching each node as an element and the centerline, and traversing and merging the centerlines originating from each vertex to generate the centerline of the parent node until the centerline of the root node is generated.

[0022] In one possible implementation, the plurality of Voronoi regions include the inner Voronoi regions of the unstructured road envelope. Determining the target distance from the path planning point to the target boundary line based on the Voronoi region where the path planning point is located includes:

[0023] Starting from the target centerline segment closest to the path planning starting point, query forward along the path planning direction to determine whether the path planning point exceeds the target centerline segment and internal Voronoi edge in the current internal Voronoi region to be judged;

[0024] If the path planning point does not exceed the target centerline segment and the internal Voronoi edge of the currently to-be-determined internal Voronoi region, the path planning point is determined to be within the currently to-be-determined internal Voronoi region;

[0025] If the path planning point exceeds the target centerline segment or internal Voronoi edge of the currently to be judged internal Voronoi region, the internal Voronoi region where the next target centerline segment is located is determined as the new currently to be judged internal Voronoi region. The execution steps are returned to determine whether the path planning point exceeds the target centerline segment and internal Voronoi edge of the currently to be judged internal Voronoi region, until the Voronoi region where the path planning point is located is determined.

[0026] After determining the internal Voronoi region where the path planning point is located, the unstructured road boundary in the determined internal Voronoi region is taken as the target boundary line, and the target distance from the path planning point to the target boundary line is determined.

[0027] As can be seen from the above scheme, by traversing the relative positional relationship between the starting point of the path planning and the target centerline segment and the internal Voronoi edge in each Voronoi region, this application embodiment can quickly determine which Voronoi region the starting point of the path planning is located in, thereby quickly matching the unstructured road boundary closest to the starting point of the path planning.

[0028] In one possible implementation, the plurality of Voronoi regions include the inner Voronoi region and the outer Voronoi region of the unstructured road envelope. Determining the target distance from the path planning point to the target boundary line based on the Voronoi region where the path planning point is located includes:

[0029] If the path planning point is located inside the unstructured road envelope, find the internal Voronoi region where the path planning point is located;

[0030] If the path planning point is located outside the unstructured road envelope, locate the external Voronoi region where the path planning point is located;

[0031] After determining the Voronoi region where the path planning point is located, the unstructured road boundary in the determined Voronoi region is taken as the target boundary line, and the target distance from the path planning point to the target boundary line is determined.

[0032] As can be seen from the above solution, by adding an external Voronoi region, this embodiment of the application can still quickly match the unstructured road boundary closest to the path planning starting point even when the path planning point is located outside the unstructured road envelope.

[0033] In one possible implementation, solving the path planning model by invoking the target distance includes:

[0034] The path planning model is solved by using the augmented Lagrange algorithm, the forward propagation algorithm, and the backpropagation algorithm, with the target distance being called.

[0035] As can be seen from the above scheme, the embodiments of this application can transform the constrained optimization problem described by the path planning model into an unconstrained optimization problem by utilizing augmented Lagrange and Voronoi regions of unstructured roads. By using forward propagation and backward propagation to iteratively solve the unconstrained optimization problem, a feasible target planning path can be obtained.

[0036] In one possible implementation, the path planning model includes:

[0037]

[0038] Wherein, the Z * and the U * Let Z and U represent the system state variables set and the system control variables set of each path planning point in the optimal solution, respectively. Let Z and U represent the system state variables set and the system control variables set of each path planning point in any solution, respectively. k and the z N Let u represent the system state variables at time k and time N, respectively. k c represents the system control quantity at time k. k (z k ) and the c N (z N ) represent the penalty functions for the system state variables at time k and time N, respectively, where d k (u k ) represents the penalty function for the system control quantity at time k, where z k+1 =f k (z k ,u k ) represents the vehicle kinematic constraints, and g k (z k )≤0 indicates the road boundary constraint, and h k (u k )≤0 indicates the vehicle control quantity constraint, and X init =X 0 An assignment expression representing the initial state of the vehicle.

[0039] Secondly, embodiments of this application provide a vehicle route planning device, the device comprising:

[0040] A segmentation unit is used to divide the unstructured road envelope based on the vehicle's forward environment perception into multiple Voronoi regions, wherein the unstructured road envelope is a polygonal envelope enclosed by irregular road boundaries;

[0041] The first determining unit is used to determine the target distance from the path planning point to the target boundary line based on the Voronoi region where the path planning point is located, wherein the target boundary line is the unstructured road boundary in the unstructured road envelope that is closest to the path planning point.

[0042] The building unit is used to construct a path planning model based on road boundary constraints, vehicle kinematic constraints, vehicle control constraints, and the initial state of the vehicle.

[0043] The solution unit is used to solve the path planning model by calling the target distance;

[0044] The second determining unit is used to determine the optimal solution of the path planning model as the target planning path.

[0045] In one possible implementation, the partitioning unit includes:

[0046] The splitting module is used to split the unstructured road envelope into multiple minimum units according to the concavity and convexity of the vertices, wherein each minimum unit includes a convex edge or multiple consecutive concave edges;

[0047] The module is used to take the smallest unit as the leaf node of the target tree, take the unstructured road envelope as the root node of the target tree, and start from the leaf node, merge the adjacent nodes of the current level as child nodes to obtain the parent node corresponding to the previous level, until the root node is reached, and the target tree is established.

[0048] The generation module is used to, starting from the leaf node in the target tree, first determine the central axis of each node in the current level, then merge the sibling nodes and their central axes in the current level to generate the central axis of the corresponding parent node in the next higher level, until the target central axis of the root node is generated.

[0049] The first determining module is used to divide the unstructured road envelope into multiple Voronoi regions based on the target centerline and the Voronoi edges determined by the auxiliary lines used in generating the target centerline.

[0050] In one possible implementation, the generation module is used for:

[0051] The elements contained in each sibling node in the current level and the central axis originating from each vertex are cached respectively as the cache information corresponding to each sibling node. The elements include convex edges, concave edges and concave points.

[0052] Clear the central axis starting from the common vertex of the sibling nodes in the cache. For each sibling node, starting from the common vertex, traverse the central axis starting from each vertex of the sibling node in the order of the vertices of the sibling node. Regenerate and cache the central axis starting from the common vertex after merging the sibling nodes, and update the central axis starting from the non-common vertex.

[0053] The regenerated central axis originating from the common vertex, the updated central axis originating from the non-common vertex, and the elements contained in all sibling nodes in the current level are used as cache information for the corresponding parent node in the previous level. The central axis originating from the common vertex in the cache information of the parent node is the central axis of the parent node.

[0054] In one possible implementation, the plurality of Voronoi regions include the inner Voronoi regions of the unstructured road envelope, and the first determining unit includes:

[0055] The matching module is used to start from the target centerline segment closest to the path planning starting point, and query forward along the path planning direction to determine whether the path planning point exceeds the target centerline segment and internal Voronoi edge of the currently to-be-determined internal Voronoi region. If the path planning point does not exceed the target centerline segment and internal Voronoi edge of the currently to-be-determined internal Voronoi region, the path planning point is determined to be within the currently to-be-determined internal Voronoi region. If the path planning point exceeds the target centerline segment or internal Voronoi edge of the currently to-be-determined internal Voronoi region, the internal Voronoi region where the next target centerline segment is located is determined as the new currently to-be-determined internal Voronoi region, and the execution step of determining whether the path planning point exceeds the target centerline segment and internal Voronoi edge of the currently to-be-determined internal Voronoi region is returned until the Voronoi region where the path planning point is located is determined.

[0056] The second determining module is used to, after determining the internal Voronoi region where the path planning point is located, take the unstructured road boundary in the determined internal Voronoi region as the target boundary line, and determine the target distance from the path planning point to the target boundary line.

[0057] In one possible implementation, the plurality of Voronoi regions include the inner Voronoi region and the outer Voronoi region of the unstructured road envelope;

[0058] The first determining unit is configured to: locate the inner Voronoi region where the path planning point is located when the path planning point is located inside the unstructured road envelope; locate the outer Voronoi region where the path planning point is located when the path planning point is located outside the unstructured road envelope; after determining the Voronoi region where the path planning point is located, use the unstructured road boundary in the determined Voronoi region as the target boundary line, and determine the target distance from the path planning point to the target boundary line.

[0059] In one possible implementation, the solution unit is used to solve the path planning model by invoking the target distance using the augmented Lagrange algorithm, the forward propagation algorithm, and the backpropagation algorithm.

[0060] In one possible implementation, the path planning model includes:

[0061]

[0062] Wherein, the Z * and the U * Let Z and U represent the system state variables set and the system control variables set of each path planning point in the optimal solution, respectively. Let Z and U represent the system state variables set and the system control variables set of each path planning point in any solution, respectively. k and the z N Let u represent the system state variables at time k and time N, respectively. k c represents the system control quantity at time k. k (z k ) and the c N (z N ) represent the penalty functions for the system state variables at time k and time N, respectively, where d k (u k ) represents the penalty function for the system control quantity at time k, where z k+1 =f k (z k ,u k ) represents the vehicle kinematic constraints, and g k (z k )≤0 indicates the road boundary constraint, and h k (u k)≤0 indicates the vehicle control quantity constraint, and X init =X 0 An assignment expression representing the initial state of the vehicle.

[0063] As can be seen from the above scheme, the embodiments of this application can not only obtain high-quality target planning paths by comprehensively considering road boundary constraints, vehicle kinematic constraints, and vehicle control constraints without relying on navigation maps, but also quickly obtain the unstructured road boundary closest to the path planning point by matching the path planning point with multiple Voronoi regions divided by the unstructured road envelope during the solution process. This can improve the solution efficiency of the path planning model and thus improve the efficiency of path planning.

[0064] Thirdly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the method as described in any possible implementation of the first aspect.

[0065] Fourthly, embodiments of this application provide an electronic device, which includes:

[0066] One or more processors;

[0067] The processor is coupled to a storage device for storing one or more programs;

[0068] When one or more programs are executed by one or more processors, the electronic device performs the method as described in any possible implementation of the first aspect.

[0069] Fifthly, embodiments of this application provide a vehicle that includes the means as described in any possible implementation of the second aspect, or includes electronic equipment as described in the fourth aspect.

[0070] In a sixth aspect, embodiments of this application provide a computer program product containing instructions that, when executed on a computer or processor, cause the computer or processor to perform the method described in any possible implementation of the first aspect. Attached Figure Description

[0071] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0072] Figure 1 A flowchart illustrating a vehicle routing method provided in an embodiment of this application;

[0073] Figure 2 An example diagram of a target tree for an unstructured road envelope provided in an embodiment of this application;

[0074] Figure 3 An example diagram illustrating the construction process of the internal Voronoi region of the left subtree provided in this application embodiment;

[0075] Figure 4 An example diagram illustrating the construction process of the internal Voronoi region of a right subtree, provided in an embodiment of this application;

[0076] Figure 5 An example diagram illustrating the process of constructing the internal Voronoi region of an unstructured road envelope, as provided in this application embodiment;

[0077] Figure 6 An example diagram illustrating the construction of the central axis between nodes in a target tree, provided in an embodiment of this application;

[0078] Figure 7 An example diagram illustrating the Voronoi region partitioning result within an unstructured road envelope, provided in an embodiment of this application;

[0079] Figure 8 An example diagram illustrating a structured road boundary search process provided in this application embodiment;

[0080] Figure 9 An example diagram illustrating the process of constructing the outer Voronoi region of a left subtree, provided in an embodiment of this application;

[0081] Figure 10 An example diagram illustrating the process of constructing the outer Voronoi region of a right subtree, provided in an embodiment of this application;

[0082] Figure 11 A block diagram illustrating the composition of a vehicle routing device provided in this application embodiment;

[0083] Figure 12 This is a schematic diagram of the structure of an electronic device or computer device provided in an embodiment of this application. Detailed Implementation

[0084] 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.

[0085] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The terms "comprising" and "having," and any variations thereof, in the embodiments and drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0086] Figure 1 This is a flowchart illustrating a vehicle routing method. This method can be applied to any scenario with an unstructured road envelope, such as valet parking or narrow passage scenarios. It can also be applied to electronic devices or computer equipment, specifically to vehicles or servers that interact with vehicles. The method may include the following steps:

[0087] S110: Divide the unstructured road envelope based on the vehicle's forward environment perception into multiple Voronoi regions.

[0088] The unstructured road envelope is a polygonal envelope formed by irregular road boundaries. During vehicle operation, image sensors, lidar, and other sensors can be used to comprehensively perceive environmental information ahead, thereby determining the unstructured road envelope of the road space that the vehicle can travel on while avoiding collisions with surrounding obstacles.

[0089] There are many methods for dividing the unstructured road envelope into multiple Voronoi regions. This application proposes a new method that draws on the theory of divide-and-conquer algorithms. First, the unstructured road envelope is split into multiple sub-envelopes, and the central axis of each sub-envelope is calculated. Then, the central axes of the sub-envelopes are merged to generate the central axis of the complete envelope, thereby generating multiple Voronoi regions. This method can improve the efficiency and accuracy of obtaining Voronoi regions.

[0090] Specifically, the method includes steps A1-A4:

[0091] A1. Based on the concavity and convexity of the vertices, the unstructured road envelope is divided into multiple minimum units.

[0092] Each minimum unit includes a convex edge or multiple concave edges. When one edge is convex, the minimum unit is that edge. When one edge is concave, the minimum unit is a broken line composed of all the concave edges.

[0093] For example, Figure 2 The polygon on the right represents the unstructured road envelope. The vertices P2, P8, P9, and P in this diagram are... 10 P 11 With concave vertices and convex vertices as the remaining vertices, the unstructured road envelope can be divided into multiple minimum units based on the concavity and convexity of the vertices, each including... and in, The whole is a smallest unit. The whole is also the smallest unit.

[0094] A2. Take the smallest unit as the leaf node of the target tree, take the unstructured road envelope as the root node of the target tree, and start from the leaf node, merge the adjacent nodes of the current level as child nodes to obtain the corresponding parent node of the previous level, until the root node is reached, and the target tree is completed.

[0095] like Figure 2 As shown, and These are leaf nodes, and Merge into internal node I1, and Merge into internal node I2, and Merge into internal node I3, and Merge internal nodes I1 and I2 into internal node I5, merge internal nodes I3 and I4 into internal node I6, and finally merge internal nodes I5 and I6 into the root node, thus generating the target tree.

[0096] A3. In the target tree, starting from the leaf node, first determine the central axis of each node in the current level, then merge the sibling nodes in the current level and their central axes to generate the central axis of the corresponding parent node in the next higher level, until the target central axis of the root node is generated.

[0097] To accelerate the generation efficiency of the central axis, the specific implementation of the step "determine the central axis of each node in the current level, and then merge the sibling nodes and their central axes in the current level to generate the central axis of the corresponding parent node in the next higher level" includes: caching the elements contained in each sibling node in the current level and the central axis originating from each vertex as cache information for each sibling node, with elements including convex edges, concave edges, and concave points; clearing the cached central axes originating from the common vertex of the sibling nodes; for each sibling node, starting from the common vertex, traversing the central axes originating from each vertex of the sibling node in the order of the vertices in the sibling node, regenerating and caching the merged central axis originating from the common vertex, and updating the central axes originating from non-common vertices; using the regenerated central axis originating from the common vertex, the updated central axis originating from non-common vertices, and the elements contained in all sibling nodes in the current level as cache information for the corresponding parent node in the next higher level, where the central axis originating from the common vertex in the parent node's cache information is the parent node's central axis.

[0098] Here, a sibling node refers to a child node at the same level as the same parent node. When the central axis originating from a certain vertex is used as a ray, it can be cached using the form of vertex + placeholder, so that when the central axis is updated later, the intersection point of the ray with other central axes can be used to replace the placeholder.

[0099] For each sibling node, starting from the common vertex, traverse the central axis originating from each vertex of the sibling node in the order of their respective vertices. Regenerate and cache the central axis originating from the common vertex after merging the sibling nodes. This includes: for each sibling node, starting from the common vertex, traversing the central axis originating from each vertex of the sibling node in the order of their respective vertices, determining the intersection points between the central axis originating from the common vertex and the central axes originating from each of the traversed vertices, and regenerating the central axis originating from the common vertex based on the set of line segments formed by the intersection points originating from the common vertex.

[0100] For example, such as Figure 3 As shown, from leaf nodes and The generated central axis is in,

[0101] The following examples illustrate the meaning of the symbols above. This represents line segment P3A1. Represent the parabola segment A1A2, This indicates that the central axis of line segments P2P3 and P3P4 is a line segment.

[0102] like Figure 3 As shown, from leaf nodes and The generated central axis is in,

[0103]

[0104] like Figure 3 As shown, the central axis generated by internal node I1 and internal node I2 is in,

[0105] like Figure 4 As shown, from leaf nodes and The generated central axis is in,

[0106] like Figure 4 As shown, from leaf nodes and The generated central axis is in,

[0107]

[0108] like Figure 4 As shown, the central axis generated by internal node I3 and internal node I4 is in,

[0109] like Figure 5 As shown, the central axis generated by internal node I5 and internal node I6 is

[0110] in,

[0111] The central axis between the internal nodes and the root node is constructed using a level-order traversal algorithm for a binary tree. For example... Figure 6 As shown, using leaf nodes and Taking the generated central axis as an example, the data structure of leaf nodes and internal nodes contains two parts: elements and the central axis. Leaf nodes The included central axis is and leaf nodes The included central axis is and □ is a placeholder. The common vertex of the two leaf nodes is P3, which needs to be cleared before rebuilding. Consider the central axis passing through vertex P3 and its sequence of operations. and If they intersect, the reconstructed result will be And there are and

[0112] A4. Based on the target centerline and the Voronoi edges determined by the auxiliary lines used in generating the target centerline, divide the unstructured road envelope into multiple Voronoi regions.

[0113] like Figure 7 As shown, the Voronoi edges include The Voronoi region includes regions Z1 to Z... 16 Regions, such as region Z1, include

[0114] S120: Determine the target distance from the path planning point to the target boundary line based on the Voronoi region where the path planning point is located.

[0115] The target boundary line is the unstructured road boundary that is closest to the path planning point in the unstructured road envelope.

[0116] Multiple Voronoi regions include the internal Voronoi regions of the unstructured road envelope, meaning that multiple Voronoi regions are divided within the unstructured road envelope. When multiple Voronoi regions include the internal Voronoi regions of the unstructured road envelope, this step is implemented as follows: Starting from the target centerline segment closest to the path planning starting point, query forward along the path planning direction to determine if the path planning point exceeds the target centerline segment and internal Voronoi edges of the currently to-be-determined internal Voronoi region; if the path planning point does not exceed the target centerline segment and internal Voronoi edges of the currently to-be-determined internal Voronoi region, determine that the path planning point is within the currently to-be-determined internal Voronoi region; if the path planning point exceeds the currently to-be-determined internal Voronoi region, determine that the path planning point is within the currently to-be-determined internal Voronoi region; if the path planning point exceeds the currently to-be-determined internal Voronoi region... If the target centerline segment or internal Voronoi edge is in the NOI region, the internal Voronoi region where the next target centerline segment is located is determined as the new current internal Voronoi region to be judged. The execution steps are returned to determine whether the path planning point exceeds the target centerline segment and internal Voronoi edge in the current internal Voronoi region to be judged, until the Voronoi region where the path planning point is located is determined. After determining the internal Voronoi region where the path planning point is located, the unstructured road boundary in the determined internal Voronoi region is taken as the target boundary line, and the target distance from the path planning point to the target boundary line is determined.

[0117] For example, in combination Figure 7 and Figure 8 It can be seen that if the path point does not exceed... and central axis segment If a path point is located within region Z1, then the path point is adjacent to the boundary of the unstructured road. The closest; otherwise, determine the path planning point and Central axis section The relative positional relationship between them.

[0118] If the path planning point has not been exceeded and central axis segment If the path point is located within region Z2, then the path planning point and the boundary of the unstructured road are... The closest; otherwise, compare the path point with the nearest. Central axis section The relative positional relationship between them.

[0119] If the path planning point has not been exceeded and central axis segment If the path planning point is located within region Z3, then the path planning point is closest to the boundary of the unstructured road P8; otherwise, determine the path planning point's distance from the boundary of the unstructured road P8. Central axis section The relative positional relationship between them.

[0120] If the path planning point has not been exceeded and central axis segment If the path planning point is located within region Z4, then the path planning point is adjacent to the boundary of the unstructured road. The closest; otherwise, determine the path planning point and Central axis section The relative positional relationship between them.

[0121] If the path planning point has not been exceeded and central axis segment If the path planning point is located within region Z5, then the path planning point is closest to the boundary of the unstructured road P9; otherwise, determine the path planning point's distance from the boundary of the unstructured road P9. Central axis section The relative positional relationship between them.

[0122] If the path planning point has not been exceeded and central axis segment If the path planning point is located within region Z6, then the path planning point and the boundary of the unstructured road... The closest; otherwise, determine the path planning point and Central axis section

[0123] The relative positional relationship between them.

[0124] If the path planning point has not been exceeded and central axis segment If the path planning point is located within region Z7, then the path planning point and the boundary of the unstructured road are... The closest; otherwise, determine the path planning point and Central axis section The relative positional relationship between them.

[0125] If the path planning point has not been exceeded and central axis segment If the path point is located within region Z8, then the path planning point and the unstructured road boundary P 10 The closest; otherwise, determine the path planning point and Central axis section The relative positional relationship between them.

[0126] If the path planning point has not been exceeded and central axis segment If the path planning point is located within region Z9, then the path planning point and the boundary of the unstructured road are... The closest; otherwise, determine the path planning point and The relative positional relationship between them.

[0127] If the path planning point has not been exceeded The path point is located in region Z. 10 Inside, the waypoints and the boundaries of unstructured roads. The closest; otherwise, determine the path planning point and Central axis section The relative positional relationship between them.

[0128] If the path planning point has not been exceeded and central axis segment The path planning point is located in region Z. 11 Within, the path planning point and the boundary of the unstructured road. The closest; otherwise, determine the path planning point and Central axis section The relative positional relationship between them.

[0129] If the path planning point has not been exceeded and central axis segment The path planning point is located in region Z. 12 If the path planning point is within the boundary of the unstructured road, then the path planning point is closest to the boundary of the unstructured road P2; otherwise, determine the path planning point's distance from the boundary of the unstructured road P2. Central axis section The relative positional relationship between them.

[0130] If the path planning point has not been exceeded and central axis segment The path planning point is located in region Z. 13 Within, the path planning point and the unstructured road boundary P 11 The closest; otherwise, determine the path planning point and Central axis section The relative positional relationship between them.

[0131] If the path planning point has not been exceeded and central axis segment The path planning point is located in region Z. 14 Within, the path planning point and the boundary of the unstructured road. The closest distance; if the path planning point is located on the central axis segment. The upper part, the path planning point is located in region Z. 15 Within, the path planning point and the boundary of the unstructured road. The closest distance; if the path planning point is located on the central axis segment. The lower part, the path point is located in region Z. 16 Within, the path planning point and the boundary of the unstructured road. Closest distance.

[0132] As can be seen from the above scheme, by traversing the relative positional relationship between the starting point of the path planning and the target centerline segment and the internal Voronoi edge in each Voronoi region, this application embodiment can quickly determine which Voronoi region the starting point of the path planning is located in, thereby quickly matching the unstructured road boundary closest to the starting point of the path planning.

[0133] In one possible implementation, the path planning process may extend beyond the unstructured road envelope, causing a path planning point to reach the outside of the unstructured road envelope. In this case, the outer region of the unstructured road envelope can be divided into Voronoi regions to obtain the outer Voronoi region. The outer Voronoi region is divided in the same direction as the inner Voronoi region, the only difference being that the left and right subtrees of the target tree need to be divided separately, instead of dividing the whole.

[0134] When multiple Voronoi regions include both inner and outer Voronoi regions of an unstructured road envelope, if the path planning point is located inside the unstructured road envelope, the inner Voronoi region where the path planning point is located is searched; if the path planning point is located outside the unstructured road envelope, the outer Voronoi region where the path planning point is located is searched. After determining the Voronoi region where the path planning point is located, the boundary of the unstructured road within the determined Voronoi region is used as the target boundary line, and the target distance from the path planning point to the target boundary line is determined.

[0135] For example, such as Figure 9 As shown, the left half of the unstructured road envelope is composed of and The system is divided into multiple regions. If a path point extends beyond the left half of the unstructured road envelope polygon, the system searches that region to find the most suitable unstructured road boundary.

[0136] like Figure 10 As shown, the right half of the unstructured road envelope is composed of and The system is divided into multiple regions. If a path point extends beyond the right half of the unstructured road envelope polygon, the system searches that region to find the most suitable unstructured road boundary.

[0137] It should be added that, since the unstructured road envelope in front of the vehicle changes continuously as the vehicle moves during the parking process, steps S110-S120 will be updated in real time to ensure the accuracy of path planning.

[0138] S130: Construct a path planning model based on road boundary constraints, vehicle kinematic constraints, vehicle control constraints, and the initial state of the vehicle.

[0139] Among them, road boundary constraints are constraints determined based on the unstructured road envelope. Their characterization is to ensure that vehicles do not exceed the unstructured road boundary and travel along the central axis of the unstructured road envelope. In path planning, the target distance corresponding to each planned path planning point can be substituted into the road boundary constraints to adjust the constraints for each path planning point.

[0140] System control constraints are constraints on control parameters such as the rate of change of front wheel steering angle and acceleration. For example, linear two-degree-of-freedom dynamic equations can be used to impose these constraints.

[0141] The method for determining vehicle kinematic constraints includes: obtaining the vehicle kinematic equations; using the fourth-order Runge-Kutta integral method to discretize the vehicle kinematic equations and obtain vehicle kinematic constraints used to limit the correlation between the system state variables at two adjacent time points.

[0142] The vehicle kinematic equations include:

[0143]

[0144] Where z = [xy θ δ v] T u = [γ a] T ,in, The system state variable at the next moment is represented by z, the system state variable at the current moment is represented by x and y, the horizontal and vertical coordinates of the path planning point are represented by θ, the yaw angle is represented by δ, the front wheel angle is represented by v, the vehicle speed is represented by γ, the rate of change of the front wheel angle is represented by a, the acceleration is represented by L, the wheelbase is represented by T, and the data calculation symbol is "transpose".

[0145] After discretizing the above vehicle kinematic equations using the fourth-order Runge-Kutta integral method, we can obtain:

[0146]

[0147] In the formula, z k Let z represent the system state quantity at time k. k+1 Let represent the system state variables at time k+1, h be the calculation step size, and K1, K2, K3, and K4 be the coefficients of the fourth-order Runge-Kutta integral method.

[0148] In one possible implementation, after obtaining the above-mentioned constraints, a path planning model can be constructed, including:

[0149]

[0150] Among them, Z * and U * Let Z and U represent the system state variables and system control variables of each path planning point in the optimal solution, respectively. k and z N Let u represent the system state variables at time k and time N, respectively. k c represents the system control quantity at time k. k (z k ) and c N (z N Let d represent the penalty function for the system state variables at time k and time N, respectively. k (u k Let z denote the penalty function for the system control quantity at time k. k+1 =f k (z k ,u k ) represents the vehicle's kinematic constraints, g k (z k )≤0 indicates road boundary constraints, h k (u k )≤0 indicates a vehicle control constraint, X init =X 0 An assignment expression representing the initial state of the vehicle.

[0151] S140: Use the target distance to solve the path planning model and determine the optimal solution of the path planning model as the target planning path.

[0152] In practical implementation, augmented Lagrange multiplication algorithm, forward propagation algorithm, and backpropagation algorithm can be used to solve the path planning model by calling the target distance. Among them, the augmented Lagrange multiplication algorithm and the Voronoi region of unstructured roads can transform the constrained optimization problem described by the path planning model into an unconstrained optimization problem. Iteratively solving the unconstrained optimization problem using forward and backpropagation can yield a feasible target planning path.

[0153] The specific solution steps include:

[0154] B1. Initialization: Select the initial point and the outer loop iteration count as variables, and initialize the Lagrange multipliers and penalty factors.

[0155] B2, Outer Loop: Initialize the weights of the regularization terms.

[0156] B3. Inner Loop: Initialize the variable recording the number of iterations for the inner loop. The backpropagation, forward process, and stopping criteria for the inner loop are as follows:

[0157] a. Backpropagation: Calculate the search direction by adjusting the regularization weights.

[0158] b. Forward propagation: Using the golden section search technique to determine the search direction.

[0159] B4. The equality constraints in the path planning model are used to establish the kinematic relationship between discrete states at adjacent time points during backpropagation. The augmented Lagrangian method only needs to transform the inequality constraints in the path planning model into the objective function.

[0160] B5. If the constraint violation degree is less than the set value or the outer loop iteration count record variable is greater than the set value, end the outer loop; otherwise, jump to B2 to continue execution.

[0161] The vehicle path planning method provided in this application not only obtains a high-quality target planning path by comprehensively considering road boundary constraints, vehicle kinematic constraints, and vehicle control constraints without relying on navigation maps, but also improves the efficiency of path planning by quickly obtaining the nearest unstructured road boundary to the path planning point by matching the path planning point with multiple Voronoi regions divided by the unstructured road envelope during the solution process.

[0162] Corresponding to the above method embodiments, another embodiment of this application provides a vehicle route planning device, such as... Figure 11 As shown, the device includes:

[0163] The segmentation unit 210 is used to divide the unstructured road envelope based on the vehicle's forward environment perception into multiple Voronoi regions, wherein the unstructured road envelope is a polygonal envelope enclosed by irregular road boundaries.

[0164] The first determining unit 220 is used to determine the target distance from the path planning point to the target boundary line based on the Voronoi region where the path planning point is located, wherein the target boundary line is the unstructured road boundary in the unstructured road envelope that is closest to the path planning point.

[0165] Building unit 230 is used to build a path planning model based on road boundary constraints, vehicle kinematic constraints, vehicle control constraints, and vehicle initial state;

[0166] Solving unit 240 is used to call the target distance to solve the path planning model;

[0167] The second determining unit 250 is used to determine the optimal solution of the path planning model as the target planning path.

[0168] In one possible implementation, the partitioning unit 210 includes:

[0169] The splitting module is used to split the unstructured road envelope into multiple minimum units according to the concavity and convexity of the vertices, wherein each minimum unit includes a convex edge or multiple consecutive concave edges;

[0170] The module is used to take the smallest unit as the leaf node of the target tree, take the unstructured road envelope as the root node of the target tree, and start from the leaf node, merge the adjacent nodes of the current level as child nodes to obtain the parent node corresponding to the previous level, until the root node is reached, and the target tree is established.

[0171] The generation module is used to, starting from the leaf node in the target tree, first determine the central axis of each node in the current level, then merge the sibling nodes and their central axes in the current level to generate the central axis of the corresponding parent node in the next higher level, until the target central axis of the root node is generated.

[0172] The first determining module is used to divide the unstructured road envelope into multiple Voronoi regions based on the target centerline and the Voronoi edges determined by the auxiliary lines used in generating the target centerline.

[0173] In one possible implementation, the generation module is used for:

[0174] The elements contained in each sibling node in the current level and the central axis originating from each vertex are cached respectively as the cache information corresponding to each sibling node. The elements include convex edges, concave edges and concave points.

[0175] Clear the central axis starting from the common vertex of the sibling nodes in the cache. For each sibling node, starting from the common vertex, traverse the central axis starting from each vertex of the sibling node in the order of the vertices of the sibling node. Regenerate and cache the central axis starting from the common vertex after merging the sibling nodes, and update the central axis starting from the non-common vertex.

[0176] The regenerated central axis originating from the common vertex, the updated central axis originating from the non-common vertex, and the elements contained in all sibling nodes in the current level are used as cache information for the corresponding parent node in the previous level. The central axis originating from the common vertex in the cache information of the parent node is the central axis of the parent node.

[0177] In one possible implementation, the plurality of Voronoi regions include the inner Voronoi regions of the unstructured road envelope, and the first determining unit 220 includes:

[0178] The matching module is used to start from the target centerline segment closest to the path planning starting point, and query forward along the path planning direction to determine whether the path planning point exceeds the target centerline segment and internal Voronoi edge of the currently to-be-determined internal Voronoi region. If the path planning point does not exceed the target centerline segment and internal Voronoi edge of the currently to-be-determined internal Voronoi region, the path planning point is determined to be within the currently to-be-determined internal Voronoi region. If the path planning point exceeds the target centerline segment or internal Voronoi edge of the currently to-be-determined internal Voronoi region, the internal Voronoi region where the next target centerline segment is located is determined as the new currently to-be-determined internal Voronoi region, and the execution step of determining whether the path planning point exceeds the target centerline segment and internal Voronoi edge of the currently to-be-determined internal Voronoi region is returned until the Voronoi region where the path planning point is located is determined.

[0179] The second determining module is used to, after determining the internal Voronoi region where the path planning point is located, take the unstructured road boundary in the determined internal Voronoi region as the target boundary line, and determine the target distance from the path planning point to the target boundary line.

[0180] In one possible implementation, the plurality of Voronoi regions include the inner Voronoi region and the outer Voronoi region of the unstructured road envelope;

[0181] The first determining unit 220 is configured to: locate the inner Voronoi region where the path planning point is located when the path planning point is located inside the unstructured road envelope; locate the outer Voronoi region where the path planning point is located when the path planning point is located outside the unstructured road envelope; after determining the Voronoi region where the path planning point is located, use the unstructured road boundary in the determined Voronoi region as the target boundary line, and determine the target distance from the path planning point to the target boundary line.

[0182] In one possible implementation, the solution unit 240 is used to solve the path planning model by calling the target distance using the augmented Lagrange algorithm, the forward propagation algorithm, and the back propagation algorithm.

[0183] In one possible implementation, the path planning model includes:

[0184]

[0185] Wherein, the Z * and the U * Let Z and U represent the system state variables set and the system control variables set of each path planning point in the optimal solution, respectively. Let Z and U represent the system state variables set and the system control variables set of each path planning point in any solution, respectively. k and the z N Let u represent the system state variables at time k and time N, respectively. k c represents the system control quantity at time k. k (z k ) and the c N (z N ) represent the penalty functions for the system state variables at time k and time N, respectively, where d k (u k ) represents the penalty function for the system control quantity at time k, where z k+1 =f k (z k ,u k ) represents the vehicle kinematic constraints, and g k (z k )≤0 indicates the road boundary constraint, and h k (u k )≤0 indicates the vehicle control quantity constraint, and X init =X 0 An assignment expression representing the initial state of the vehicle.

[0186] The vehicle path planning device provided in this application embodiment can not only obtain a high-quality target planning path by comprehensively considering road boundary constraints, vehicle kinematic constraints, and vehicle control constraints without relying on navigation maps, but also quickly obtain the unstructured road boundary closest to the path planning point by matching the path planning point with multiple Voronoi regions divided by the unstructured road envelope during the solution process. This can improve the solution efficiency of the path planning model and thus improve the efficiency of path planning.

[0187] Based on the above method embodiments, another embodiment of this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method as described in any of the above embodiments.

[0188] Based on the above method embodiments, another embodiment of this application provides an electronic device, such as... Figure 12 As shown, the electronic device includes:

[0189] One or more processors 310;

[0190] The processor 310 is coupled to a storage device 320, the storage device 320 being used to store one or more programs;

[0191] When the one or more programs are executed by the one or more processors 310, the electronic device or computer device performs the method as described in any of the above embodiments.

[0192] Based on the above method embodiments, another embodiment of this application provides a vehicle that includes the device as described in any of the above embodiments, or includes the electronic device as described above.

[0193] Based on the above embodiments, another embodiment of this application provides a computer program product, which includes instructions that, when executed on a computer or processor, cause the computer or processor to perform the method described in any of the above embodiments.

[0194] The above-described apparatus embodiments correspond to the method embodiments and have the same technical effects. For detailed descriptions, please refer to the method embodiments. The apparatus embodiments are derived from the method embodiments; detailed descriptions can be found in the method embodiments section, and will not be repeated here. Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of one embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing this application.

[0195] Those skilled in the art will understand that the modules in the apparatus of the embodiments can be distributed in the apparatus of the embodiments as described in the embodiments, or they can be located in one or more devices different from this embodiment with corresponding changes. The modules of the above embodiments can be combined into one module, or they can be further divided into multiple sub-modules.

[0196] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A vehicle route planning method, characterized in that, The method includes: The unstructured road envelope based on the vehicle's forward environment perception is divided into multiple Voronoi regions, wherein the unstructured road envelope is a polygonal envelope enclosed by irregular road boundaries; Based on the Voronoi region where the path planning point is located, determine the target distance from the path planning point to the target boundary line, wherein the target boundary line is the unstructured road boundary in the unstructured road envelope that is closest to the path planning point; A path planning model is constructed based on road boundary constraints, vehicle kinematic constraints, vehicle control constraints, and the initial state of the vehicle. The target distance is used to solve the path planning model, and the optimal solution of the path planning model is determined as the target planning path.

2. The method according to claim 1, characterized in that, The unstructured road envelope based on vehicle-front environment perception is divided into multiple Voronoi regions, including: Based on the concavity or convexity of the vertices, the unstructured road envelope is divided into multiple minimum units, wherein each minimum unit includes a convex edge or multiple consecutive concave edges. The smallest unit is taken as the leaf node of the target tree, the unstructured road envelope is taken as the root node of the target tree, and starting from the leaf node, the adjacent nodes of the current level are merged as child nodes to obtain the parent node corresponding to the previous level, until the root node is reached, and the target tree is completed. In the target tree, starting from the leaf node, the central axis of each node in the current level is first determined, and then the sibling nodes in the current level and their central axes are merged to generate the central axis of the corresponding parent node in the previous level, until the target central axis of the root node is generated. The unstructured road envelope is divided into multiple Voronoi regions based on the target centerline and the Voronoi edges determined by the auxiliary lines used in generating the target centerline.

3. The method according to claim 2, characterized in that, Merge sibling nodes and their central axes in the current level to generate the central axis of the corresponding parent node in the next higher level, including: The elements contained in each sibling node in the current level and the central axis originating from each vertex are cached respectively as the cache information corresponding to each sibling node. The elements include convex edges, concave edges and concave points. Clear the central axis starting from the common vertex of the sibling nodes in the cache. For each sibling node, starting from the common vertex, traverse the central axis starting from each vertex of the sibling node in the order of the vertices of the sibling node. Regenerate and cache the central axis starting from the common vertex after merging the sibling nodes, and update the central axis starting from the non-common vertex. The regenerated central axis originating from the common vertex, the updated central axis originating from the non-common vertex, and the elements contained in all sibling nodes in the current level are used as cache information for the corresponding parent node in the previous level. The central axis originating from the common vertex in the cache information of the parent node is the central axis of the parent node.

4. The method according to claim 1, characterized in that, The plurality of Voronoi regions include the internal Voronoi regions of the unstructured road envelope. Based on the Voronoi region where the path planning point is located, the target distance from the path planning point to the target boundary line is determined, including: Starting from the target centerline segment closest to the path planning starting point, query forward along the path planning direction to determine whether the path planning point exceeds the target centerline segment and internal Voronoi edge in the current internal Voronoi region to be judged; If the path planning point does not exceed the target centerline segment and the internal Voronoi edge of the currently to-be-determined internal Voronoi region, the path planning point is determined to be within the currently to-be-determined internal Voronoi region; If the path planning point exceeds the target centerline segment or internal Voronoi edge of the currently to be judged internal Voronoi region, the internal Voronoi region where the next target centerline segment is located is determined as the new currently to be judged internal Voronoi region. The execution steps are returned to determine whether the path planning point exceeds the target centerline segment and internal Voronoi edge of the currently to be judged internal Voronoi region, until the Voronoi region where the path planning point is located is determined. After determining the internal Voronoi region where the path planning point is located, the unstructured road boundary in the determined internal Voronoi region is taken as the target boundary line, and the target distance from the path planning point to the target boundary line is determined.

5. The method according to claim 4, characterized in that, The multiple Voronoi regions include the inner Voronoi region and the outer Voronoi region of the unstructured road envelope. Based on the Voronoi region where the path planning point is located, the target distance from the path planning point to the target boundary line is determined, including: If the path planning point is located inside the unstructured road envelope, find the internal Voronoi region where the path planning point is located; If the path planning point is located outside the unstructured road envelope, locate the external Voronoi region where the path planning point is located; After determining the Voronoi region where the path planning point is located, the unstructured road boundary in the determined Voronoi region is taken as the target boundary line, and the target distance from the path planning point to the target boundary line is determined.

6. The method according to claim 1, characterized in that, Solving the path planning model by calling the target distance includes: The path planning model is solved by using the augmented Lagrange algorithm, the forward propagation algorithm, and the backpropagation algorithm, with the target distance being called.

7. The method according to any one of claims 1-6, characterized in that, The path planning model includes: Wherein, the Z * and the U * Let Z and U represent the system state variables set and the system control variables set of each path planning point in the optimal solution, respectively. Let Z and U represent the system state variables set and the system control variables set of each path planning point in any solution, respectively. k and the z N Let u represent the system state variables at time k and time N, respectively. k c represents the system control quantity at time k. k (z k ) and the c N (z N ) represent the penalty functions for the system state variables at time k and time N, respectively, where d k (u k ) represents the penalty function for the system control quantity at time k, where z k+1 =f k (z k ,u k ) represents the vehicle kinematic constraints, and g k (z k )≤0 indicates the road boundary constraint, and h k (u k )≤0 indicates the vehicle control quantity constraint, and X init =X 0 An assignment expression representing the initial state of the vehicle.

8. A vehicle route planning device, characterized in that, The device includes: A segmentation unit is used to divide the unstructured road envelope based on the vehicle's forward environment perception into multiple Voronoi regions, wherein the unstructured road envelope is a polygonal envelope enclosed by irregular road boundaries; The first determining unit is used to determine the target distance from the path planning point to the target boundary line based on the Voronoi region where the path planning point is located, wherein the target boundary line is the unstructured road boundary in the unstructured road envelope that is closest to the path planning point. The building unit is used to construct a path planning model based on road boundary constraints, vehicle kinematic constraints, vehicle control constraints, and the initial state of the vehicle. The solution unit is used to solve the path planning model by calling the target distance; The second determining unit is used to determine the optimal solution of the path planning model as the target planning path.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-7.

10. An electronic device, characterized in that, The electronic device includes: One or more processors; The processor is coupled to a storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the electronic device performs the method as described in any one of claims 1-7.

11. A vehicle, characterized in that, The vehicle includes the device as described in claim 8, or the electronic device as described in claim 10.

Images