Travel control method for vehicle and vehicle

Abstract

The embodiment of the application provides a driving control method of a vehicle and the vehicle, and the method comprises the following steps: acquiring a predicted driving track of the vehicle in a future time period; constructing risk area information of an obstacle based on the predicted driving track and obstacle information of at least one obstacle on a current driving road where the vehicle is located; constructing a search tree based on the risk area information and a current driving state variable of the vehicle; screening a driving path node of the vehicle in the future time period from a plurality of sampling nodes based on a cost index corresponding to each of the plurality of sampling nodes in the search tree; and controlling the vehicle to drive according to the driving state variable corresponding to the driving path node in the future time period. The application solves the technical problem of low reliability of speed limit control of the vehicle in the related art.

Claims

1. A method for controlling the movement of a vehicle, characterized in that, include: Obtain the predicted driving trajectory of the vehicle within a future time period; Based on the predicted driving trajectory and obstacle information of at least one obstacle on the current driving road of the vehicle, risk area information of the obstacle is constructed, wherein there is a collision risk between the obstacle and the vehicle, and the risk area information is used to characterize the area range occupied by the obstacle in the future time period. Based on the risk area information and the vehicle's current driving state variables, a search tree is constructed, wherein the search tree includes multiple sampling nodes, and the sampling nodes are used to characterize the driving state variables of the vehicle at the corresponding time points in the future time period. Based on the cost indicators corresponding to the multiple sampling nodes in the search tree, the driving path nodes of the vehicle in the future time period are selected from the multiple sampling nodes. The cost indicators are used to characterize the comprehensive cost of the vehicle in multiple dimensions when driving according to the driving state variables corresponding to the sampling nodes. Within the future time period, the vehicle is controlled to drive according to the driving state variables corresponding to the driving path nodes.

2. The method according to claim 1, characterized in that, Based on the predicted driving trajectory and obstacle information of at least one obstacle on the current driving road of the vehicle, risk area information of the obstacle is constructed, including: Based on the obstacle information of the obstacle, the obstacle type of the obstacle is determined, wherein the obstacle type is either a static obstacle type or a dynamic obstacle type; Based on the type of obstacle, determine the construction strategy for constructing the risk area information; According to the construction strategy, based on the predicted driving trajectory and the obstacle information of the obstacle, the risk area information of the obstacle is constructed.

3. The method according to claim 2, characterized in that, Based on the obstacle type, a strategy for constructing the risk area information is determined, including: In response to the obstacle type being the static obstacle type, the construction strategy is determined to be the first construction strategy; According to the first construction strategy, based on the predicted driving trajectory and the obstacle information of the obstacle, the risk area information of the obstacle is constructed, including: Based on the obstacle information, the head position and tail position of the obstacle are projected onto the predicted driving trajectory, respectively, to determine the first longitudinal distance between the head and tail of the obstacle on the predicted driving trajectory; and based on the obstacle information, the first lateral distance between the boundary point of the obstacle that is closest to the predicted driving trajectory and the predicted driving trajectory is determined. Based on the first longitudinal distance, the first lateral distance, the angle between the orientation of the obstacle and the tangent direction of the predicted driving trajectory, and the time range corresponding to the future time period, the risk area information of the obstacle is constructed.

4. The method according to claim 2, characterized in that, Based on the obstacle type, a strategy for constructing the risk area information is determined, including: In response to the obstacle type being the dynamic obstacle type, the construction strategy is determined to be the second construction strategy; According to the second construction strategy, based on the predicted driving trajectory and the obstacle information of the obstacle, the risk area information of the obstacle is constructed, including: Based on the obstacle information, a first predicted driving trajectory of the obstacle is predicted within the future time period; Based on the predicted driving trajectory and the first predicted driving trajectory, a second lateral distance between the obstacle and the vehicle in the future time period is determined; The risk area information of the obstacle is constructed based on the second lateral distance between the obstacle and the vehicle in the future time period.

5. The method according to claim 1, characterized in that, Based on the risk area information and the vehicle's current driving state variables, a search tree is constructed, including: Based on the vehicle's current driving state variables, determine the current sampling node; Based on the current driving state variable, identify whether the current sampling node is within the area range corresponding to the risk area information, and obtain the identification result; In response to the identification result indicating that the current sampling node is within the area corresponding to the risk area information, the acceleration sampling step size corresponding to the current sampling node is determined as a first acceleration sampling step size. Based on the first acceleration sampling step size and a preset sampling time interval, at least one extended node of the current sampling node is determined. The extended node includes a driving state variable, a parent node pointer, and a child node pointer. The driving state variable characterizes the driving state of the vehicle at the extended node. The parent node pointer points from the extended node to the previous state node of the extended node, and the child node pointer points from the extended node to the next state node of the extended node. Alternatively, In response to the identification result indicating that the current sampling node is outside the area range corresponding to the risk area information, the acceleration sampling step size corresponding to the current sampling node is determined to be the second acceleration sampling step size, and at least one of the extended nodes of the current sampling node is determined according to the second acceleration sampling step size and the preset sampling time interval, wherein the first acceleration sampling step size is smaller than the second acceleration sampling step size; The extended node is taken as the current sampling node, and the driving state variable of the extended node is taken as the current driving state variable. The following steps are returned to start execution: Based on the current driving state variable, it is identified whether the current sampling node is within the area range corresponding to the risk area information, and the identification result is obtained until the sampling time point corresponding to the extended node is the end time point of the future time period. The tree structure composed of multiple sampling nodes is determined as the initial search tree. The initial search tree is pruned to obtain the final search tree.

6. The method according to claim 5, characterized in that, The initial search tree is pruned to obtain the search tree, including: Based on the driving speed variable in the driving state variables corresponding to the multiple sampling nodes in the initial search tree, a first sampling node whose driving speed variable is greater than a preset maximum speed and a second sampling node whose driving speed variable is less than a preset minimum speed are determined among the multiple sampling nodes. The search tree is obtained by removing the first sampling node and the second sampling node from the plurality of nodes in the search tree.

7. The method according to claim 1, characterized in that, Before selecting the vehicle's travel path nodes for the future time period from the multiple sampling nodes based on the cost indicators corresponding to the multiple sampling nodes in the search tree, the method further includes: The distance cost index, speed cost index, acceleration cost index, and obstacle cost index corresponding to each of the sampling nodes in the search tree are determined respectively. The distance cost index is used to characterize the longitudinal distance deviation between the sampling node and the end point of the predicted driving trajectory, and the corresponding path completion penalty cost. The speed cost index is used to characterize the speed deviation between the speed variable corresponding to the sampling node and the speed limit value of the lane in which the vehicle is traveling, and the corresponding speed compliance and comfort cost. The acceleration cost index is used to characterize the absolute value of the acceleration corresponding to the sampling node, and the corresponding longitudinal handling change penalty cost. The obstacle cost index is used to characterize the distance between the sampling node and the obstacle, and the corresponding safety cost to the driver of the vehicle. The cost index is obtained by fusing the distance cost index, speed cost index, acceleration cost index and obstacle cost index.

8. The method according to claim 7, characterized in that, Based on the cost indicators corresponding to the multiple sampling nodes in the search tree, the driving path nodes of the vehicle in the future time period are selected from the multiple sampling nodes, including: Based on the cost indicators corresponding to the multiple sampling nodes in the search tree, the path costs corresponding to the multiple leaf nodes at the end of the search tree are determined, wherein the path cost is the travel cost corresponding to the travel path formed from the root node to the leaf node in the search tree; Based on the path costs corresponding to the multiple leaf nodes, the target leaf node with the lowest path cost is selected from the multiple leaf nodes; Backtracking is performed based on the parent node pointer corresponding to the target leaf node until the root node of the search tree is reached; The root node, the target leaf node, and the sampling node between the root node and the target leaf node are determined as the driving path nodes.

9. The method according to any one of claims 1 to 8, characterized in that, Controlling the vehicle's movement according to the travel path nodes within the future time period includes: The vehicle's speed is controlled according to the driving speed variable in the driving state variables corresponding to the driving path nodes within the future time period.

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

Images