Unmanned vehicle path planning method and device

Abstract

The embodiment of the invention provides an unmanned vehicle path planning method and device. According to the method, high-precision map information is analyzed, repulsion is calculated for each gridaccording to map obstacle information after the map is rasterized, the repulsion serves as the searching cost, the obstacle avoidance efficiency of a hybrid A* algorithm can be effectively improved,and the generated trajectory conforms to vehicle dynamics constraints. Meanwhile, the generated trajectory is smoothed by using a gradient descent algorithm, and finally an optimal trajectory is output. Vehicle kinematics constraints are considered, so that the planned path has drivability; a potential field is calculated for each grid, so that the path searching time can be reduced to a certain extent; the potential field method enhances the tolerance of sensing errors to a certain extent, and improves the robustness of path planning; different repulsion levels are allocated to different types of obstacles, and repulsion weights are dynamically set for the boundary width of the lane-level road; and the gradient descent smoothing module is adopted to smooth the generated path, so that thepath is more suitable for vehicle driving.

Description

technical field [0001] The embodiments of the present invention relate to the technical field of automatic driving, and in particular to a method and device for path planning of an unmanned vehicle. Background technique [0002] Path planning is a very important part in the driving process of self-driving vehicles. During the walking process, unmanned vehicles need to plan a trajectory that can avoid obstacles and meet the vehicle dynamics constraints. It is a path tracking control module. Foundation. [0003] Traditional path planning algorithms usually use A* algorithm, potential field method, RRT algorithm, etc. to plan paths based on map obstacle information. These algorithms do not consider vehicle dynamics parameters, and the generated trajectories cannot satisfy the constraints of vehicle dynamics parameters, so autonomous vehicles may not be able to drive steadily according to the established trajectory. The traditional hybrid A* algorithm considers vehicle dynamic...

AI Technical Summary

Problems solved by technology

These algorithms do not consider vehicle dynamics parameters, and the generated trajectories cannot satisfy the constraints of vehicle dynamics parameters, so autonomous vehicles may not be able to drive steadily according to the established trajectory

The traditional hybrid A* algorithm considers vehicle dynamics during the search process, but does not consider the cost priority of each grid during the search process, resulting in low search efficiency and the generated trajectory is not smooth enough

Images