Smart Vehicle Path Planning Based on Modified PRM Algorithm

Aiming at the problems of lack of guidance in selecting sampling points, low reuse rate of road signs and low search efficiency of probabilistic map algorithm（PRM）, a pseudo-random sampling strategy based on the spatial main axis as the reference axis was designed, which optimized the generation mode of sampling points and removed redundant sampling points. The distance threshold between road points was set, and the two-way incremental method was used for collision detection, which optimized the call times of collision detection. The key points of the planned path were extracted as the discrete control points of the Bezier curve, and the path was smoothed to make the generated path more in line with the driving conditions of the vehicle. MATLAB and ROS were used to build a test platform to verify and analyze the correctness of the modified PRM algorithm. Compared with the basic PRM algorithm, the modified PRM algorithm had obvious advantages in the construction speed of road map, path planning speed and path length.