Path Planning Research for Outdoor Mobile Robot

Abstract

We present an improved A* algorithm based on road boundary constrainted for road scenes such as campuses, communities, and industrial parks. Firstly, the LIDAR scans the surrounding environment and saves it in the point cloud data, pre-processes the point cloud and acquires the road boundary, then translates the acquired road boundary to generate a minimal cost feasible domain, and finally adds the feasible domain to the A* algorithm for path planning. To verify the feasibility of the method, experiments were conducted in both simulated and real environments. The experimental results show that the method can improve the problem of the paths planned by the A* algorithm being too close to the road edges and reduce the influence of road boundaries when the mobile robot is moving outdoors.