Path Planning Algorithm for Mobile Robots based on A* Algorithm with Adaptive Search Strategy

Abstract

Path planning is one of the core functions of autonomous navigation. A* algorithm is widely used in the study of path planning for robots. Traditional A* algorithm is limited by the direction of node expansion. In some cases, it would generate some unnecessary bends, which makes the path not suit the motion characteristics of the robot and results in bad affection on its path. An improved A* algorithm with search strategy is proposed. The strategy of neighborhood point searching for A* algorithm is presented, so that A* algorithm has a more reasonable node search direction in a wider area. To further solve the unfavorable bending problem, the heuristic function takes the turns of the path into account in the cost function, which can reduce certain types of path turning that affect the robot’s motion. Finally, the improved A* algorithm is verified under simulation environment. Comparison tests with traditional A* algorithm, the path length of the proposed algorithm is reduced by 13.0 %, and the number of turns is reduced by 42.9 %. Compared with Dijkstra algorithm, the total time cost of the proposed algorithm is shortened by 12.3 %.