Global Dynamic Path Planning based on Fusion Improved A-star And Dynamic Window Method

Abstract

Aiming at the mobile robot can quickly plan out a better global, and can meet the needs of real-time obstacle avoidance, this paper proposes a fusion algorithm based on improved A-star and dynamic window method. Firstly, A-star algorithm is optimized in path search. Although adding heuristic function to A-star algorithm can reduce a large number of useless search nodes, experiments show that there are still many useless search nodes, which increases the search time of the algorithm, Moreover, a weighting factor is introduced into the heuristic function, which makes the improved A-star algorithm further reduce the search useless nodes in the process of global path planning and improve the efficiency of the algorithm. Then aiming at the problem that A-star algorithm is not suitable for real-time obstacle avoidance in dynamic obstacle environment, the dynamic window algorithm is integrated, and the reasonable evaluation function is set according to the actual environment, which increases the flexibility and reliability of the whole system, so that the robot can reach the target point safely and reliably. The simulation results show that: compared with the traditional A-star algorithm, the improved algorithm can plan the optimal global path more quickly, reduce the search time by about 60%, and the path result is smoother and can avoid obstacles in real time. The improved algorithm is tested in the actual environment, and the test results show that it can meet the actual needs.