UAV Dynamic Path Planning using Anytime Repairing Sparse A* Algorithm and Targets Motion Estimation (IEEE/CSAA GNCC)

This paper presents an effective method for unmanned aerial vehicle (UAV) dynamic path planning considering moving-target and obstacle-avoidance constraints. In the process of dynamic path planning, moving-target positions are predicted using Kalman filtering algorithm on the receding horizon. Then, anytime repairing sparse A
^*
algorithm (AR-SAS) is customized to generate feasible paths from the staring positions to the moving-target positions. By rolling planning, UAVs are able to track the moving targets efficiently. Simulation results demonstrate that the proposed method can generate feasible paths within limited time and has high robustness.