UAV Path Planning Based on Improved A ∗ and DWA Algorithms

Abstract

This work proposes a path planning algorithm based on A $\ast$ and DWA to achieve global path optimization while satisfying security and speed requirements for unmanned aerial vehicles (UAV). The algorithm first preprocesses the map for irregular obstacles encountered by a UAV in flight, including grid preprocessing for arc-shaped obstacles and convex preprocessing for concave obstacles. Further, the standard A $\ast$ algorithm is improved based on UAV’s flight environment information and motion constraints. Further, the DWA algorithm’s limitations regarding local optimization and long planning time are mitigated by adaptively adjusting the evaluation function according to the UAV’s safety threshold, obstacles, and environment information. As a result, the global optimal path evaluation subfunction is constructed. Finally, the key points of the global path are selected as the subtarget points of the local path planning. Under the premise of the optimal path, the UAV real-time path’s efficiency and safety are effectively improved. The experimental results demonstrate that the path planning based on improved A $\ast$ and DWA algorithms shortens the path length, reduces the planning time, improves the UAV path smoothness, and enhances the safety of UAV path obstacle avoidance.