Cooperative Path Planning for Multiple UAVs Based on APF B-RRT* Algorithm

Abstract

Aiming at the path planning problem of an unmanned aerial vehicle (UAV) in a complex unknown environment, this paper proposes a cooperative path planning algorithm for multiple UAVs. Using the local environment information, several rolling path plannings are carried out by the Artificial Potential Field Bidirectional-Rapidly exploring Random Trees (APF B-RRT*) algorithm. The APF B-RRT* algorithm optimizes the search space by pre-sampling and adapts with an adaptive step while fusing with the APF algorithm for guiding sampling. Then, the generated path is trimmed and smoothed to obtain the optimized path. Then, through the sampling constraint, several paths can be planned at the same time, which are guaranteed not to collide. The model predictive control (MPC) is used to realize the cooperative control of the UAVs, that is, the UAVs reached the destination simultaneously along the planned path. This algorithm achieves some progress in solving the problems of slow convergence speed, an unstable result and an unsmooth path in UAV path planning. Simulation and comparison show that the APF B-RRT* algorithm has certain advantages in algorithm performance.