Sample-based Path Planning for Small UAV Obstacle Avoidance

Abstract

The obstacle avoidance path planning ability of small UAV is the basis for its safe flight. In recent years, the sampling-based obstacle avoidance path planning algorithm has been widely used because of its superior performance. For example, RRT* algorithm can guarantee probability completeness while possessing asymptotic optimality. However, the addition of optimization process reduces the convergence rate of the algorithm. In order to ameliorate this problem, an improved RRT* algorithm based on biased sampling is proposed in this paper. The algorithm improves the convergence speed of the algorithm and accelerates the obstacle avoidance path search by performing the partial concentration sampling in the relative area of the goal point and the path point. The simulation results show that the proposed algorithm can obtain an optimized small UAV obstacle avoidance path in a shorter time.