UAV Path Planning Algorithm Based on Improved Rapidly-Exploring Random Tree

In this paper, a UAV path planning algorithm based on improved rapidly-exploring random tree(RRT) is proposed to solve the problem of high randomness of sampling points and poor smoothness of planned path in the RRT path planning algorithm. In the process of tree expansion,the detection field is developed to limit the generation range of sampling points, and the generation strategy of sampling points is changed, so that the algorithm has the ability to achieve faster convergence to target and round obstacles. In addition, the improved algorithm also cuts the redundant points of the initial track, and uses the B-spline curve method to improve the smoothness of the track. The simulation results show that the UAV path planning algorithm based on the improved RRT reduces the average path length by 21.75%, the average number of tree nodes by 75.78%, and the average calculation time by 48.04% compared with the traditional RRT UAV path planning algorithm.