Obstacle Avoidance and Path Planning for UGVs Using Improved RRT Algorithm

An improved Rapidly-exploring random trees (RRT) algorithm based on the contour of obstacles is proposed to solve the problems of excessive search blindness and easy to fall into local dead zones in the path planning of unmanned ground vehicles (UGVs) in complex environments. The algorithm firstly reduces the rejection rate of new nodes by limiting the sampling range, optimizes the pathfinding direction by combining node expansion with target points, then speeds up the planning speed by combining obstacle contour, and finally reduces the path distance by multi-layer optimization of the planned path. Simulation results show that this algorithm can effectively improve the speed of path planning, shorten the path distance, and is widely applicable to different environments.