UAV Track Planning Based on IRRT Algorithm

To overcome the problems of low search efficiency, many redundant points and large turning angle in the track planning process of unmanned aerial vehicle (UAV) with Rapidly-exploring Random Trees (RRT) algorithm, This paper presents an Improved Rapidly-exploring Random Trees (IRRT) algorithm, First, the target probability sampling method is used to reduce the number of invalid sampling points; Secondly, the strategy of target oriented expansion is used to generate new nodes to enhance the orientation of the target; Then, Dijkstra algorithm is used to optimize the searched feasible path, and the redundant points in the track are removed to obtain a shortest track; Finally, the Bezier curve is used to smooth the shortest track and smooth the larger corner in the shortest track to get a track that meets the kinematic flight of UAV, In simple environment, narrow passage environment and a large number of complex irregular obstacles, Several experiments were conducted in simple environment, narrow channel environment and a large number of complex irregular obstacles, the results show that the IRRT algorithm performs better than the basic RRT algorithm in terms of node utilization, running time and track length, thus verifying the effectiveness of the algorithm.