Path Planning Rapid Algorithm Based on Modified RRT* for Unmanned Surface Vessel
Aiming at the weak purposiveness of rapidly exploring random tree algorithm in USV path planning,a modified rapid algorithm is proposed.The artificial potential field method is improved and the force analysis in four directions is added to comprehensively calculate the resultant force on USV.The calculation method of steering angle is redefined to avoid entering the local optimal trap and can reach the target point smoothly to obtain an initial path.The initial path is used to set the random point sampling area of rapidly exploring random tree algorithm.By reducing the probability of random points generated in worthless area during sampling,the purpose and timeliness of the algorithm are improved,and the quadratic programming path is obtained.The redundant points of the path planned by rapidly exploring random tree algorithm is removed,in which the path cost can be further reduced while the path nodes are reduced,and the final planned path is obtained.Experimental results show that,compared with the original rapidly exploring random tree algorithm,the modified algorithm can lower the running time and the number of sampling nodes by 84.14%and 70.09%,respectively,when obtaining a path with a similar cost.The proposed algorithm has better quality and higher running efficiency.
NETL
NSTL
万方数据
