Dynamic Path Planning of Robots Based on Improved Bi-RRT and Dynamic Window Method
In complex environments,the bidirectional fast expanding random tree(Bi-RRT)algorithm,exists problems of low sampling efficiency,redundant nodes and inability in avoiding obstacles,hence,an algorithm based on the fusion of improved Bi-RRT and improved dynamic window method(DWA)is proposed.Firstly,the improved Bi-RRT algorithm adopts the strategy of bidirectional adaptive expansion,which accelerates the convergence speed and improves the sampling efficiency,and introduces the node optimal selection and path optimization strategy based on the A*algorithm to improve the goal orientation and shorten the path length.Then,the evaluation function in the DWA algorithm is improved,and the global yaw angle and global path distance are added to the evaluation function,so as to diversify the evaluation of the trajectory direction and realize the efficient dynamic obstacle avoidance of the robot.Finally,the improved Bi-RRT and improved DWA algorithm are fused,based on MATLAB and ROS platform,the fused algorithm is simulated to verify that the proposed algorithm has a short planning path,high efficiency and can effectively avoid unknown obstacles.
万方数据
维普
