Research on Obstacle Avoidance Path Planning of Intelligent Transfer Device Based on Improved RRT Algorithm

As an important assistive device for people with lower limb dysfunction, the intelligent transfer device (IDS) has high requirements for path planning safety and movement efficiency during autonomous navigation. In obstacle spatial path planning, the rapidly exploring random trees (RRT) algorithm has the problems of environment exploration blindness and slow convergence speed. Therefore, a path planning method that improves the RRT algorithm is proposed. Firstly, a sampling strategy combining restricted region search and probabilistic goal bias is used to improve the orientation of path planning. Then, a dynamic variable step size expansion strategy is adopted to solve the poor convergence speed of the random tree. Finally, the path node reconnection and fitting optimization based on safe distance is proposed to solve the redundancy of nodes and the discontinuity of curvature in the initial path. The results show that the improved RRT algorithm reduces the average search time by 99.70%, 61.56%, and 60.22%, respectively, compared to the RRT, RRT*, and Informed RRT* algorithms.