Path Planning of Redundant Series Manipulators Based on Improved RRT Algorithms

摘要

Aiming at the path planning problem of the redundant manipulator with seven degrees of freedom, this paper presents a new algorithm based on improved RRT (Rapidly-exploring Random Tree algorithm), Gm-RRT algorithm. The new algorithm performs path planning in the joint space of the robot to improve the efficiency of the planning. The new algorithm combines the target probability offset to improve the orientation of the growth of extended random trees; uses the method of generating random points instead of single random points to improve the computational efficiency of path planning; and introduces the method of adaptive virtual gravity of targets to ensure the balance of the new algorithm in target orientation and exploration to the surrounding space. Lastly, a simulation experiment is carried out on the MATLAB platform. The results show that the improved algorithm has advantages in time and keeps the algorithm concise.