Path Planning of Manipulator Based on Improved RRT-Connect Algorithm
Ying Zhu, Yanfeng Tang, Yawan Zhang, Yujun Huang · 2021
Aiming at the problems of randomicity, blindness and low efficiency of RRT-Connect Algorithm, an improved RRT-Connect Algorithm is proposed in this paper. First, the idea of artificial potential field is introduced to guide the random tree to grow in the direction of the target, to reduce the blindness of path searching, to increase the guiding force and to accelerate the planning speed, finally, the b-spline curve is used to smooth the planned path, to eliminate the non-smooth points in the path, and to obtain the continuous trajectory of the manipulator, it can avoid the problems such as the quick change of direction when the manipulator performs the grasping task, improve the execution precision and prolong the service life of the manipulator. Matlab simulation results show that this algorithm has some advantages in the path planning of complex environment. Compared with the traditional RRT-Connect, the improved RRT-Connect can complete the path planning task accurately and efficiently.