MDA + RRT: A general approach for resolving the problem of angle constraint for hyper-redundant manipulator. (1st May 2022)

• Record Type: Journal Article
• Authors: Jia, Longfei; Huang, Yuping; Chen, Ting; Guo, Yaxing; Yin, Yecheng; Chen, Jing
• Is Part Of: Expert systems with applications. Volume 193(2022)
• Publication Date: 2022-05-01
• Subjects: Hyper-redundant manipulator -- RRT algorithm -- Maximum deflection angle -- Path planning; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2021.116379
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: The MDA + RRT algorithms guarantee that planned path satisfies the constraint of angle. Compared with other algorithms, this algorithm does not increase the computation amount. The feasibility and optimization of MDA + RRT are preferable to the corresponding RRT. This method is a universal method that can be associated with any kind of RRT algorithm. Abstract: In this paper, a cable-driven hyper-redundant manipulator with 17 degrees of freedom is presented for narrow and complex environments firstly. Based on the mechanical structure, the MDA + RRT algorithms is proposed for path planning considering the maximum deflection angle of joint. First, the transformation relationship between the path parameters and manipulator parameters is obtained through the theorem of sine of triangles and two extreme points, which can convert limitation of the deflection angle of the joint to the limitation of the deflection angle of the path. Then, three corresponding improved algorithms are proposed: MDA-RRT, MDA-RRT*, MDA-QRRT* on the basis of the traditional RRT, RRT* and Q-RRT* algorithms. Finally, six algorithms are applied to plan the path of avoiding three different obstacles respectively by simulation. The results demonstrate that, three improved algorithms can guarantee that the planned path satisfies the constraint of deflection angle of joint without increasing the computational amount compared with the three traditional algorithms, only by limiting the selection range of the next random vertex. Among the six algorithms, the feasibility and optimization of the MDA-QRRT* algorithm are the best and the feasibility and optimization of the three improved algorithms are also better than the corresponding traditional algorithms.