Operational-space wrench and acceleration capability analysis for multi-link cable-driven robots

Abstract

Multi-link cable-driven robots (MCDRs) have potential advantages in confined spaces exploration because of their redundancy and flexibility. Operational space wrench and acceleration capability analysis of MCDRs is important for their design, manipulability optimization, and motion planning. However, existing works mainly focus on capability analysis in the joint space. In this paper, we present a zonotope-based iterative method and a simplified capability zonotope to analyze the operational-space wrench and acceleration capability of MCDRs. In the iterative method, the capability generated by some cables can be iteratively added to the initial capability zonotope based on the Minkowski sum. In the simplified zonotope capability representation, a threshold is put forward to reduce redundant vertices and faces with little volume loss. Finally, simulations on a 24 DOFs MCDR are performed to verify the effectiveness of the developed method. The results demonstrate that our iterative algorithm can easily generate the capability zonotope with a few MB ROM, while traditional operational wrench capability evaluation without our iterative algorithm needs 18432 GB ROM. Furthermore, our simplified representation reduces the vertices and faces from 1260 and 2516 to 88 and 172, respectively, but with only 3.3% volume loss, which decreases the constraints of the robot and is conducive to manipulability optimization and motion planning.