Cable Attachment Optimization for Reconfigurable Cable-Driven Parallel Robots Based on Various Workspace Conditions
Hung Hon Cheng, Darwin Lau
Abstract
This article proposes a novel method to determine the optimal cable attachment configuration for reconfigurable cable-driven parallel robots (RCDPRs) considering different workspace conditions. It is shown that wrench-feasible, wrench-closure, and interference-free conditions can be formulated into inequality constraints by considering the cable attachment points as polynomial functions or variables. Furthermore, the proposed method determines the optimal cable attachment location that minimizes the cable force or maximizes the tension factor kinematically at each pose. The proposed formulation can be resolved by different optimization techniques, such as semidefinite programming relaxation and multivariable gradient-based optimization solvers. The proposed approach can be implemented on RCDPRs from low to high degrees-of-freedom and a wide range of obstacles. The proposed formulation can be widely applied for different reconfiguration mechanisms, such as rails, UGV, and UAV.