Wrench Feasibility Workspace Analysis and Adaptive Rotation Algorithm of Cable-Driven Upper Limb Rehabilitation Robot

Cable-driven rehabilitation robot is an important branch of cable-driven parallel robots (CDPRs). The ability of CDPRs to generate wrench determines their performance in task execution, and some CDPRs employ reconfigurable structures to enhance their wrench capabilities. This paper proposes a cable-driven 4- degrees of freedom (DOF) upper limb rehabilitation robot with an adaptive dynamic structure (DAS) to alter the distribution of cable attachment points (CAPs). Then the available wrench set (AWS) and wrench feasibility workspace (WFW) of the robot are analysed. Furthermore, an adaptive rotation algorithm based on Bayesian optimization is proposed to adjust the rotation angle of the DAS. Thereby modifying the distribution of the CAPs, and significantly improving the WFW of robot. Examples of simulation are presented to demonstrate the effectiveness of the adaptive rotation algorithm in increasing the robot's WFW.