Optimization and analysis of a redundant 4R spherical wrist mechanism for a shoulder exoskeleton

This paper presents a redundant 4-revolute (4R) spherical wrist mechanism for a shoulder exoskeleton, which overcomes several major issues with the 3R mechanisms used in the past. An analysis of the 3R mechanism is done to highlight the limitations in its range of motion and problems caused by operating near singular configurations. To ensure that the redundancy in the 4R mechanism is efficiently utilized, genetic algorithm is used to optimize the mechanism design and identify the optimal operating configurations of the mechanism. The capability to reach the entire shoulder workspace is guaranteed and the joint velocities are minimized by considering the joint displacements required to move the end-effector throughout the workspace and the condition number of joint configurations for reaching 89 positions in the workspace. Analysis of the 4R mechanism obtained from the optimization process indicates that it can move throughout the entire shoulder workspace with feasibly low joint velocities.