Control strategies for enhancing manipulability in tensegrity-based redundant robots and manipulators

Abstract

This paper explores the potential of movable tensegrities and nonlinear cable-driven structures for designing lightweight robots. These mechanisms exhibit both drive and degrees of freedom redundancy, allowing additional conditions on pure end-effector motion control. The study focuses on using the redundancy in tensegrity mechanisms to maximize manipulability within a workspace. The methodology is demonstrated using a planar two-stage cable-driven tensegrity robot as a benchmark. The study begins with a description of self-stress analysis. Actuation planning is adapted for workspace exploration and position interpolation in trajectory planning. The control strategy for the manipulator involves trajectory planning, motion control, and implementing computed torque control. Actuator redundancy is addressed using singular value decomposition and the least squares method. Controller gains are optimized based on different test trajectories. A key contribution of this study is the development of a manipulability optimization methodology based on nonlinear dynamics. After meeting the end-effector motion requirements, the objective function, created by combining extreme singular values and the condition number of the Jacobian matrix, is optimized. Simulation experiments demonstrate the robustness of the algorithm, showing a significant improvement in the objective function.