Adaptive Robust Posture Control of Parallel Manipulator Driven by Pneumatic Muscles With Redundancy

Abstract

This paper presents an adaptive robust posture controller for a parallel manipulator driven by pneumatic muscles (PMDPM) with a redundant DOF. Rather severe parametric uncertainties and uncertain nonlinearities exist in the dynamics of the PMDPM. To deal with these uncertainties effectively, the recently developed adaptive robust control strategy is applied. Furthermore, the developed control strategy explicitly takes into account the particular physical properties of the system studied. Specifically, the symmetric geometric structure of the parallel manipulator driven by identical pneumatic muscles and no external moments around symmetry-axis direction of the parallel manipulator make the rotation angle of the parallel manipulator around its symmetry axis direction negligible and a nonfactor during normal operations. As such, the axial rotation angle is not measured and controlled when the PMDPM are used in practice, leading to a single-DOF redundancy in synthesizing the precise posture controller for rotation angles around other axes. To make full use of this redundancy, an equivalent average-stiffness-like desired constraint is introduced in the development of the adaptive robust posture controller to achieve precise posture tracking while reducing control input chattering caused by measurement noise. Experimental results are obtained to verify the validity of the proposed controller for the redundant PMDPM.