Design of an Integral Suboptimal Second-Order Sliding Mode Controller for the Robust Motion Control of Robot Manipulators

Abstract

The formulation of an integral suboptimal second-order sliding mode ((ISSOSM) control algorithm, oriented to solve motion control problems for robot manipulators, is presented in this paper. The proposed algorithm is designed so that the so-called reaching phase, normally present in the evolution of a system controlled via the sliding mode approach, is reduced to a minimum. This fact makes the algorithm more suitable to be applied to a real industrial robot, since it enhances its robustness, by extending it also to time intervals during which the classical sliding mode is not enforced. Moreover, since the algorithm generates second-order sliding modes, while the model of the controlled electromechanical system has a relative degree equal to one, the control action actually fed into the plant is continuous, which provides a positive chattering alleviation effect. The assessment of the proposal has been carried out by experimentally testing it on a COMAU SMART3-S2 anthropomorphic industrial robot manipulator. The satisfactory experimental results, also compared with those obtained with a standard proportional-derivative controller and with the original suboptimal algorithm, confirm that the new algorithm can actually be used in an industrial context.