Nonlinear Robust Control of Unknown Robot Manipulator Systems With Actuators and Disturbances Using System Identification and Integral Sliding Mode Disturbance Observer

Abstract

In this study, nonlinear robust control of robot manipulator systems with completely unknown manipulator dynamics using a system identification and integral sliding mode disturbance observer (SI-ISMDOB)-based sliding mode control (SMC) method is proposed. Because the unknown robot manipulator system - owing to the actuator dynamics, unknown manipulator parameters, frictional forces, and unknown load torques - cannot be controlled using the existing model-based methods, they need to be identified; subsequently, the identified manipulator system should be controlled in a robust manner. To this end, the manipulator-actuator dynamics at the motor shaft are first proposed by considering the unknown robot manipulator dynamics as torque disturbances and are subsequently identified using the system identification. An SI-ISMDOB-based SMC method is then developed using the identified manipulator-actuator dynamics and torque-disturbance estimates from the ISMDOB. The simulation and experimental results on an actual planar manipulator system are presented to verify the improved performance of the proposed method over the SMC method and SMDOB-based SMC method via a performance indicator.