Time delay control with integral sliding surface and its application to an excavator system

Abstract

The robust control method is required for controlling nonlinear systems. For example, the control of robotic excavator is difficult from the standpoint of the following problems: parameter variations in mechanical structures, various nonlinearities in hydraulic actuators, and disturbance due to the contact with the ground. In mechanical structures, the inertial force and gravitational force varies largely with joint motions. Hydraulic actuators, massively coupled and complexly connected, have various nonlinear components. For such reasons, the robust controller is required for obtaining the good control performance in nonlinear systems. Recently the Time Delay Control with Switching Action (TDCSA) method has been proposed as a promising technique in the robust control area, where the plant has unknown dynamics with parameter variations and substantial disturbances are present. More specifically, TDCSA consists of time delay control (TDC), which estimates the amounts of an unknown nonlinear dynamics and unexpected uncertainties and cancels it, and a switching action based on sliding mode control (SMC). The switching action, the discontinuous input of sliding mode control, keeps the tracking errors, which is occurred by the time delay estimation error, on the predefined sliding surface and improves the robustness. Its effectiveness has been demonstrated through a DC motor and a linear motor. This thesis attempted to remedy the problems of TDCSA using an integral sliding surface and to show the excellence of this controller. Firstly, when TDCSA is applied to nonlinear system having saturation element and high frequency resonances, TDCSA reveals two problems: the one is windup problem and other is chattering problem or undesired vibration. These two problems should be solved. Secondly, the effectiveness of TDCSA has been shown through experiments on simple nonlinear system like as motors. So experiments on highly nonlinear systems are necessary to s...