Nonlinear control of flexible link robots and nonholonomic robotic systems

Abstract

In this thesis, effective nonlinear control schemes are investigated for two typical robotic systems: flexible link robots and nonholonomic robotic systems, by utilizing the inherent properties of the system dynamics, which are of academic interests and practical values. Derived from the basic energy-work relationship of flexible robot systems, adaptive robust and model-free control strategies are proposed to eliminate the problems associated with truncated model-based controllers. Employing the active vibration suppression ability of smart materials, composite control strategies are presented for the tracking control of flexible robots. All the strategies are applicable to both single-link and multi-link cases. In the second part of the thesis, state and output feedback stabilizations are presented for uncertain nonholonomic systems firstly. To address the nonholonomic control problem at the dynamic level, two different control strategies are presented for uncertain nonholonomic robotic systems with different configurations. Desired performances are achieved using the proposed controllers.