DEVELOPMENT OF TERMINAL SLIDING MODE CONTROL METHODS FOR UNCERTAIN NONLINEAR SYSTEMS AND THEIR APPLICATIONS TO ROBOTIC MANIPULATORS

Abstract

The study reported in this thesis develops TSMCs for uncertain nonlinear systems and their applications to robotic manipulators that plays a very critical role in modern control technology. The central motivation of this thesis is to significantly improve trajectory tracking precision and to overcome the limitations of SMC-based methods and TSMC-based methods for several classes of uncertain nonlinear systems in presence of external disturbances and uncertain dynamics, or even undesired faults. These proposed control methodologies are developed based on SMC, SC, TSMC, NFTSMC, FLS, STA, NNs, observer-based controllers, and AC. The fundamental theoretical procedure is the foundation of the asymptotic stability based Lyapunov theory underpinned by the Lipschitz condition in the ordinary differential equations and finite time control method. The main applications of the proposed control methodologies are to apply to uncertain mechanical systems and robotic systems, in which external disturbances and uncertain dynamics are required to be bounded and to satisfy the suitable condition.

The proposed control algorithms are designed to achieve the following major advantages such as simple design, fast transient response, defined time convergence, robustness against uncertainties, high tracking accuracy, and stabilization with small steady-state errors. These proposed control algorithms can reject some/all of the limitations in conventional SMC or TSMC such as reaching phase glitch and the singularity problem. They can also avoid/ eliminate/ attenuate the effects of chattering behavior and the requirement for prior information about the upper bound of external disturbances and uncertain dynamics as well as the necessity for an exact mathematical model. Especially, some the designed controllers have estimate ability and fault tolerance.

The proposed control algorithms were applied for trajectory tracking control and FTC of parallel and serial robotic manipulators, or synchronization problem in motion controlling. The computer numerical simulation and experiment results are performed for 2-DOF planar parallel manipulator, 3-DOF planar parallel manipulator, 2-DOF serial robotic manipulator, and 3-DOF Puma560 robot manipulator to demonstrate the effectiveness and applicability of the proposed systems and to validate the theoretical derivation. Moreover, the designed control methodologies can be extended their applications to uncertain high-order MIMO systems.