Some generalizations to the theory of adaptive robust control and its application

Amit Mohanty, Purdue University

Abstract

The research work presented in this dissertation focuses on some generalizations to adaptive robust control (ARC) of a class of uncertain nonlinear systems transformable to parametric semi-strict feedback form with unknown rates of states gains. The class of systems considered in this work is larger than the class of systems considered previously in the literature. Special emphasis was put on the generalizations of indirect adaptive robust control (IARC) and integrated direct/indirect adaptive robust control (DIARC) algorithms. Two application examples—one for generalized IARC and one for generalized DIARC—are considered. First, the effectiveness of the IARC algorithm in obtaining better parameter estimates as compared to the direct adaptive robust control (DARC) algorithm is demonstrated by applying the generalized IARC algorithm for trajectory tracking control of an electro-hydraulic robotic manipulator. Second, the effectiveness of generalized DIARC in obtaining better tracking accuracy as compared to generalized IARC algorithm is shown by applying both the algorithms for trajectory tracking of an electro-hydraulic robotic arm with input valve deadband. In the second part the dissertation, indirect adaptive robust control of a class of MIMO systems are considered. These uncertain nonlinear MIMO systems are motivated from practical mechanical engineering systems, such as robotic manipulators. Finally, different areas for extended applications of ARC-based controller and observer designs were explored. Specifically, an adaptive robust observer was designed and implemented for velocity estimation on an electro-hydraulic system and an adaptive compensation-based sensor fault-tolerant adaptive robust controller was designed for a class of linear systems in a parametric output feedback form.

Degree

Ph.D.

Advisors

Yao, Purdue University.

Subject Area

Mechanical engineering

COinS