"A new model reference adaptive control design method with guaranteed transient performance using neural networks is proposed in this thesis. With this method, stable tracking of a desired trajectory is realized for nonlinear system with uncertainty, and modified state observer structure is designed to enable desired transient performance with large adaptive gain and at the same time avoid high frequency oscillation. The neural network adaption rule is derived using Lyapunov theory, which guarantees stability of error dynamics and boundedness of neural network weights, and a soft switching sliding mode modification is added in order to adjust tracking error. The proposed method is tested by different theoretical application problems simulations, and also Caterpillar Electro-Hydraulic Test Bench experiments. Satisfying results show the potential of this approach"--Abstract, page iv.
Balakrishnan, S. N.
Sarangapani, Jagannathan, 1965-
Landers, Robert G.
Mechanical and Aerospace Engineering
M.S. in Mechanical Engineering
National Science Foundation (U.S.)
Ames Research Center
Missouri University of Science and Technology
Journal article titles appearing in thesis/dissertation
- Development and time domain analysis of a new model reference adaptive controller.
- New model reference adaptive controller in missile autopilots design.
- Electro-hydraulic piston control using a new model reference adaptive controller
x, 92 pages
© 2010 Yang Yang, All rights reserved.
Thesis - Open Access
Library of Congress Subject Headings
Adaptive control systems -- Design
Neural networks (Computer science) -- Design
Print OCLC #
Electronic OCLC #
Link to Catalog Record
Yang, Yang, "Neural MRAC based on modified state observer" (2010). Masters Theses. 5433.