Nonlinear Robust Control Theory and Applications
Abstract
Model based control methods are commonly used in the design of large, complex systems. Specifically, a mathematical model of the system is constructed, utilizing, for example, first principles analysis and experimental data, which is then used for subsequent control system design and analysis. For the purposes of feedback control highly accurate models are desired. However, such accuracy often requires that complicated high order models be used, which in turn lead to more difficult control design problems from both an engineering and a computational perspective. The emphasis of this research is on the development of methods for reducing the size and complexity of the model while retaining the essential features of the system description. The main goal of these methods is to find a simplified system model which describes the physical system accurately enough so that controllers designed based on this simplified model perform well when implemented on the real system. Directly related to the topic of model reduction are the realization theory concepts of minimality and its converse reducibility, which are also addressed in detail in this thesis.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 18, 1997
- Accession Number
- ADA336238
Entities
People
- John Doyle
Organizations
- California Institute of Technology