Adaptive Control of Systems with Uncertain Nonlinearities.
Abstract
Our broad research program in robust and adaptive control has culminated in the development of three new design methodologies presented in two books totaling 850 pages, four book chapters, 26 major journal publications, and more numerous conference papers. Our new inverse optimality approach to robust nonlinear control avoids the need to solve the HJI equation, but still achieves the desirable properties of optimal systems. A systematically constructed robust Lyapunov function is at the same time the upper value function of a meaningful differential game. For systems with unknown nonsmooth nonlinearities in actuators and sensors - hysteresis, dead-zone and backlash - we have developed an adaptive inverse approach. Adaptive controllers of this type compensate for the undesirable acts of the nonlinearities and with imperfect components achieve high tracking performance. Modular adaptive designs lead to new controllers applicable to larger classes of nonlinear systems. In the linear case, when other adaptive schemes are also applicable, our designs result in superior transient and steady-state performance. The results of this research have been well received by the research community and industrial applications are beginning to appear.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 30, 1995
- Accession Number
- ADA304391
Entities
People
- Peter V. Kokotovic
Organizations
- University of California, Santa Barbara