ANALYSIS AND SYNTHESIS OF SELF ADJUSTING SYSTEMS USING CORRELATION TECHNIQUES.
Abstract
An analytical study was made of a class of Self-Adjusting Systems, using correlation techniques, with the ultimate objective of synthesizing an adaptive controller for on-line control of dynamical systems. Work was carried out in the following five areas: Comparison of different methods for generating the gradient of a performance index with respect to the adjustable parameters of the system; The study of parameter perturbation methods and the possibility of using a single perturbation signal for generating the entire gradient in parameter space; Study of two time scales and multiple time scales in 'Adaptive' systems caused by slow changes in the environment; Computer simulation of on-line adaptive control systems; A preliminary investigation of stochastic automata theory for use in adaptive control.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 1969
- Accession Number
- AD0690938
Entities
People
- Kumpati S. Narendra
Organizations
- Yale University