Input conditions for continuous-time adaptive systems problems

S Dasgupta; B.D.O Anderson; A.C Tsoi

doi:10.1109/9.45148

Back

Input conditions for continuous-time adaptive systems problems

Journal article

Peer reviewed

Input conditions for continuous-time adaptive systems problems

S Dasgupta, B.D.O Anderson and A.C Tsoi

IEEE transactions on automatic control, Vol.35(1), pp.78-82

01/1990

DOI: 10.1109/9.45148

View Online

Abstract

The authors develop persistence-of-excitation conditions for the exponential convergence of continuous-time adaptive algorithms. Exponential convergence is important for robustness. Adaptive algorithms without such convergence can behave unacceptably in the presence of modeling inadequacies. Conditions for convergence are usually framed as spanning conditions on a regressor vector involving the output of the unknown system. In this study the authors translate these conditions into ones involving the system input only.< >

Feedback loop

Programmable control

Adaptive systems

Adaptive algorithm

Transfer functions

Control systems

Robustness

Adaptive control

Frequency domain analysis

Convergence

Details

Title: Subtitle: Input conditions for continuous-time adaptive systems problems
Creators: S Dasgupta - Dept. of Electr. & Comput. Eng., Iowa Univ., Iowa City, IA, USA
B.D.O Anderson
A.C Tsoi
Resource Type: Journal article
Publication Details: IEEE transactions on automatic control, Vol.35(1), pp.78-82
Publisher: IEEE
DOI: 10.1109/9.45148
ISSN: 0018-9286
eISSN: 1558-2523
Language: English
Date published: 01/1990
Academic Unit: Electrical and Computer Engineering
Record Identifier: 9984083818002771

Metrics

7 Record Views

7 Times Cited - Web of Science