Investigations into Boundedness Properties and Nonlinear Effects of Adaptive Output Error Estimation Algorithms

摘要

The thesis examines local instability and global asymptotic stability of both discrete-time and continuous-time output error adaptive estimation algorithms. By analyzing both local instability and global stability of these schemes with respect to adaptation gain, two boundedness conjectures are proposed which state that all signals within these adaptive systems as well as the output error and parameter estimates remain globally bounded for all time despite any locally unstable behavior. This conjecture, though not formally proven, is motivated through the use of several examples and heuristic arguments based on known local and global theory of adaptive systems.