TITLE:
Adaptation in Stochastic Dynamic Systems—Survey and New Results I
AUTHORS:
Innokentiy V. Semushin
KEYWORDS:
Linear Stochastic Systems, Parameter Estimation, Model Identification, Change Point Detection, Learning and Adaptive Control
JOURNAL NAME:
International Journal of Communications, Network and System Sciences,
Vol.4 No.1,
January
28,
2011
ABSTRACT: This paper surveys the field of adaptation mechanism design for uncertainty parameter estimation as it has developed over the last four decades. The adaptation mechanism under consideration generally serves two tightly coupled functions: model identification and change point detection. After a brief introduction, the pa-per discusses the generalized principles of adaptation based both on the engineering and statistical literature. The stochastic multiinput multioutput (MIMO) system under consideration is mathematically described and the problem statement is given, followed by a definition of the active adaptation principle. The distinctive property of the principle as compared with the Minimum Prediction Error approach is outlined, and a plan for a more detailed exposition to be provided in forthcoming papers is given.