Model-set adaptation in variable-structure MM estimation by hypothesis testing
X.R. Li · 1998
The key component in variable-structure multiple-model (MM) estimation is model-set adaptation (MSA). This paper formulates MSA as hypothesis testing problems and provides effective solutions, which have some desirable optimality properties. The hypotheses tested are in general composite, N-ary, multivariate, and worst of all, not necessarily disjoint. The results form a theoretical foundation and guideline for developing good and practical MSA algorithms.