Efficient Model Selection in Switching Linear Dynamic Systems by Graph Clustering

The computation required for a switching Kalman Filter (SKF) increases exponentially with the number of system operation modes. In this paper, computationally tractable graph representation is proposed linear dynamic (SLDS) along solution minimum-sum optimization problem clustering to reduce mode cardinality offline, before collecting measurements. It shown that upon perfect detection, induced error caused by can be quantified exactly in terms dissimilarity measures structure. Numerical results verify based on framework effectively reduces model complexity given uncertain detection and well approximated if underlying assumptions are satisfied.