Metric selection for information theoretic sensor management

Abstract

Different information theoretic sensor management approaches are compared in a Bayesian target-tracking problem. Specifically, the performance using the expected Renyi divergence with different parameter values is compared theoretically and experimentally. Included is the special case in which the expected Renyi divergence is equal to the expected Kullback-Leibler divergence, which is also equivalent to both the mutual information and the expected change in differential entropy for this Bayesian updating problem. The example problem involves a single target moving in a circle, four bearing-only sensors, and two time-delay sensors. A particle filter based tracker is used.