Decision-Theoretic Sensor Resource Management

Abstract

Sensor resource management is usually formulated as an optimization problem under uncertainty. We use a decision-theoretic model to show that the objective function should not be just tracking or target identification performance. Instead, it should represent the expected value of the outcome from using the collected data. This outcome depends on how the collected data and fusion results are used in making other resource management decisions such as weapon to target assignment. We argue the need for considering the integrated sensor/weapon management problem and propose a decomposition to make the solution feasible