Optimal multiobjective placement of distributed sensors against moving targets

We consider the optimal deployment of a sparse network of sensors against moving targets, under multiple conflicting objectives of search. The sensor networks of interest consist of sensors which perform independent binary detection on a target, and report detections to a central control authority. A multiobjective optimization framework is developed to find optimal trade-offs as a function of sensor deployment, between the conflicting objectives of maximizing the Probability of Successful Search (PSS) and minimizing the Probability of False Search (PFS), in a bounded search region of interest. The search objectives are functions of unknown sensor locations (represented parametrically by a probability density function), given sensor performance parameters, statistical priors on target behavior, and distributed detection criteria. Numerical examples illustrating the utility of this approach for varying target behaviors are given.