Sensor scheduling for multiple target tracking and detection using passive measurements

An algorithm for scheduling and control of passive sensors is proposed. This algorithm is based on a partially observed Markov decision process and an expected short- or long-term reward given by the sum of Renyi information divergences between Gaussian densities. This allows effective and efficient implementations and is demonstrated on simulations of situation scenarios of practical interest. The different situations scenarios are composed of multiple unmanned aerial vehicles (UAVs), equipped with passive radar sensors, and multiple unknown targets using active radars. The rewards, in terms of measurement information gains, are then maximized to achieve the objectives of scheduling different sensors and trajectory controls.