Random Set Tracker Experiment on a Road Constrained Network with Resource Management

This paper describes the application of finite set statistics (FISST) to real-time multiple target road constrained tracking problems. We studied specific test problems where multiple modality wireless sensor networks monitored road networks of interest. Acoustic and radar detections updated a global density that tracked the number and positions of targets. The global density determines "information states" that form the basis of a closed-loop Markov Decision Process resource management procedure that controls sensor operation.