Coordination of sensor platforms for tracking and identifying objects: Performance evaluations

This paper presents an evaluation of algorithms to coordinate multiple sensor platforms to track and identify multiple objects, where the system demonstrations and experiments have been performed in a simulation environment. The paper describes how object tracking and decision making methods are applied to collaborative control in a maritime channel surveillance and protection vignette. It delineates important system components and the performance evaluation metrics. The experimental results on the algorithms developed for information fusion (unscented kalman filter, bayesian identity management) and decision making (heuristic genetic algorithm) are presented. The results show that the collaborative control approach provides effective and non-overlapping sequential allocations of the scarce sensor platforms to multiple objects from different viewing angle quadrants, and increases the track management performance.