Awareness-based decision making for search and tracking

This paper focuses on the management of mobile sensor agents for the search and tracking of multiple objects of interest. In the case where such objects are in greater numbers than available agents, search and tracking are two competing demands since a sensor agent can perform either the tracking mission or the search mission, but not both at the same time. A sensor agent has to decide on whether to continue searching or stop and track once it finds an object of interest. Based on a novel dynamic awareness model and assuming static objects, a decision-making and control strategy is developed to guarantee the full coverage of a domain of interest, and, equivalently, the detection of all objects of interest in the domain with probability one. The strategy also guarantees the tracking of each object's "state" for a minimum guaranteed amount of time tau
_c
. Centralized and decentralized implementations are described. Numerical simulations are provided to demonstrate the performance of the strategies.