abstract
- In this chapter, we consider the problem of managing a network of sensors with particular application to multisensor multitarget tracking. We study the problem of decision based control of a network of sensors carrying out surveillance over a region that includes a number of moving targets. The objective is to maximize the information obtained and to track as many targets as possible with the maximum possible accuracy. Uncertainty in the information obtained by each sensor regarding the location of the targets is addressed in the problem formulation. The chapter presents a number of solutions for centralized and decentralized tracking involving sensor management and distributed information flow control. We consider a distributed data fusion system consisting of sensors that are decentralized, heterogenous, and potentially unreliable. The objective function for sensor management is based on the Posterior Cramer-Rao lower bound and constitutes the basis of a reward structure for Markov decision processes that are used, together with decentralized lookup substrate, to control the data fusion process. In distributed sensor network fusion, we analyze three distributed data fusion algorithms: associated measurement fusion, tracklet fusion and track-to-track fusion. The chapter also provides a detailed analysis of communication and computational load in distributed tracking algorithms. In centralized sensor network fusion, we introduce a multi-level hierarchy of MDPs to control each of the sensors in the network. Simulation results are presented on a representative multitarget tracking problem using a network of sensors showing a significant improvement in performance compared to the existing algorithm.