Information-Based Control of Decentralised Sensor Networks
David Nicholson, Sarvapali D. Ramchurn, Alex Rogers · Birkhäuser Basel eBooks · 2008
This chapter describes how formal information measures can be used as the basis for enabling decentralised, intelligent and autonomous control of large-scale sensor network resources, with widespread application throughout the military and security domain. These information measures are the result of filtering and fusing local sensor observations, assimilating the products over a communication network, and interpreting them in the wider context to infer underlying states of interest to the military or security operation. Information provides a currency against which a constrained set of sensing and communication actions can be valued, resulting in a single action or sequence of actions being executed. This is known as Information-Based Control (IBC). The main focus of this chapter is the problem of decentralised IBC in a large-scale sensor network, and its solution in terms of multi-agent system methodologies. Examples and applications, relevant to the military world, are used to highlight a number of important practical considerations. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.