Sequential Monte Carlo methods for collaborative multi-sensor tracking

Localization, tracking, and navigation (or geolocation) are essential in many commercial, security, public safety, and military applications. It is widely known that accurate and reliable geolocation is extremely difficult to achieve in complex multipath environments such as indoor and urban areas. With the recent advances in sensor networks, it has become possible to form dynamic multihop networks using many sensor nodes and thus to accomplish collaborative sensing and processing with distributed sensor nodes, which provides unprecedented opportunity to accomplish reliable localization in complex application scenarios. Collaboration of resource-constrained, unreliable sensor nodes is extremely important to achieving substantial sensing and processing capability in the aggregate and to providing collectively reliable network behavior in mission-critical applications. In this paper, we present a general framework for collaborative localization and tracking of mobile sensor nodes using sequential Monte Carlo methods. Various simulation results are presented to demonstrate the performance of the proposed approach.