Multi-Static Radar Target Tracking Using Information Consensus Filters
David W. Casbeer, Randal W. Beard · 2009
In this paper, the information consensus filter (ICF) is adapted to non-linear process and measurement models by employing an extended information filter (EIF). The nonlinear ICF is a decentralized estimator, where each node in the sensor network maintains a local estimate. A node’s local observations are fused with the local estimate, and the new information is communicated throughout the network by a consensus filter. We apply the non-linear ICF to a multi-static radar tracking system, where unmanned aerial vehicles (UAVs) act as mobile adaptive radar receivers. Each UAV maintains a local track of the target using the non-linear ICF, and for a small UAV team, it is shown that the accuracy of the decentralized non-linear ICF is similar to that of the estimate in an centralized extended information filter.