Cooperative Bayesian target detection on a real road network using aerial vehicles
Brett Barkley, Derek A. Paley · 2016
A cooperative track-before-detect algorithm for multiple ground targets is presented for fixed-wing Unmanned Air Vehicles (UAVs) with a finite field of view. The road network forms a graph whose nodes indicate the target likelihood ratio. Target observations are assimilated by a Bayesian likelihood ratio tracker in which likelihood diffuses according to the graph Laplacian of the road network. Using the likelihood ratio in a composite potential along with attractive and repulsive terms, the algorithm directs UAVs modeled as Dubins cars towards nodes with high likelihood. Results from numerical simulations are included to illustrate the algorithm.