Application of Action Selection, Information Gathering, and Information Evaluation Technologies to UAV Target Tracking

Abstract

AbstractThis paper illustrates agent technologies applied to unmanned aerial vehicle (UAV) target tracking. The combination of the three technologies presented in this paper provide UAVs with functionality needed for coordinated autonomous operation, from building up accurate beliefs, efficiently gathering information, to acting rationally. In the UAV target tracking domain, communication among agents is necessary for building beliefs about target locations. Reliable information provisioning networks are constructed through selection of appropriate information sources and trust evaluations are applied to belief revision. Also, a macro-based action selection scheme is deployed for efficient coordination of the target tracking activity among agents.KeywordsTarget LocationMultiagent SystemUnmanned Aerial VehicleAction SelectionTarget TrackingThese 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.