Skip to Main Content
For the sake of higher level of autonomy, based on dynamic Bayesian networks (DBNs), a novel awareness frame is proposed for achieving cooperative control of multiple unmanned aerial vehicles (UAVs) in dynamic environment. With learning and inference algorithms for DBNs, the awareness in dynamic environment could be accomplished using parameterspsila change in relative DBNpsilas transition networks based on derived fusion signal sequences. A cooperative path optimization algorithm against pop-up threats is provided for multiple UAVs under this awareness frame, which exploits knowledge about the given problem, together with a simple but efficient form of coordination variable among independent UAVs. Learning and inference for this awareness DBN are also discussed. Simulation results demonstrating the feasibility of this approach are presented.