Decentralized Real-Time Estimation and Tracking for Unknown Ground Moving Target Using UAVs

This paper presents a decentralized cooperative tracking strategy based on information filtering with consensus analysis and model predictive control (MPC) for multiple unmanned aerial vehicles (UAVs), tracking unknown ground moving target. For unknown target, squared-root cubature information filtering (SCIF) is designed to estimate the target states based on the measurement from the onboard sensor at each UAV. To eliminate the difference between estimations of UAVs, the consensus algorithm, hybrid consensus on measurement-consensus on information is applied for more accurate estimation of target. A fast MPC method is introduced to obtain the UAVs' path, where collision avoidance between UAVs and the change of communication topology among UAVs are taken into account. Finally, the simulation results demonstrate the effectiveness of the proposed method.