Active sensing based cooperative target tracking using UAVs in an urban area

Abstract

To the problem of cooperative ground moving target tracking using UAVs in an urban area, a novel method is presented, which based on target state fusion estimation and prediction, as well as on-line trajectory planning. Firstly, based on active sensing, a solving framework was presented, the formal representation and system model of cooperative target tracking using a team UAVs was established; secondly, a filter, named unscented information filter, was suggested for fusion estimation and prediction of the target state; thirdly, with a predicted target state, an on-line trajectory planning algorithm was presented, which based on receding horizon control and genetic algorithm. Simulation results demonstrate that the performance of target tracking was improved with regard to the predicted target state.