Ground Target Motion Estimation based on visual measurements from a UAV platform
Luis Miguel del Pozo Lopez, Miguel Sabaris Boullosa, Juan Jose Navarro Corcuera · 2021
This paper presents an on-board target motion estimation solution to the problem of standoff tracking of a moving ground target using unmanned aerial vehicles (UAV). As target motion estimation error is sensitive to choice of the UAV flight path, the estimation problem is solved in conjunction with a UAV guidance law. A tactical fixed-wing UAVequipped with a gimbaled camera is considered. The onboard estimator is designed such that the UAV can track a moving ground target and provide the target's motion estimation solution in real time. An estimator solution based on Kalman Filtering is proposed with a guidance law based on inverse kinematics. Additional results based on other estimators proposed in the literature will be shown for comparison purposes Numerical simulations are performed to verify the feasibility and benefit of the approach. The results show the filters exhibit stable convergence periods for both position and velocity estimates in the presence of tracking loss events. These filters show similar performances in relatively fairy conditions. The Kalman filter approach, integrating a dynamic model and measurements, is preferred since it may provide an early stability in the estimation of the target's state and it is capable to cover tracking loss events.