A Relative Navigation Method For Spatial Non-cooperative Maneuvering Targets Based On Extended State Observer

Aiming at the problem of high-precision relative navigation for space non-cooperative targets with diverse maneuvering capabilities, a highly robust relative navigation method based on the idea of an extended state observer(ESO for short) is proposed in this paper. Considering the characteristics of non-cooperative targets, such as the lack of effective cooperative measurement information, strong uncertainty of dynamic models, and complex space disturbances, the relative kinematic equation is established by uniformly treating the target maneuver and external disturbances as additional states. An extended state observer is designed to achieve synchronous online estimation of relative motion parameters and lumped disturbances. Moreover, a feedforward compensation mechanism is adopted to effectively suppress the influence of target maneuver and other disturbances on the relative motion estimation. Simulation experiments show that, compared with traditional extended Kalman filter and augmented Kalman filter method, the relative navigation accuracy of this method is improved by more than $50 \%$ under target maneuvering conditions, verifying the effectiveness of the proposed method in complex space environments. The research findings can provide reliable technical support for space situation awareness, on-orbit servicing of spacecraft,space debris removal tasks, and related applications.