Analytic continuation extended Kalman filter framework for perturbed orbit estimation using a network of space-based observers with angles-only measurements

This work presents a new method for space-based angles-only orbit estimation. The approach relies on the integration of a novel and highly accurate Analytic Continuation technique with a new measurement model for multiple observers for inertial orbit estimation. Analytic Continuation computes the perturbed orbit dynamics, as well as the perturbed state transition matrix (STM), in the inertial frame. A new measurement model is developed for simultaneous measurements using a constellation of low-cost observers with monocular cameras for angles-only measurements. Analytic Continuation and the new measurement model are integrated in an Extended Kalman Filter (EKF) framework, where the Analytic Continuation method is used to propagate the perturbed dynamics and compute the perturbed STM and error covariance, with the measurements obtained via the new measurement model. Two case studies comprising small and large constellations of observers are presented, along with cases of sparse measurements and a study of the computational efficiency of the proposed approach. The results show that the new approach is capable of producing highly accurate and computationally efficient perturbed orbit estimation results compared with classical EKF implementations.