Orbit maneuver detection is critical for tracking space objects. In previous studies, the solutions generally suffer from the problems of weak sensitivity, time delay, and difficulty in selecting detection thresholds, which are more serious for high-altitude near-circular orbit missions when only line-of-sight (LOS) angle measurements are available. To this end, this study furthers the development of an orbital maneuver detection method for noncooperative targets in high-altitude orbits, where a highly sensitive maneuver characterization quantity with an adaptive detection threshold is introduced. First, nonlinear relative dynamics and a line-of-sight measurement model are established in the local vertical horizontal frame, and a square-root cubature Kalman filter with strong tracking capability is used to estimate the relative orbital parameters. Second, the concept of relative angular momentum is introduced, and a data sequence for maneuver detection is constructed by performing sliding window variance calculation and moving average smoothing on the relative angular momentum, thereby obtaining a highly sensitive detectable quantity. Subsequently, least-squares polynomial fitting is used for prediction and to construct a comparative fitting quantity. Furthermore, under the constraint of false dismissal probability, the Neyman–Pearson (N–P) criterion is introduced to solve the maneuver-detection threshold of the predicted fitting quantity sequence. Additionally, the concept of a delay window was introduced to enhance the sensitivity of the detection method for locating the maneuver time point. Finally, a Monte Carlo simulation system is used to verify the effectiveness and evaluate the performance of the proposed algorithm. The simulation results show that the proposed algorithm achieves a higher detection success rate, lower detection error, and shorter delay window length while exhibiting greater potential for detecting small-thrust maneuvers of relative long-range targets in high-altitude orbits.

Relative navigation is a key enabling technology for space missions such as on-orbit servicing and space situational awareness. Given that there are several special advantages of space relative navigation using angles-only measurements from passive optical sensors, angles-only relative navigation is considered as one of the best potential approaches in the field of space relative navigation. However, angles-only relative navigation is well-known for its range observability problem. To overcome this observability problem, many studies have been conducted over the past decades. In this study, we present a comprehensive review of state-of-the-art space relative navigation based on angles-only measurements. The emphasis is on the observability problem and solutions to angles-only relative navigation, where the review of the solutions is categorized into four classes based on the intrinsic principle: complicated dynamics approach, multi-line of sight (multi-LOS) approach, sensor offset center-of-mass approach, and orbit maneuver approach. Then, the flight demonstration results of angles-only relative navigation in the two projects are briefly reviewed. Finally, conclusions of this study and recommendations for further research are presented.