中文摘要：

针对仅含角度测量信息的单个天基平台可观测性较弱的问题,提出了一种含脉冲机动检测的空间非合作目标跟踪算法,并设计了非合作目标实时跟踪数据处理流程.该算法利用抗差估计技术和UKF（Unscented Kalman Filter,无迹卡尔曼滤波）算法构造目标跟踪滤波器,并综合残差多项式拟合和新息分布特征等方法实现目标机动检测,在天基平台观测信息类型有限和观测几何较差的情况下,可以同时排除孤立野值和成片测量野值的影响,实现非合作机动目标的连续稳定跟踪.数值实验验证了算法的可行性和有效性,也表明了跟踪精度和可靠性与测量精度密切相关.

英文摘要：

A new tracking algorithm with pulse maneuver detection for non-cooperative space objects is proposed to make up for the poor observability of a single space-based platform, for which only angular measurement information is available. A real-time tracking data processing flow is designed for non-cooperative space objects. The algorithm introduces a robust estimation technique to unscented filter for real-time tracking purpose. Then, a maneuver detection method is developed by integrated polynomial fitting of measurement residuals and statistical character analysis of filter innovation. In case of poor observation conditions caused by deficient information or poor observation geometry for a single space-based platform, the algorithm can suppress the influence of isolated outliers as well as patched outliers, and obtain a continuous and roust tracking of non-cooperative objects. Simulation results validate the feasi- bility and efficiency of the proposed approach, and show that the tracking accuracy and robustness is highly correla- ted with measurement accuracy.