基于子系统重构的鲁棒滤波算法及其应用

复杂环境下航天器姿态确定系统存在系统参数不确定性,并且可能存在观测数据的间断丢失情况.对此,本文研究了一种新的鲁棒滤波算法.该算法采用线性预测子系统重构补偿丢失的观测信号,并利用其参与到测量更新阶段,采用极小极大理论推导出具有观测数据丢失的不确定性系统的鲁棒状态估计的递推形式.将该方法用于复杂环境下的航天器姿态确定系统中,通过仿真验证了算法的有效性.

Robust filtering algorithm based on subsystem reconstruction and its application

Spacecraft attitude determination systems in complex environment system have the problem of parametric uncertainties, and observation data discontinuity. A new robust filtering algorithm is adopted to solve it. In this algorithm, linear prediction system is reconstructed to compensate for the loss of observing signals, and participated in the measurement update phase. Minimax theory is used to deduce recursion form of robust state estimation for uncertain system with observation data loss. This method is used for spacecraft attitude determination system in complex environments, and simulations verify the validity of the algorithm.