基于进化规划的区间系统卡尔曼滤波

Kalman Filtering for Interval System Using Evolutionary Programming

对不确定系统的最优鲁棒滤波问题 ,提出了一种新的方法———进化规划 (EP)卡尔曼滤波 .该方法采用EP全局寻优技术 ,搜索最优估计区间 ;也采用EP全局最优技术 ,确定工程中可实现的最优估计标称值 .这种新方法的假设条件与标准的卡尔曼滤波完全相同 ,并具有标准的卡尔曼滤波相同的递推结构、相同的最优性 .最后 ,给出计算仿真的例子 ,并与文 [1]的仿真结果进行了比较 .结果表明 ,本文提出的新方法更精确 。

This paper develops a robust Kalman filtering algorithm by incorporating with the evolutionary programming (EP) technique for interval systems containing uncertainties. Based on the global optima_searching capability of EP, the new filtering algorithm is able to find the optimal Kalman filtering results at every iteration. The upper and lower boundaries and the nominal trajectory of the optimal estimates of the system state vectors are computed by the new algorithm, under the same statistical conditions while yielding the same optimal estimates as the conventional Kalman filtering scheme. A typical computer simulation example is included for comparison with the interval Kalman filtering method, which shows that the new algorithm is more accurate and less conservative.