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 ,有较少的保守性 A new method called evolutionary programming (EP) Kalman filter is proposed for the optimal robust filtering of uncertain systems. This method uses EP global optimization technique to search for the optimal estimation interval, and also adopts EP global The best technique to determine the optimal estimated nominal value that can be achieved in a project.The assumption of this new method is exactly the same as the standard Kalman filter and has the same recursive structure as the standard Kalman filter with the same optimal Finally, an example of computational simulation is given and compared with the simulation results in [1]. The results show that the proposed method is more accurate and less conservative