针对基于单一性能指标进行组合导航滤波算法品质评估不全面的缺陷, 提出了多性能指标融合的滤波算法品质评估方法。首先, 定义了描述滤波算法准确性、实时性、稳定性和可靠性的性能指标; 然后,建立了基于投影寻踪的组合导航滤波算法评估模型, 采用遗传算法求解出最优投影方向, 给出组合导航滤波算法总的评价指数; 最后, 对常规Kalman滤波和自适应滤波算法进行了仿真实验, 结果表明, 该方法可用于滤波算法的品质评估。

A method for evaluating the quality of the filter algorithm for integrated navigation based on the projecting pursuit theory is proposed, which fuses multiple capability indexes, so as to overcome the disadvantage of only using a single capability index in evaluating the quality of the filter algorithm.Firstly, the capability indexes such as precision, real-time performance, stability and reliability of the filter algorithm are defined.Then, the quality evaluating model of the filter algorithm based on projecting pursuit is established, the optimal projection direction is solved by using the genetic algorithm, and then the comprehensive evaluating index is given.Finally, simulation tests are carried out, in which the adaptive filter is compared with the general Kalman filter, and the results indicate that the proposed method is valid in evaluating the quality of the filter algorithm.