组合导航中软故障难以检测，致使卡尔曼滤波精度降低甚至发散．为提高滤波的容错性，提出了一种基于遗传模糊控制的智能自适应滤波算法．首先针对软故障提出一种模糊自适应滤波算法，算法中通过监测观测量新息及其变化率，应用模糊控制系统计算观测质量因子，并对滤波器量测噪声阵进行在线自适应调整，从而抑制软故障对滤波的影响，保证滤波的精度，提升容错性能．然后，利用自适应遗传算法对隶属度函数的参数进行优化，从而进一步提高算法的整体精度．利用本文提出的算法在SINS／CNS／GPS导航平台上进行了定位实验，结果显示该算法有效，在软故障存在时，定位精度小于2131，测速精度小于0．1m／s．%Since the soft fault was difficult to be detected, Kalman filter used in the integrated navigation system tended to appear a drastic declining of accuracy, or even a divergence. Therefore, an intelligent adaptive filtering algorithm based on fuzzy control optimized by genetic algorithm was proposed in order to improve the fault-tolerant ability of filter. Firstly, a fuzzy adaptive filtering algorithm was presented to deal with the soft fault. By monitoring the residual and its rate, the observation quality gene was obtained by the fuzzy control system and the filtering measurement noise matrix was adaptively adjusted on-line. The fihering effects arising from the gradual changing fault could be prevented to a large extent, and then a nice filtering accuracy and improved fault-tolerant ability were achieved. The adaptive genetic algorithm was applied to optimize the membership functions so as to enhance the entire accuracy of algorithm. At last, the proposed algorithm was used to develop a positioning experiment based on a SINS/CNS/GPS integrated navigation platform. The results show that the proposed algorithm is valid, and the positioning accuracy is less than 2 m and the velocity accuracy is less than 0.1 m/s while the soft fault exists.
展开▼
