一种新型模糊卡尔曼滤波数据融合算法

A New Fuzzy Kalman Filter Data Fusion Algorithm

复杂条件下,某些特殊数据的融合问题将不再满足卡尔曼滤波算法的误差特性假设。模糊卡尔曼滤波通过引入模糊控制规则自适应调整噪声协方差矩阵,使模型中的噪声能够更加靠近真实噪声水平。通过对误差特性的分析,提出依据不同的误差来源采用不同推理规则和协方差矩阵调整策略的模糊卡尔曼滤波方法。仿真实验验证了方法的合理性和有效性。

Under complex conditions, the fusion of some special data will no longer satisfy the error characteristics hypothesis of the Kalman filter algorithm. Fuzzy Kalman filter adaptively adjusts the noise covariance matrix by introducing fuzzy control rules, so that the noise in the model can get closer to the real noise level. Based on the analysis of the error characteristics, this paper proposes fuzzy Kalman filtering method based on different error sources, using different inference rules andcovariance matrix adjustment strategy. The simulation experiment proves the rationality and effectiveness of the method.