Multi-sensor information fusion multi-stage algorithm under the unknown noisy environment

In the self-tuning kalman filtering process, In order to get the unbiased filtering results, the estimations of the unknown noises statistics information in the multi-sensor system should be unbiased. Based on the autoregressive and moving average model, a multi-stage information fusion identification algorithm is presented in this paper. This algorithm can be used to get the unbiased estimations of the unknown parameters and noises variance. The estimations could be taken into the Kalman filter to get a self-tuning filter that has good convergence to the optimal Kalman filter. An example shows the effectiveness of the algorithm.