Fuzzy adaptive Kalman filter for multi-sensor system

The present paper proposes a new adaptive Kalman filter-based multisensor fusion to satisfy the real time performance requirements. The adaptive scheme of Kalman filter based on fuzzy logic is developed to prevent the filter from divergence and to avoid the need of accurate knowledge of statistical values of noise for both process and measurement noises. To reach this objective, first each measurement coming from each sensor is fed to a fuzzy-adaptive Kalman filter to estimate the covariance measurement noise matrix. Then it is applied to a set of Kalman filters. The fuzzy similarity is calculated between each sensor´s measurement values and the multiple sensors´ objective values to determine the importance weight of each sensor in fusion algorithm. An applied example is given to confirm that the algorithm can give priority to the highest stability and highest reliable sensors.