2019 IEEE International Conference on Artificial Intelligence and Computer Applications (ICAICA) | 2019
Globally Optimal Weighted Fusion Kalman Filter with Colored Measurement Noises
Abstract
In this paper, the multisensor systems with ARMA coloured measurement noise are converted to those with the same local dynamic model and uncorrelated noises by using the state augmented method. Furthermore, the globally optimal weighted measurement fusion Kalman filter is given. Compared with the existing methods transforming the multisensor systems into those with the same or different local dynamic models and correlated noises, the transformed system models in this paper are applicable to the weighted measurement fusion algorithm. Although the dimension of augmented state is higher than that in the existing references, the computational load and complexity of the whole optimal fusion Kalman filter are better than the existing fusers. The simulation example for a 3-sensors radar tracking system with colored measurement noises shows its effectiveness.