Kalman Filter-based Multisensor Data Fusion

In this paper, we evaluate the performance of twomultisensor data fusion techniques for state estimation byapplying Kalman filter. These methods are state vector fusionand measurement fusion. The comparisons of these methods aredemonstrated for a target tracking problem and analysis isperformed by means of the components of the error covariancematrix. According to environmental conditions, we should selectone of the fusion architectures in order to fuse data obtainedfrom respective sensors. The simulation results show that themeasurement fusion methods generally have better stateestimation performance over the state vector fusion methods

Keywords:

State Estimation , Discrete Kalman Filter , DataFusion Architectures , Multisensor Data Fusion