Application of the block Kalman filter to multisensor estimation with uncertain measurements

The multisensor estimation problem in which measurements from N sensors are acquired sequentially is considered. The Block Kalman Filter is shown to be a natural way of fusing sequential data in a variety of situations involving multisensor estimation. This technique is then extended to the case of multisensor tracking applications where false measurements are often present, by applying the probabilistic data association filter (PDAF) technique to the blocked state model.