Consensus-Based Kalman filter for multi-target tracking with integrated measurements^∗

A distributed Kalman filter for integrated measurement of interactive multi-target tracking is proposed. The consistency Kalman filter is a distributed information fusion algorithm. Each target is estimated by the absolute measurement information and the relative measurement information of neighbor target. The absolute measurement information is the measurement information of multiple sensors. The estimated value is obtained by the consistency fusion. That is, in wireless sensor network, the sensor node and neighbor node carry out the data exchange. All nodes get consistent and high-precision estimates. The design of the gain matrix is obtained by minimizing the upper bound of the estimation error covariance matrix. The effectiveness of the algorithm is verified by numerical simulation.