Distributed CPHD Filtering Under Robust Set-Theoretic Information Flooding Mechanism

This article studies the distributed multitarget tracking problem. A cardinality probability hypothesis density filter under a robust set-theoretic information flooding mechanism is proposed to perform targeted communication to reduce the communication burden, and increase the robustness. In particular, the collecting set is locally designed to represent known local estimation of each neighbor node, and when receiving the data packet, each node selects the unknown local estimations for storage and updates its local collecting set. Then, in each iteration, the local node selectively sends its neighboring nodes the local estimations that they do not know. When the node knows the local estimations of all the nodes, the information fusion will be performed and the received information will no longer be processed. Finally, the tracking performance of the proposed algorithm is verified by comparison studies.