Distributed optimal consensus filter for target tracking in heterogeneous sensor networks

In this paper, we address the problem of distributed consensus filter design for target tracking problems using heterogeneous sensor networks with two types of sensors. The type-I sensors have more computation power, while the type-II sensors are low-end ones. The main objective of this paper is to design distributed optimal consensus filters for these two types of sensors, respectively, to estimate the state of the target based on the noisy measurements. Our derivation of the optimal filter is based on the use of minimum principle of Pontryagin (for type-I sensors) coupled with the Lagrange multiplier method and the results of generalized inverse of matrices (for type-II sensors). Simulation studies are presented to validate the performance of the proposed filters.