Hybrid Consensus-Based Cubature Information Filter and Covariance Intersection Strategy for Hierarchical Distributed State Estimation

In this paper, a novel hybrid consensus-based distributed cubature information filter and sequential covariance intersection estimator (CDCIF-SCI) is proposed for hierarchical distributed state estimation. A comprehensive hierarchical fusion structure is presented, utilizing the CDCIF filter at the local sensor layer and the SCI fuser at the cluster head layer. At the local sensor layer, the hybrid consensus strategy is integrated into the cubature information filter to achieve distributed state estimation. At the cluster head layer, the global fusion center employs a sequentially implemented multi-sensor covariance intersection fusion among the estimates from each cluster to enhance accuracy. Furthermore, an overall performance analysis of the CDCIF-SCI is provided, offering theoretical insights into the estimator's accuracy. Numerical simulation is conducted to validate the effectiveness of the estimator.