An Event-Triggered Hybrid Consensus Filter for Distributed Extended Object Tracking

Abstract

Motivated by the unique state characteristics of the extended object and energy constraints in distributed sensor networks, this letter proposes a novel event-triggered hybrid consensus filter for distributed extended object tracking, achieving balanced estimation-communication performance. This parallel consensus mechanism processes three consensus operations on the prior information pair, novel information pair of kinematic state, and shape parameter information pair of extent state, enabling enhanced consensus and propagation of extended object characteristics across the network. To reduce data transmission while preserving estimation performance, the proposed event-triggered strategy contains three distinct transmission tests, performed in parallel on corresponding information pairs to evaluate information loss. Simulation results of demonstrate the superior performance of the proposed filter compared with conventional triggered filters.