Adaptive Consensus-based Multi-domain Collaborative Asynchronous Multi-target Tracking

For the asynchronous measurements and communication delays in the integrated space-air-ground multi-domain collaborative multi-platform fusion systems, a multi-source information fusion method for multi-target tracking (MTT) based on adaptive consensus is proposed. Firstly, measurement models are established for ground-based radars, airborne early warning platforms and space-based satellites, with local MTT implemented through labeled multi-Bernoulli (LMB) filters. Secondly, a consensus posterior time is defined to enable the forward prediction across detection platforms, achieving temporal alignment between local LMB multi-target posteriors. Subsequently, a distributed fusion is performed through the multi-domain collaborative network, where an asynchronous metric function is derived to dynamically determine consensus fusion weights according to asynchronous characteristics of each platform. Finally, the results of comprehensive simulation experiments demonstrate that the proposed approach successfully accomplishes MTT under conditions of asynchronous measurements, exhibiting superior adaptability to target maneuvers. Compared to conventional consensus fusion methods, this approach demonstrates superior tracking accuracy and enhanced stability.