ABSTRACT This research addresses distributed multi‐sensor group target fusion tracking under time misalignment, specifically considering sensors with unknown initial operating time offsets and sensors with asynchronous multi‐sampling rates. In contrast to prior studies that assume known time offsets between sampling frames, this work develops a fusion approach that does not require such information. The whole methodology within a standard multi‐sensor fusion process: local trajectory Poisson multi‐Bernoulli (TPMBM) filtering results are first used to estimate time offsets, followed by the prediction of posterior densities using these offsets, and finally, a sequential fusion strategy is applied for group target tracking. To estimate time offsets, the posterior density at the retrodiction moment is derived using a continuous‐discrete model, and an optimisation function that incorporates time offsets is formulated. The time region is discretised, and iterative methods are used to search for time‐offset solutions. Simulation results confirm the effectiveness and feasibility of the proposed time‐offset estimation and distributed fusion‐tracking method.
Xiong et al. (Thu,) studied this question.