An Approximate Implementation of Multi-sensor Generalized Labeled Multi-Bernoulli Filter for Multiple Target Tracking

摘要

Multi-sensor generalized labeled multi-Bernoulli (MS-GLMB) filter provides an analytic solution for multi-sensor multiple target tracking problems, which outputs the number of targets and the trajectories of their states by using multiple sensor observations. However, the complicated data association makes MS-GLMB filter intractable. In this paper, we propose an approximate implementation MS-GLMB filter to effectively track multiple targets in multi-sensor systems. The proposed MS-GLMB filter is compared with the sequential version of the MS-GLMB filter, and simulation results show that our method has improved tracking performance.