Uncertainty-Aware δ-GLMB Filtering for Multi-Target Tracking

The δ-GLMB filter is an analytic solution to the multi-target Bayes recursion using in multi-target tracking. Theoretically, the δ-GLMB filter handles uncertainties in measurements in its filtering procedure. However, in practice, degeneration of the measurement quality affects the performance of this filter. In this paper, we will discuss the effects of increasing measurement uncertainty on δ-GLMB filter and also will propose two heuristic methods to improve the performance of the filter in such conditions. The base idea of proposed methods is to utilize the information stored in the history of the filtering procedure, which can be used to decrease the measurements uncertainty effects on the filter. Since the GLMB filters have shown good results in the field of multi-target tracking, an uncertain immune δ-GLMB can play as a strong tool for this area. In this study, the results indicate that the proposed heuristic ideas can improve the performance of the filtering in presence of uncertain observations.