Passive localization based on multi-sensor GLMB filter Using a TDOA Approach
Xudong Wang, Weifeng Liu, Yimei Chen · 2017
The problem concerning the methods of passive localization based on multi-sensor has attracted widespread attentions in recent years. Traditional target localization estimate techniques assume that the number of targets in the localized areas is consistent and known at any moment. This paper considers a more realistic situation that the number of targets is unknown and time-varying under cluttered environment. In the first stage, we use the method of time difference of arrival to describe the observation model. In the second stage, we apply the random finite set theory to deal with such a more complicated localization problem from the perspective of set, by using the generalized labeled multi-Bernoulli filter with the assumption of independence of all targets. Finally, we estimate the position and the number of the time-varying targets simultaneously by using a sequential Monte Carlo implementation. To verify the proposed algorithm, the final experiment provides three pairs sensors to track two targets, which moving in a 2-dimension plane with constant velocity.