Monte Carlo realisation of a distributed multi-object fusion algorithm

We consider the problem of distributed target tracking in a multi-object, multi-sensor scenario in which the structure of the joint distribution of the estimate between different nodes is unknown. In this paper we present a preliminary implementation of Generalised Covariance Intersection (GCI) fusion rule for multi-object posteriors through a Monte Carlo realisation. We discuss the subtleties in the case of multi-object distributions and derive a scheme for sampling from Exponential Mixture Densities which are at the heart of the GCI. We demonstrate the improvement in localisation of multiple targets in a simulation scenario.