Overview on evidential fusion approaches in the context of collaborative perception for occupancy modeling

The objective of this paper is to provide a review of the literature on collaborative perception, with a special focus on evidential fusion and combination approaches for both reliable and unreliable information sources. In addition, a study on uncertainty estimation tools is considered in order to integrate them in the global fusion architecture. To do so, we focus on the powerful tools of data fusion and conflict management offered within the Dempster-Shafer framework. Hence, we start by modeling occupancy on scenarios from the Carla simulator using belief functions by varying combination operators proposed in the literature. We add some experimentation using probabilistic approach using the same scenario to extract more conclusions.