Integration of shape constraints in data association filters

Many solutions have been proposed in literature to deal with the tracking and data association problem. A common assumption made in the proposed algorithms is the independence of targets. There are however many interesting applications in which targets exhibit some sort of coordination, they satisfy shape constraints. In the current work a general and well formalized method which allows to learn and embed such constraints into probabilistic data association filters is proposed. The resulting algorithm performs robustly in challenging scenarios compared to a commercial tracking system based on a standard filter (JPADF).