Disambiguating localization symmetry through a Multi-Clustered Particle Filtering

Distributed Particle filter-based algorithms have been proven effective tools to model non-linear and dynamic processes in Multi Robot Systems. In complex scenarios, where mobile agents are involved, it is crucial to disseminate reliable beliefs among agents to avoid the degradation of the global estimations.We present a cluster-based data association to boost the performance of a Distributed Particle Filter. Exploiting such data association, we propose a disambiguation method for the RoboCup scenario robust to noise and false perceptions. The results obtained using both a simulated and a real environment demonstrate the effectiveness of the proposed approach.