Towards Designing Multi-Agent Coverage Systems Capable of Anticipation and Tight Coordination with Detailed Environmental and Perception Models

Loïs Vanhée (1) , Laurent Jeanpierre (1) , Abdel-Illah Mouaddib (1)

1 Equipe MAD - Laboratoire GREYC - UMR6072

This paper introduces a new approach for modelling coverage problems , capable of taking into consideration 1) evolving points of interest, 2) environmental dynamics, including the influence of agent actions, and 3) detailed perception models. Such coverage problems requires tight coordination between agents while anticipating the consequences of their actions for maximizing the covered areas over time while avoiding adverse situations (e.g. collisions).

Domaines: Système multi-agents [cs.MA]

Loïs Vanhée, Laurent Jeanpierre, Abdel-Illah Mouaddib. Towards Designing Multi-Agent Coverage Systems Capable of Anticipation and Tight Coordination with Detailed Environmental and Perception Models. International Conference on Autonomous Agents and Multiagent Systems, Aug 2018, Stockholm, France. ⟨hal-01893896⟩