Paolo Robuffo Giordano

Sampling-Based Coordination-Informed Multi-Objective Multi-Robot Reinforcement Learning

Antonio Marino, Esteban Restrepo, Soon-jo Chung, Paolo Robuffo Giordano, Claudio Pacchierotti

Multi-robot systems must simultaneously optimize competing objectives while maintaining coordinated behavior. Existing multi-agent reinforcement learning approaches often rely on fixed or centralized coordination, which limits adaptability and violates distributed constraints. Th…Read more