Safe Multi-Agent Reinforcement Learning for Formation Control without Individual Reference Targets

Open in new window