Safe Multi-Agent Isaac Gym benchmark for safe multi-agent reinforcement learning research. - View it on GitHub