Safety-Aware Multi-Agent Learning for Dynamic Network Bridging
This addresses safety in cooperative multi-agent tasks for robotics or communication networks, but it appears incremental as it combines existing methods.
The paper tackles the problem of maintaining a communication path between moving targets in safety-critical multi-agent systems under partial observability, and shows that integrating a safety filter with reinforcement learning improves coordination, though no concrete numbers are provided.
Addressing complex cooperative tasks in safety-critical environments poses significant challenges for multi-agent systems, especially under conditions of partial observability. We focus on a dynamic network bridging task, where agents must learn to maintain a communication path between two moving targets. To ensure safety during training and deployment, we integrate a control-theoretic safety filter that enforces collision avoidance through local setpoint updates. We develop and evaluate multi-agent reinforcement learning safety-informed message passing, showing that encoding safety filter activations as edge-level features improves coordination. The results suggest that local safety enforcement and decentralized learning can be effectively combined in distributed multi-agent tasks.