Multi-agent rendezvous in fluid flows via reinforcement learning
For researchers in multi-agent systems and fluid dynamics, this work demonstrates that MARL can exploit fluid kinematics to enhance coordination, addressing a key challenge in real-world aquatic or aerial swarm applications.
This study uses multi-agent reinforcement learning (MARL) to develop physics-informed rendezvous strategies in vortical flows, achieving significantly improved rendezvous rates compared to naive strategies. The learned strategies are transferable across varying vortex intensities, scales, and swarm sizes, and a heuristic extracted from them also outperforms the naive approach.
Rendezvous is a critical task for multi-agent systems, requiring agents to coordinate to meet at an unspecified location. However, achieving this in fluid environments presents a challenge, as it remains unclear how agents can exploit underlying fluid kinematics to facilitate convergence. In this study, we adopt a multi-agent reinforcement learning (MARL) approach to develop physics-informed rendezvous strategies in vortical flows. Compared to a naive strategy, where agents navigate toward their counterparts, MARL strategies significantly improve the rendezvous rate. MARL strategies also show transferability across varying vortex intensities, vortex scales, and swarm sizes. By breaking the symmetry of the state-action map, MARL strategy leverages a non-intuitive mechanism that prevents agents from becoming trapped in separate vortices, thereby enhancing rendezvous success. Additionally, a heuristic strategy is extracted from the learned strategy and also outperforms the naive strategy. Furthermore, a theoretical analysis demonstrates that fluid deformation impedes the rendezvous process. Large finite-time Lyapunov exponents identify where fluid effects separate adjacent agents, suggesting that targets should be planned in weak-deformation regions. Our findings reveal the important role that agent-fluid interactions play in multi-agent tasks and highlight the MARL capability to explore swarm intelligence in complex flow environments.