Goal-Oriented Multi-Agent Reinforcement Learning for Decentralized Agent Teams
This addresses coordination challenges for decentralized agent teams like autonomous vehicles in real-world settings, representing an incremental advance in MARL for specific domains.
The paper tackles the problem of coordinating autonomous vehicles in dynamic, unpredictable environments with limited communication and partial observability by proposing a decentralized Multi-Agent Reinforcement Learning framework with goal-aware communication, resulting in significantly improved task success rates and reduced time-to-goal compared to baselines, with stable performance as agent numbers increase.
Connected and autonomous vehicles across land, water, and air must often operate in dynamic, unpredictable environments with limited communication, no centralized control, and partial observability. These real-world constraints pose significant challenges for coordination, particularly when vehicles pursue individual objectives. To address this, we propose a decentralized Multi-Agent Reinforcement Learning (MARL) framework that enables vehicles, acting as agents, to communicate selectively based on local goals and observations. This goal-aware communication strategy allows agents to share only relevant information, enhancing collaboration while respecting visibility limitations. We validate our approach in complex multi-agent navigation tasks featuring obstacles and dynamic agent populations. Results show that our method significantly improves task success rates and reduces time-to-goal compared to non-cooperative baselines. Moreover, task performance remains stable as the number of agents increases, demonstrating scalability. These findings highlight the potential of decentralized, goal-driven MARL to support effective coordination in realistic multi-vehicle systems operating across diverse domains.