Decentralized Autonomous Traffic Management through Corridor Networks
2026-06-22 • Multiagent Systems
Multiagent SystemsArtificial IntelligenceEmerging TechnologiesRobotics
AI summaryⓘ
The authors address the problem of managing many autonomous aircraft flying in special air corridors without a central controller. They use a method called multi-agent reinforcement learning to teach each aircraft how to navigate independently while still following traffic rules. Their approach, trained in simple corridor setups, works well even when applied to more complex corridor networks with merges and splits, without needing retraining. The results show that aircraft can safely maintain separation, follow routes, and keep good speeds by coordinating only locally. This suggests decentralized methods can effectively manage busy air traffic corridors.
Autonomous aircraftAdvanced Air Mobility (AAM)Decentralized traffic managementMulti-agent reinforcement learning (MARL)Air corridorsTrajectory planningTraffic flowInter-aircraft separationZero-shot transferTraffic density
Authors
Jasmine Jerry Aloor, Aadarsh Govada, Hamsa Balakrishnan
Abstract
As autonomous aircraft are introduced at scale and traffic density increases, centralized management becomes insufficient to coordinate the large numbers of crewed and uncrewed aircraft. Dedicated Advanced Air Mobility (AAM) corridors have therefore been proposed for organizing high-density autonomous traffic flows. The desire to scalably provide autonomous aircraft flexibility in trajectory planning motivates the development of decentralized approaches to traffic management in AAM corridors. In this work, we extend a multi-agent reinforcement learning (MARL) approach to address the challenge of decentralized traffic flow management in air corridor networks. We test policies trained in a single-corridor setting on increasingly complex multi-corridor networks with combinations of merges and splits in a zero-shot manner. Experimental results demonstrate that learned behaviors transfer well to scenarios with varying traffic density, network geometry, and heterogeneous vehicle performance, without needing centralized coordination or model retraining. We evaluate system-level performance in terms of conformance to corridor boundaries, completion rates, average speeds, distance traveled, and maintenance of inter-aircraft separation. We find that although our policies require only locally coordinated entry, traversal, and exit behaviors, they collectively produce desirable traffic flows through the corridor network.