About this document
Multi-Agent Learning for CAV Merging Control by jonathan.tam5565 is a document available to read on EtoBox.
The document discusses a study on cooperative merging control for connected and automated vehicles (CAVs) using multi-agent deep reinforcement learning (MADRL). It addresses the challenge of aligning local benefits of CAV users with system optimum strategies, proposing a reward switching mechanism and the incorporation of incentives to enhance cooperation. The study demonstrates that this approach can improve traffic flow efficiency, reduce energy consumption, and enhance safety during highway on-ramp opera
- Author
- jonathan.tam5565
- Language
- EN