Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces a Momentum-Based Reward Function (MBRF) for DRL-based traffic signal control, designed to incentivize continuous vehicle movement rather than solely penalizing congestion. MBRF incorporates a momentum term that rewards vehicles for maintaining speed, leading to more stable and long-sighted policies. Experiments in SUMO demonstrate that MBRF achieves superior throughput-emission trade-offs and learning stability compared to delay/queue-based rewards and traditional controllers.
Stop obsessing over congestion: rewarding continuous vehicle movement in DRL traffic control yields better throughput and lower emissions.
Urban traffic congestion is a growing global issue contributing significantly to long commute times and environmental pollution. Traditional traffic signal control systems often fail to adapt to dynamic traffic conditions. Adaptive traffic signal control can improve urban traffic without changing road infrastructure. Deep Reinforcement Learning (DRL) has shown strong performance for this task, but existing delay and queue-based rewards often produce short-sighted or unstable policies. This paper proposes a Momentum-Based Reward Function (MBRF) that encourages vehicles to keep moving rather than penalizing congestion alone. The method is evaluated in SUMO (Simulation of Urban MObility) using standard traffic metrics such as waiting time, queue length, throughput, and CO2 emissions. Results show that the proposed reward produces better throughput-emission trade-offs and more stable learning behavior than delay or queue-based rewards, as well as classical controllers such as Max Pressure and LQF.