AI-Driven Traffic Signal Control System to Reduce CO2 Emissions
YAISD-WS 2025 · CEUR Workshop Proceedings 3974, pp. 18-27
Editorial summary
A reinforcement-learning controller adjusts traffic-light phases using queue lengths, vehicle speeds and arrival rates. Its training objective balances emissions, delays and throughput. Tests in a traffic simulator compare the learned policy with conventional timing under changing demand and sensor noise.
Bibliographic reference
Use the publisher’s record for the citation format required by your journal or organisation.
Oleksander Ryzhanskyi, Vitaliy Pavlyshyn, Pavlo Radiuk, Eduard Manziuk, Oleksander Barmak, Iurii Krak. "AI-Driven Traffic Signal Control System to Reduce CO2 Emissions". YAISD-WS 2025 · CEUR Workshop Proceedings 3974, pp. 18-27, 2025.
Test a method against your business task.
Describe the decision your system needs to support and the data you have. We assess whether a method fits, define a prototype and agree how to measure its quality, response time and operating cost.