Published Paper Details:

Congestion during rush hours remains the emergent challenge in cities globally. Conventional traffic management systems, largely dependent on fixed timing or basic rule-based reasoning, fail to respond to dynamic traffic information and provide minimal assistance in extreme situations such as emergency vehicle travel. In this paper we propose a smart traffic management model integrating Reinforcement Learning and Blockchain technology to develop an even more responsive, transparent, and decentralized solution. Our RL agents are trained based on real traffic patterns in a simulated environment and adjust traffic signal timing dynamically to optimize traffic flow. Meanwhile, blockchain smart contracts on a blockchain system record the systems performance, trigger emergency responses, and ensure data integrity without the presence of any central authority. When an emergency vehicle is detected, the system overrides the RL agents' decisions briefly to give way, maintaining safety in a way that does not compromise learning. Simulation results demonstrate enormous reductions in waiting times of automobiles, clearing congestion, and handling emergencies appropriately. Simulation results present remarkable reductions in waiting times of automobiles, clearing congestion, and addressing emergencies appropriately.