With the advent of smart urban spaces, efficient and flexible traffic management has become a necessity. The changing nature of urban traffic makes traditional traffic management systems—which rely on pre-established systems—inefficient. This study presents a deep reinforcement learning (DRL) model for real-time traffic management in smart cities. To optimize traffic flow and reduce congestion, the model uses deep-rooted reinforcement learning to automatically adapt to changing traffic conditions Urbanized several simulated systems representing a range of vehicle scenarios were used to evaluate our technology. According to the results, the traffic volume increased significantly and the typical waiting time decreased compared to the traditional traffic management system. The program can also successfully accommodate unexpected traffic incidents such as accidents or roadblocks. This study demonstrates the potential of DRL to transform urban traffic management and create a more sustainable, efficient and effective smart city.
Deep Reinforcement Learning (DRL) for Real-Time Traffic Management in Smart Cities
23.11.2023
528473 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Real-Time Traffic Monitoring with AI in Smart Cities
Springer Verlag | 2025
|TrafficIntel: Smart traffic management for smart cities
IEEE | 2017
|