To address the potential traffic congestion in urban transportation systems, an intelligent dynamic Route Guidance System (RGS) is proposed, which calculates the road weights with real-time traffic flow predictions using the reinforcement learning method. Additionally, a training strategy is proposed for the Road Agents (RAs), which optimizes RA through interaction with the traffic environment. The effectiveness of the proposed method is verified through simulation on the Urban Mobility Simulation (SUMO) platform, and it is shown that the proposed method performs well in complex and changing urban traffic environments with different driver participation rates. Therefore, it provides a feasible solution to the congestion problem and will push forward the development of intelligent transportation systems.
Dynamic Route Guidance System Based on Real-time Vehicle-Road Collaborations with Deep Reinforcement Learning
2023-10-10
1972329 byte
Conference paper
Electronic Resource
English
Dynamic route guidance method based on vehicle-road communication
European Patent Office | 2015
|A Road Traffic Guidance Service Based on Deep Reinforcement Learning
British Library Conference Proceedings | 2020
|