Variable speed limit control strategy for freeway tunnels based on a multi-objective deep reinforcement learning framework with safety perception
Published in Expert Systems with Applications, 2024
Recommended citation: Jin, J., Huang, H., Li, Y., Dong, Y., Zhang, G., & Chen, J. (2024). Variable speed limit control strategy for freeway tunnels based on a multi-objective deep reinforcement learning framework with safety perception. Expert Systems with Applications, 126277. https://www.sciencedirect.com/science/article/abs/pii/S0957417424031440
’– Developing an RCR prediction model based on random deep and cross networks. This model aims to achieve the safety perception function of VSL systems in freeway tunnels, taking into account the current traffic detection conditions of most freeway tunnels. – Integrating the RCR prediction model into the reinforcement learning process of VSL agents to design an MDRL framework with safety perception. – Developing an application-oriented VSL strategy for traffic safety and efficiency improvement of freeway tunnels based on the safety-perception MDRL framework for the first time.’
Recommended citation: Jin, J., Huang, H., Li, Y., Dong, Y., Zhang, G., & Chen, J. (2024). Variable speed limit control strategy for freeway tunnels based on a multi-objective deep reinforcement learning framework with safety perception. Expert Systems with Applications, 126277.