Enhancing Communication-Based Train Control Systems Through Train-to-Train Communications

High reliability and low latency are crucial for urban rail transits. In this paper, we introduce communication strategies for communication-based train control (CBTC) systems using long-term evolution for metro (LTE-M) to improve the reliability and latency. To be specific, the FlashLinQ-based Train-to-Train (T2T) communication schemes are introduced considering both the transmission delay and the packet drop. The quantified resilience is also introduced as a system metric to evaluate the preservation and recovery performance of CBTC systems. First, a novel urban rail transit wireless communication model is established using FlashLinQ-based T2T communications. Then, we introduce a novel cognitive control scheme based on LTE-M with T2T communication to enhance the quality of service and the resilience of multi-train CBTC systems. In the introduced scheme, Q-learning is used to generate optimal control strategies considering both wireless communication parameters adaption and train control parameters. Extensive simulations are conducted and the results show that the resilience of CBTC systems can be enhanced using the introduced scheme. Furthermore, using the introduced scheme, not only the gaps in optimal velocity versus distance curve are smaller, but also the unplanned traction and breaking are reduced as well.