A Cognitive Control Approach to Communication-Based Train Control Systems

Abstract

Communication-based train control (CBTC) is an automated train control system using bidirectional train-ground wireless communications to ensure the safe operation of rail vehicles. Due to unreliable wireless communications and train mobility, the train control performance can be significantly affected by wireless networks. Although some works have been done to study CBTC systems from both train-ground communication and train control perspectives, these two important areas have traditionally been separately addressed. In this paper, with recent advances in cognitive dynamic systems, we take a cognitive control approach to CBTC systems considering both train-ground communication and train control. In our approach, the notion of information gap is adopted to quantitatively describe the effects of train-ground communication on train control. Moreover, unlike the existing works that use network capacity as the design measure, in this paper the linear quadratic cost for the train control performance in CBTC systems is considered in the performance measure. Reinforcement learning is applied to obtain the optimal policy based on the performance measure, which includes linear quadratic cost and information gap. In addition, the wireless channel is modeled as finite-state Markov chains with multiple state transition probability matrices, which can demonstrate the characteristics of both large-scale and small-scale fading. The channel state transition probability matrices are derived from real field measurement results. Simulation results show that the proposed cognitive control approach can significantly improve the train control performance in CBTC systems.