Towards an adaptive and Intelligent automatic train operations (A.I. – ATO)


While rail-base urban rail transit is an effective and sustainable mass public transport system for cities, the significant energy consumption from running the train services is non-negligible. Beijing Metro is the city’s biggest electricity user, and 50% of the energy consumption is used by train traction force.

Following on from an earlier collaboration with BJTU and TCT on the development of energy-efficient train speed controls for Beijing Metro Line 7, the current project is to exploit state-of-the-art control technologies and machine learning in railway, and to develop energy-efficient optimal train speed control for ATO that is dynamic and adaptive to respond to real-time operation conditions, and with an intelligent self-learning process to automatically adjusting control parameters and functions to achieve the optimal performances in terms of energy efficiency, tracking precision and safety.