An automatic train operation (ATO) system is a crucial part of the communication-based train control (CBTC) system. It enables calculation and adjustment of train operation curve, which guarantees the efficient operation of urban rail transits. With limited computation resources, the present onboard ATO equipment cannot meet the latency-critical requirements of train speed profile calculation. Edge computing is considered as an effective approach to improve calculation capabilities of computation resource-constrained devices. This paper proposes an edge computing-enabled ATO system. A game-theoretic method is used to optimize the computation offloading process for the proposed system. We formulate the offloading decisions of multiple trains as a multi-user potential game. An optimal computation offloading scheme is obtained when multiple trains achieve the Nash equilibrium as game players. The numerical results certify the effectiveness of our proposed approach.
A Potential Game Based Offloading Scheme for Edge Computing-Enabled Automatic Train Operation Systems
2021-09-19
490494 byte
Conference paper
Electronic Resource
English