System model building and dynamic online control of traffic flow

Abstract

The Underground transportation system plays a vital role in public transportation and exhibits complicated dynamics. The model predictive control approach for underground train dispatching is proposed in this paper. The system modelling technique spatially aggregates trains to generate the traffic flow model in a network of interconnected control units. The state-space model for the underground train traffic of a metro line is investigated. Simulation results are reported to demonstrate the effectiveness of the optimization control model.