A Robust Model Predictive Control for Virtual Coupling in Train Sets

In recent decades, the demand for rail transport has been growing steadily and faces a double problem. Not only must capacity be increased, but also more flexible service is needed to meet real demand. Both objectives can achieved through virtual coupling (VC), which an evolution of current moving block systems. Trains under VC run much closer together, forming what called virtually coupled train set (VCTS). this paper, we propose approach in implemented via model predictive control (MPC). For purpose, define robust controller that predict, based on dynamic train, state system at later moments time make appropriate decisions. A MPC (RMPC) obtained by introducing two uncertain variables. The first variable added acceleration equation model, while second used uncertainty positioning. To test RMPC coupling, simulation cases are performed metro line, analysing influence both uncertainties. all cases, results show safer operation without significantly affecting service.