In this paper, the advisory train cruise control stability with a driver in the loop is investigated, where the driver's state is measured by an external device. A novel stability criterion is proposed, aware of the varying behaviour of the driver in the system. For this purpose, the driver is modelled as a time varying system, and the dynamic model of the train is formed by considering rolling and aerodynamic resistance forces. In order to ascertain stability, a sampled- data based state feedback controller is considered. Converting the sampling period into a bounded time-varying delay, the addressed problem is transformed to the problem of stability analysis of time-varying delayed system. Based on Lyapunov stability theory, a novel Lyapunov Krasovskii Functional (LKF) is designed to provide sufficient conditions for the existence of L2 stability of the system in terms of solvable Linear Matrix Inequalities. A case study is given to illustrate the effectiveness of the proposed method.