In this paper, a sliding mode controller based on linear matrix inequality(LMI) is designed to solve the path tracking problem for autonomous vehicle systems with mismatched uncertainties. Firstly, the nonlinear system model is established based on the vehicle kinematics and dynamics equations, taking into account of the uncertainty caused by the sideslip of the tire, the constraint of vehicle sideslip and random interference. Secondly, a linear sliding mode path tracking control method is proposed for the linearized model. LMI is used to provide sufficient conditions for the existence of sliding mode surface which gurantees the quadratic stability of the closed-loop system. Finally, in the serret-Frenet coordinate system, the effectiveness of the method proposed in verified. Simulation results show that the reference path is tracked accurately using the controlled designed in this paper, which provides strong robustness.