In this paper an integrated feed-forward lateral dynamics controller, which considers the actuator dynamics and input delays, is presented. The proposed approach uses an optimization based model inversion of a nonlinear two-track model with magic formula tire model to describe the vehicle dynamics. The model is extended by the actuator models containing delays, which are identified with measurement data. A compensation of the input delays is achieved by a prediction of the driver demand, which is used for the generation of the desired output trajectory with a linear single-track model. The control allocation is formulated as a quadratic optimization problem by using the linearized dynamics of the vehicle model. A control allocation algorithm which is suitable for the application in production vehicles is presented and compared with the solution of a quadratic programming solver in a simulation study with recorded handling maneuvers. The simulation study shows, that the consideration of the actuator dynamics as well as the prediction of the driver demand yield extensive improvements of the driving dynamics and the proposed control allocation algorithm leads to accurate results for the constrained optimization problem.