Disturbance compensation methods are widely used to design the robust controller. In order to achieve the robust parallel control, how to implement the disturbance compensation in parallel control laws is studied in this paper. First, the key points are the estimations of the total inputs via inverse systems and the application of system state derivatives. Then, the inverse system based parallel control (ISPC) method is proposed for the optimal control of nonlinear systems. The basic structure of the inverse system based parallel control method is explained and compared with the traditional parallel control methods. The adaptive dynamic programming (ADP) method based on an approximate value function is used to solve the parallel control law. Finally, numerical simulations demonstrate the feasibility of the inverse system based parallel control method.