In this paper, a Physics-Informed Neural Network (PINN) is presented for control purposes. The dynamics of the system are integrated into the training process of the neural network to improve training accuracy. In the control loop, an optimization-based algorithm is also used, which ensures the safe operation of the system. One of the main requirements against the optimization-based algorithm is the low computational cost. Moreover, in the paper, a comparative analysis is presented of the conventionally trained and the PINN-based results for control purposes. The results are illustrated through a vehicle-oriented problem, considering parameter uncertainty and nonlinear effects.