Multi-Terminal HVDC systems is one of the most promising technologies for large scale integration of electrical grids. The control of this type of systems is a challenging regulation problem with multiple inputs and multiple outputs. This paper presents a model predictive approach for this task where states are determined by a measurement as well as estimation when measurement is unavailable. Delay in communications as well as noise in the measurements are considered in the control. Simulations results on the CIGRE five terminals HVDC grid demonstrate the performance of the proposed control under different scenarios including N-1 contingencies in both the transmission lines and the communications.