The contribution of this work is the numerical simulation and the experimental assessment of a network distributed control system using an unmanned aerial vehicle and a remote master controller connected by a wireless network. A novel multi-input multi-output long-range predictive horizon minimum variance control approach is developed and implemented in altitude, heading, lateral and longitudinal velocities control problems under the influence of process noise, measurement noise and time-delay. The proposed autopilots exhibited damped and fast response speeds with optimal control and output variance minimization, outperforming a classic model predictive control approach in convincing experimental field tests in real-world environments and scenarios.