Unmanned Aerial Vehicles (UAVs) are currently very interesting due to the numerous applications and possibilities that they make possible, and consequently, they have a significant presence in the research community. They have to deal with an important limitation, which is their autonomy. UAVs are usually vehicles of reduced dimensions and weight, consisting of different electronic subsystems, like the autopilot subsystem, the communication equipment, the servos and actuators or the propulsion subsystem. All of this required energy is provided for the on-board batteries. There is a trade-off between boarded equipment weight and its total consumption. This compromise is finally reflexed in the autonomy of the vehicle. In this project we will focus on the auto-pilot system consumption. Despite the relevance of the auto-pilot power consumption is relative to the whole consumption, it will be more critical for small UAVs. This problem can be considered from different points of view. In this project we will analyse what is the influence of the control strategy over the final power consumption. It is desired a reduction of the power keeping a correct control of the vehicle. In the assessment of the selected control strategies different type of UAVs will be considered. For the control application it will be needed mathematical models of the UAVs. Considering four different groups of UAVs, which are fixed-wing, flexible-wing, quad-rotors and helicopters, we will elaborate one unique and representative model for each group of UAVs. Regarding the control techniques, we will consider optimal control, PID control and adaptive control. The fact of introduce them in UAVs will allows us to study what is their influence on the power consumption. Finally, once the techniques will have been applied, it will be seen the advantages that Adaptive control has over the rest of strategies. This technique will show the best results regarding the power consumption, providing important reductions respect the initial subsystem power consumption.