Trade Offs between Energy Consumption and Control Quality in Unmanned Aerial Vehicles

Unmanned Aerial Vehicles (UAVs), particularly autonomous versions, are increasingly relevant in several fields, including human activity, nature observation, and actuation. This work explores an often overlooked question: the impact of flight control algorithms and their parameterization on the energy efficiency of UAV electronics. Using a highly configurable and comprehensive simulation environment, it evaluates the energy required to run specific control algorithms with varying parameters in a processor, while other UAV tasks are accomplished in other processors of a multiprocessor core. Experiments in energy saving mode show it is feasible to decrease the energy taken to run the control algorithm alone by 16.67%, while a maximum performance mode allows for more precise UAV operation, whenever required.