Design, Analysis, and Learning Control of a Robotic Wind Turbine

Wind power represents one of the most promising sources of renewable energy, and improvements to wind turbine design and control can have a significant impact on energy sustainability. Although wind energy has not been a canonical research area in robotics, we argue here that many robotics techniques (physical system design, modeling and control, and policy learning techniques) can in fact have a large impact on wind energy research. Pursuing this goal, in this paper we develop a small, fully functional robotic wind turbine, capable of variable speed operation and individual pitch actuation. We analyze the power output of the turbine under different operating conditions, comparing to predictions from aerodynamic simulators. Using inverse dynamics control we demonstrate high-frequency blade pitch control synchronized to rotor angle. Finally, using a novel Reinforcement Learning policy search algorithm, we show that the system can quickly optimize its energy output in an online and fully model-free manner.