Controller Design for Hybrid Drones that Hover like Helicopters and Fly like Planes

Hybrid unmanned aerial vehicles (UAVs) are drones that can hover like a helicopter as well as fly like a plane. They combine the advantages of multicopters and fixed-wing planes: vertical take-off, landing, and low energy use. However, the control systems for hybrid drones can be difficult to design. A system developed at MIT  uses a neural network to design the controller, running many simulations to continually improve the system. Their method automates the design process by training a mode-free, model-agnostic neural network controller for hybrid UAVs. The researchers demonstrate the efficacy of the proposed controller both in simulation and in their custom-built hybrid UAVs. The experiments show that the controller is robust to exploit the complex dynamics when both rotors and wings are active in flight tests.