Modeling, Simulation and Control of a Tailsitter Tiltrotor MAV

Abstract

Technological advances in micro aerial vehicles have led to a considerable interest in the customization of such vehicles for numerous new applications. Especially attractive are drones with vertical take-off and landing capabilities. This paper discusses the modeling, simulation and control of the longitudinal dynamics of an unconventional tailsitter-tiltrotor MAV intended for environmental sampling missions. The rotor forces, including the effect of tilt-deflections, are modelled using combined momentum and blade element theory. The aerodynamic modeling takes nonlinearities into account by blending a flat-plate approximation with lifting line theory. The nonlinear flight dynamics model is trimmed and linearized at specific operating points, to evaluate the open-loop dynamics and facilitate controller development. A longitudinal control architecture is then designed for the two main flight regimes, aircraft mode and helicopter mode. To cover the transition phases between modes, the two separate control architectures are fused by means of a transition number, which itself is a measure of the extent to which the aircraft is in one phase of flight or another. The devised control framework is tested in a simulation environment constructed out of the developed flight dynamics model. Results show effective control in both helicopter and aircraft mode, as well as demonstrate successful transition flight. Controllability challenges and limitations specific to the tailsitter-tiltrotor design without aerodynamic control surfaces have also been identified.