Predictive Guidance and Control for a Tail-Sitting Unmanned Aerial Vehicle

This paper is concerned with the research and development of a predictive control and guidance system for a tail sitting, unmanned aerial vehicle (UAV). The vehicle in question is the T-wing and it is a tail-sitting UAV. The T-Wing can fly in vertical and horizontal flight modes, using propeller wash over the control surfaces for control in vertical flight. These two flight conditions and the fundamentally different vehicle dynamics between them give rise to a challenging control problem. The predictive control and guidance strategy developed in this research synthesises non-linearity by creating a new linear model from a non-linear vehicle model at every time step. This methodology allows the use of efficient quadratic programming techniques to find the control solution. The continuous linearization ensures that the current linear problem is always a close approximation to the current nonlinear situation. This predictive controller has been successfully flight-tested in the vertical flight mode, running in real-time on a 400 MHz, PC-104 flight computer using xPC Target software.