Testing and System Identification of an Ornithopter in Longitudinal Flight

There is currently a large effort underway to understand the flight dynamics of avian-based flapping-wing vehicles, or ornithopters, as they represent a critical intersection between existing biological flyers and the need for small aerial robots to conduct a variety of mission scenarios. Efforts to model the flight dynamics of these vehicles for feedback control have been complicated by a number of factors including nonlinear flight motions, unsteady aerodynamics at low Reynolds numbers, and limited sensor payload capacity. This paper presents data for a 0.45 kg ornithopter research platform, flown in straight and level mean flight. A visual tracking system was employed to follow retroreflective markers on the ornithopter and reconstruct state measurements. A multibody model of the flight dynamics was used to investigate the spatial distribution of kinematic variables over the duration of a wing stroke, and system identification techniques were employed to extract models for the lift, thrust, and pitching moment coefficients. Two methods of parameter estimation showed good results for relatively simple aerodynamic models that can be used for feedback control.