Adaptive backstepping control for miniature air vehicles

There are a variety of potential military and civil applications for miniature air vehicles (MAVs). However, commercial viability will require that low level control loops do not require tuning for each individual airframe. The objective of this paper is to explore the use of nonlinear model reference adaptive control for pitch and roll attitude hold. We derive algorithms that are well suited for implementation on MAVs and demonstrate their effectiveness through flight tests on a 1.2 meter wingspan airframe using the Kestrel autopilot. The performance is compared to specially tuned PD controllers on the same airframe