Design and Evaluation of a Model-Based Controller for Flapping-Wing Micro Air Vehicles

Abstract

Because of the highly integrated nature of their dynamics, flapping-wing micro air vehicles exhibit significant coupling, including interactions between aeroelastic wing behavior, actuator dynamics, and control systems. An existing model-based controller is demonstrated to exhibit reduced tracking performance and even closed-loop instability when exposed to these effects, due to differences between the models used for control design and control evaluation. Motivated by this analysis, a controller is developed that provides improved robustness despite model uncertainty by decoupling the normal form of the vehicle dynamics, which accounts for coupling of the forces and moments acting on the vehicle and enables enhanced tuning capabilities. Closed-loop stability is achieved when the control system is evaluated on a richer model that includes the effect of wing flexibility and limited actuator capabilities, despite a reduction in control effectiveness. Using the same control design methodology, a minimally actuated configuration is achieved by removing wing bias manipulation from the control parameters. It is shown that the controller is still able to stabilize the resulting system, although with a reduction in robustness against unmodeled dynamics, which is expected due to the increased dependence of the control effort on the remaining control parameters.