Closed-Loop Identification and Real-Time Control of a Micro Quadcopter

This article presents a novel model-based control design scheme for a commercially-available micro quadcopter. A simple yet effective attitude dynamics model was identified using the closed-loop instrumental variable method with an iterative minimization search algorithm, since the actual model of the onboard attitude controller was unknown. Model-based integral-linear quadratic regulator, backstepping, and nested saturation controllers were proposed and compared for hovering control of the micro quadcopter based on the identified attitude dynamics model and its theoretical translational model. The proposed hovering controllers were first evaluated in simulations and then validated experimentally. For the conditions used in this research, it was observed that the nested saturation controller generally had the best performance among the three proposed controllers when hovering in the presence of external wind disturbances, while the backstepping controller exhibited the worst performance.