Experimental Verification of Linear and Adaptive Control Techniques for a Two Degrees-of-Freedom Helicopter

Abstract

This paper presents the design procedure and experimental results of a high performance adaptive augmentation technique applied to a controller derived based on linear quadratic methods. The Quanser two degrees-of-freedom (2DOF) helicopter was chosen as the experimental platform on which these controllers were implemented. The paper studies the implementation of each of these controllers standalone as well as in the augmented scheme, and discusses its performance and robustness for cases with parametric uncertainties, and unmodeled dynamics. An attempt is made to combine linear quadratic tracker's reliability with the adaptive augmentation's robustness toward modeling uncertainties. It is found that appropriate tuning of parameters in the adaptive framework is key to its performance and thus the process of choosing the parameters is elaborated along with guidelines for choosing a reference model. Tuning considerations for controller implementation on the experimental setup as compared to the same on the numerical model are also addressed. The experiments performed on this system serve as a suitable research test and evaluation basis for robotics and flight control applications.