Design Study of an Autonomous Unmanned Air Vehicle Controller

Abstract

The purpose of this document is to both quantitatively and qualitatively compare three varying approaches to control of unmanned air vehicles: dynamic inversion, classical gain scheduling, and robust gain scheduling through H∞ synthesis. The quantitative comparisons include robust performance and robust stability measures of the aircraft and controller linearized about a trim operating point. A second study will look at the time response of the system with varying perturbation to the nominal plant dynamics. The qualitative analysis looks at the complexity of the controller and time required to implement, this included comparison of iterative solving methods for system gains both in the frequency and time domains. A full nonlinear form of the dynamic inversion controller was implemented with full non-linear simulation in Simulink and shows comparable results to the linearized studies. The results have shown that a dual loop controller with a dynamic inversion inner loop and H∞ outer loop has the largest robust performance over the flight envelope but one of the more advanced forms of control for implementation.