A Quasi-LPV Approach for Gain Scheduling Cascaded NDI-based Controllers for Hypersonic Glide Vehicles

Abstract

Hypersonic glide vehicles (HGVs) have emerged as crucial assets with applications in both civil and military sectors, prompting significant research and development efforts. Ensuring the stability and precision of these vehicles, especially in the face of complex and uncertain physical effects, necessitates well-designed autopilots and robust control systems. Among the control methodologies explored, nonlinear dynamic inversion (NDI) stands out as a promising approach, offering the ability to handle highly nonlinear systems effectively. However, challenges such as sensitivity to model uncertainties and closed-loop instabilities pose significant hurdles. Addressing these challenges requires careful consideration of closed-loop dynamics, either through robust control approaches or tailored pole locations. This paper proposes a systematic methodology for determining suitable gains for cascaded nonlinear controllers, leveraging the similarities between NDI and quasi-linear parameter-varying (quasi-LPV) systems. By transforming nested nonlinear dynamics into a quasi-LPV framework, the proposed approach enables the utilization of linear control tools, providing stability and performance across the entire flight envelope. The contributions of the paper include discussions on the transformation of nonlinear dynamics, validations of assumptions, insights into potential applications of quasi-LPV control, and simulation results demonstrating the effectiveness of the proposed methodology.