Robust Two-Degree-of-Freedom Missile Autopilot Design via Scheduled μ-Synthesis and LPV Methods

This paper presents the design of a lateral acceleration autopilot for a high-agile dual pulse surface-to-air missile. A thorough investigation of two dissimilar robust control design approaches is performed, namely an industry proven gain-scheduling via $\mu$-synthesis and a grid-based linear parameter-varying (LPV) controller design. Both autopilots are synthesized within the same mixed sensitivity framework and implemented in a two-degree-of-freedom configuration in order to improve robust stability and tracking performance. A comparison of both autopilots is presented via $\mu$-analysis and nonlinear simulations throughout the flight envelope. The paper also presents a novel modeling and synthesis approach, the former utilizing the $\nu$-gap metric and an extended quasi-LPV modeling via function substitution. Tuning for the 2-DoF controller weighting functions is performed in a holistic optimization-based approach that incorporates fundamental controllability considerations.