Robust observer-based controllers for systems with non-homogeneous uncertainties

The author presents a unified approach to the stabilization problem for uncertain systems using a reduced order observer based on state feedback. A well-developed algorithm is presented that not only yields an optimum Lyapunov matrix in a systematic technique, but also expands the allowable set of tolerable uncertainty through introducing the concept of a nonhomogeneous polyhedron of large parameter variations. A comparative example involving the AFIT/F16 flying at 3000 ft and Mach 0.6 is used to demonstrate the proposed technique while displaying its efficiency. The proposed technique yields wider ranges of parameter variations than those reported in the literature.<>