Adaptive output feedback stabilization of nonlinear systems

Output feedback control design techniques are required in practice due to the limited number of sensors/measurements available for feedback. This thesis focuses on output feedback controller design techniques for nonlinear systems subject to different system restrictions. The problem of controlling the heart dynamics in a real time manner is formulated as an adaptive learning output-tracking problem. For a class of nonlinear dynamic systems with unknown nonlinearities and non-affine control input, a Lyapunov-based technique is used to develop a control law. An adaptive learning algorithm is exploited that guarantees the stability of the closed-loop system and convergence of the output tracking error to an adjustable neighborhood of the origin. In addition, good approximation of the unknown nonlinearities is also achieved by incorporating a persistent exciting signal in the parameter update law. The effectiveness of the proposed method is demonstrated by an application to a cardiac conduction system modelled by two coupled driven oscillators.;The output feedback control design for a class of non-minimum phase nonlinear systems with unknown nonlinearities is studied. The proposed approach is able to combine the two previous design methods and provide a stabilizing output feedback control law. The performance of the proposed method is demonstrated by simulation results.;An output feedback design technique is developed to achieve semi-global practical stabilization for a class of non-minimum phase nonlinear systems, subject to parameter uncertainties. This work provides a constructive controller design method for an auxiliary system, whose existence is crucial, but is only assumed in (Isidori, 2000). The control design technique is used to regulate the benchmark van de Vusse reactor. Simulation results demonstrate satisfactory controller performance.