It is well known that designing controllers for plants with nonlinear dynamics and modeling uncertainties are quite challenging. This presentation focuses on the controller fragility and performance deterioration issues due to inaccuracies in controller implementation. This calls for redesigning the controller to ensure satisfactory performance. In this regard, the presentation is composed of two parts: a theoretical part and an application part. In the theoretical part, it addresses the problem of non-fragile adaptive control problem for a class of continuous-time systems with state-delays and norm-bounded uncertainties against controller gain variations. In the application part a robust redesign technique for the control of a class of nonlinear systems with particular emphasis on the inverted wedge. The inverted wedge is an under-actuated planar robotic system with two degrees of freedom and a single control input. The control objective is to design a feedback controller that regulates the system to zero-state, that is, to balance the wedge in the inverted position by shifting the weight on the top of the wedge.
Group for Research in Decision Analysis