Improved robust adaptive controller for a class of partially known nonlinear systems

A class of partially known systems, in this case robot manipulators, is studied. The known dynamics are separated out and used to perform a feedback linearization on the nonlinear system. An adaptive controller is designed for the remaining unknown portion of the plant. The objective is to design a robust adaptive controller with signal synthesis to improve the convergence rate, reduce transient oscillation, and yield better disturbance rejection. The adaptive controller maintains the computed torque structure with a parameter estimation scheme; in addition, an auxiliary signal is synthesized. The authors also attempt to develop a design technique by which the convergence rate of the system can be improved by picking a design parameter α, the degree of stability of the model