Output L∞ Neural Dynamic Surface Control for Large Inertia Servo Systems

This paper presents a precise positioning intelligent control scheme for large inertia servo mechanism on the basis of effective compensation of input deadzone nonlinearity. A novel neural observer for estimating system unmeasured states is proposed which is incorporated into dynamic surface control (DSC) algorithm. The output feedback controller and adaptive laws are developed based on Lyapunov stability analysis while Nussbaum function is employed to determine the control direction. L∞ stability validation guarantees both the transient and steady state performance of overall closed loop system signals. The proposed control scheme is demonstrated numerically by applying it to a second order nonlinear system with unknown input deadzone nonlinearity.