Robust adaptive control of a class of nonlinear systems with unknown dead-zone. (Q1428128)

Practical control systems may have dead zones in physical components. The presence of the dead zones, typically with unknown parameters, can severely limit the performance of the control system. This paper proposes a robust adaptive control architecture for a class of continuous time nonlinear dynamic systems with a dead zone. The authors use a new description of a dead zone and thereby avoid the need to construct a dead zone inverse to minimize the effects of the dead zone. The dead zone model used here is a piecewise linear function \(w(t)\) that is zero in the interval \([b_\ell, b_r]\) and with identical positive slope \(m\) outside this interval. It is assumed that \(w(t)\) is not available for measurement, and that \(b_\ell\), \(b_r\) and \(m\) are unknown. This is a static approximation of a physical phenomenon with negligible fast dynamics, and serves as a good model for a servo motor or a hydraulic servo valve. The authors prove global stability of the entire system and demonstrate stabilization and tracking within a desired precision. Simulation results are provided that illustrate their approach.