Enhanced maximal nonsaturating effort adaptive control

Abstract

An enhanced maximal nonsaturating effort adaptive control (EMNEAC) scheme for servo-control systems that have input saturation and deadbands is described. The proposed EMNEAC builds on two key online enhancement techniques: a knowledge-enhanced parameter identification (KEPI) algorithm and a maximal nonsaturating effort selection (MNES) algorithm. The KEPI algorithm uses a priori information about the characteristics of the physical system being controlled. It generates parameter estimates which lie within admissible ranges of physical values. The EMNEAC scheme incorporates an automatic self-selection of suitable closed-loop performance specifications for online self-tuning control design. It uses heuristic information derived from computed control input and designs a controller whose output level is kept with the admissible set of control. The result is that the EMNEAC scheme puts out maximal control effort for high tracking performance without saturating the control input of the system. It is more robust in terms of stability and achieves superior performance over other standard self-tuning control algorithms.<>