Anti-Windup Recursive Least Squares Method for Adaptive Lookup Tables with Application to Automotive Powertrain Control Systems

摘要

This paper presents the parameter estimation methods for updating adaptive lookup tables with application to automotive powertrain control systems. Powertrain control systems are moving toward incorporating more and better adaptive features to continue improving robust performance. Lookup tables offer a computationally efficient means for storing models and controls parameters. Adaptive lookup tables are especially effective due to their capability to adapt to uncertainties that differ over time such as tolerance variations and system aging. Motivated by increasing interest, the adaptation methods of lookup tables based on anti-windup recursive least squares are considered. The effectiveness of the methods are illustrated with application to adaptation of the volumetric efficiency lookup table for a robust air-fuel ratio control of gasoline engines. The proposed adaptation methods enable fast adaptation without covariance windup even in the absence of persistent excitation.