Modeling scheme for vehicle longitudinal control

Abstract

A modeling scheme for vehicle longitudinal control based on system input/output data is developed. A family of linear fractional transformation (LFT) models is used to describe vehicle dynamics. Each of the LFT models is valid over a given dynamic range around its operating point. With these LFT models available, robust controllers can be designed using available tools. Strategies and algorithms for building LFT models of vehicle longitudinal control are introduced. Attention is given to excitation signals for the system input, H/sub infinity / identification methods, and coverage of the system uncertainties. Experiments with a test vehicle have been conducted using these techniques, and a family of LFT models for vehicle longitudinal control has been constructed based on the experimental data. The modeling method is illustrated by using simulation data. >