Automotive systems identification: application of neuro-hybrid models

  • Authors:
  • Ventura Assuncao

  • Affiliations:
  • IAV GmbH, Gifhorn, Germany

  • Venue:
  • MS'06 Proceedings of the 17th IASTED international conference on Modelling and simulation
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

The identification of automotive systems requires big efforts regarding time and resources that are not always included in project planning. Nevertheless during project execution a lot of measurements are performed directly on the vehicle's ECU (Engine Control Unit) which contain by themselves useful information concerning different vehicle systems. The translation of such information into system models would imply a large time saving when doing calibration work. In order to fulfil this objective the so called neuro-hybrid models are introduced. Neurohybrid models have a structure where a parametric and a nonparametric model are connected together by a signal demodulator with the demodulator providing for an optimal retention of the behaviour of the system under identification.