System Identification for Automotive Systems: Opportunities and Challenges
Without control many essential targets of the automotive industry could not be achieved. As control relies directly or indirectly on models and model quality directly influences the control performance, especially in feedforward structures as widely used in the automotive world, good models are needed. Good first principle models would be the first choice, and their determination is frequently difficult or even impossible. Against this background methods and tools developed by the system identification community could be used to obtain fast and reliably models, but a large gap seems to exist: neither these methods are sufficiently well known in the automotive community, nor enough attention is paid by the system identification community to the needs of the automotive industry. This introduction summarizes the state of the art and highlights possible critical issues for a future cooperation as they arose from an ACCM Workshop on Identification for Automotive Systems recently held in Linz, Austria.
KeywordsModel Predictive Control Homogeneous Charge Compression Ignition Principle Model Hammerstein Model Optimal Experiment Design
Unable to display preview. Download preview PDF.
- 1.Fliess, M., Normand-Cyrot, D.: On the approximation of nonlinear systems by some simple state-space models. In: Bekey, G., Saridis, G. (eds.) Proceedings of the Sixth IFAC Symposium on Identification and System Parameter Estimation 1982. IFAC, vol. 1, pp. 511–514. Pergamon, Oxford (1983); Proceedings of the Sixth IFAC Symposium on Identification and System Parameter Estimation 1982, Washington, DC, USA, June 7-11 (1982)Google Scholar
- 4.Hirsch, M., Alberer, D., del Re, L.: Grey-box control oriented emissions models. In: Proc. 17th IFAC World Congress, Seoul, South Korea (2008)Google Scholar
- 6.Hjalmarsson, H., Mårtensson, J.: Optimal input design for identification of non-linear systems: Learning from the linear case. In: American Control Conference, New York City, USA (2007)Google Scholar
- 8.Krener, A., Isidori, A., Respondek, W.: Partial and robust linearization by feedback. In: Proceedings of the 22nd IEEE Conference on Decision and Control, New York, NY, USA, vol. 1, pp. 126–130 (1983); Proceedings of the 22nd IEEE Conference on Decision and Control, San Antonio, TX, USA, December 14-16 (1983)Google Scholar
- 9.Larsson, C., Rojas, C., Hjalmarsson, H.: MPC oriented experiment design. In: 18th IFAC World Congress, Milano, Italy (to appear, 2011)Google Scholar
- 10.Ljung, L.: System Identification: Theory for the User. Prentice-Hall, Upper Saddle River (1998)Google Scholar
- 11.Oppenauer, K.S., del Re, L.: Hybrid 2-zone Diesel combustion model for NO formation. In: SAE ICE 2009 – 9th International Conference on Engines and Vehicles, SAE 2009-24-0135. Capri, Italy (2009)Google Scholar
- 12.Parrilo, P., Ljung, L.: Initialization of physical parameter estimates. In: van der Hof, S.W.P., Wahlberg, B. (eds.) Proc. 13th IFAC Symposium on System Identification, pp. 1524–1529. Rotterdam, The Netherlands (2003)Google Scholar
- 13.Röpke, K., von Essen, C.: DoE in engine development. Quality and Reliability Engineering International 24(6), 643–651Google Scholar