Gaussian processes are a probabilistic, non-parametric approach to modelling that allows easy merging of ordinary measured data and local linear models. This can be of particular importance in the identification of non-linear dynamic systems from experimental data, where there is usually more data available around the equilibrium points and only sparse data is available far from them. The utility of the Gaussian process model for predictive control is investigated in this paper and is illustrated on a simple first-order vehicle dynamics.
Allgöwer, F. and Zheng, A., editors, 2000: Nonlinear model predictive control. Progress in system and control theory, Volume 26. Birkhäuser Verlag.
2.
Allgöwer, F., Badgwell, T.A., Qin, S.J., Rawlings, J.B. and Wright, S.J.1999: Nonlinear predictive control and moving horizon estimation - an introductory overview. In Frank, P.M., editor. Advances in control: highlights of ECC'99 . Springer, 391-449.
3.
Bishop, C.M. and Tipping, M.E.2003: Bayesian regression and classification. In Suykens, J., Horvath, G. , Basu, S., Micchelli, C. & Vandewalle, J., editors. Advances in learning theory: methods, models and applications. NATO Science Series III: Computer and Systems Sciences. IOS Press, 190.
4.
Girard, A.2004: Approximate methods for propagation of uncertainty with Gaussian process models. PhD thesis, University of Glasgow , Glasgow. http://www.dcs.gla.ac.uk/agathe/ThesisFinal.ps
5.
Girard, A., Rasmussen, C.E. and Murray-Smith, R.2002: Multi-step ahead prediction for non linear dynamic sytems - a Gaussian process treatment with propagation of the uncertainty. In Becker, S.,Thrun, S. & Obermayer, K., editors. Advances in neural information processing systems. Volume 15. MIT Press, 545-52.
6.
Gregorčič, G. and Lightbody, G.2003: An affine Gaussian process approach for nonlinear system identification. System Science29, 47-63.
7.
Johansen, T.A., Hunt, K.J., Gawthrop, P.J. and Fritz, H.1998: Off-equilibrium linearization and design of gain scheduled control with application to vehicle speed control. Control Engineering Practice6, 167-80.
8.
Johansen, T.A., Shorten, R. and Murray-Smith, R.2000: On the interpretation and identification of dynamic Takagi-Sugeno fuzzy models. IEEE Transactions on Fuzzy Systems8, 297-313.
9.
Kocijan, J. and Girard, A.2005: Incorporating linear local models in Gaussian process model. In Proceedings of IFAC World Congress 2005 , Prague.
10.
Kocijan, J. and Leith, D.J.2004: Derivative observations used in predictive control. In Proceedings IEEE Melecon 2004, Dubrovnik .
11.
Kocijan, J. and Murray-Smith, R.2005: Nonlinear predictive control with a Gaussian process model. In Murray-Smith, R. and Shorten, R., editors. Switching and learning in feedback systems. Springer.
12.
Kocijan, J., Girard, A., Banko, B. and Murray-Smith, R.2005: Dynamic systems identification with Gaussian processes. Mathematical and Computer Modelling of Dynamic Systems11, 411-24.
13.
Kocijan, J., Girard, A. and Leith, D.J.2003: Incorporating linear local models in Gaussian process models. IJS report DP-8895. Jozef Stefan Institute.
14.
Kouvaritakis, B. and Cannon, M., editors, 2001: Nonlinear predictive control, Theory and practice. IEE Control Engineering61. IEE.
15.
Leith, D.J., Leithead, W.E., Solak, E. and Murray-Smith, R.2002: Divide & conquer identification: Using Gaussian process priors to combine derivative and non-derivative observations in a consistent manner. In Conference on Decision and Control 2002, Las Vegas, 624-29.
16.
Likar, B. and Kocijan, J.2007: Predictive control of a gas-liquid separation plant based on a Gaussian process model. Computers & Chemical Engineering31, 142-52.
17.
Maciejowski, J.M.2002: Predictive control with constraints. Pearson Education Limited.
18.
Murray-Smith, R., Johansen, T.A., and Shorten, R. (1999): On transient dynamics, off-equilibrium behaviour and identification in blended multiple model structures. In European Control Conference 1999, Karlsruhe , BA-14.
19.
Rasmussen, C.E. and Williams, C.K.I.2006: Gaussian processes for machine learning. MIT Press.
20.
Richalet, J.1993: Industrial applications of model based predictive control . Automatica19, 1251-74.
21.
Solak, E., Murray-Smith, R., Leithead, W.E., Leith, D.J. and Rasmussen, C.E.2002: Derivative observations in Gaussian process models of dynamic systems. In Becker, S., Thrun, S & Obermayer, K., editors. Advances in neural information processing systems, Volume 15. MIT Press, 529-36.
