It sounds as if this problem would be well suited to analysis using Bayesian catastrophe theory. Here are some references: Harrison, P.J. and Smith, J.Q. (1980) Discontinuity, decision and conflict (with discussion). In Bayesian Statistics, Proceedings of the First Valencia Meeting May 18-June 2, 1979. pp.99-140 Smith, J.Q. (1979) Mixture catastrophes and Bayesian decision theory, Proc. Cambridge Philos. Society 86, 91-101. Smith, J.Q., Harrison, P.J. and Zeeman, E.C. (1980) The analysis of some discontinuous decision processes. Europeans Journal of Operations Research Zeeman, E.C. (1977) Catastrophe Theory, selected papers, 1972-77, New York Addison-Wesley. On Thu, 29 Jun 2000, Herman Bruyninckx wrote: > We want to estimate a very non-linear function, i.e., the hysteresis effect > of friction in robots and machine tools. What are the most appropriate > tools to do this estimation? > > We were looking in the direction of Bayesian methods, because we have quite > extensive models of the hysteresis curve. The problems we have are that > the nonlinearities are of two difficult origins: (i) multiple valued, and > (ii) with non-smooth ``switching'' behavior... More details about the > problem and the model are to be found in a PDF file at > <http://www.mech.kuleuven.ac.be/~bruyninc/hyster.pdf> > > Thanks for any hints :-) > > PS BTW where are the archives of this mailinglist? > > -- > [EMAIL PROTECTED] (Ph.D.) Fax: +32-(0)16-32 29 87 > Dept. Mechanical Eng., Div. PMA, Katholieke Universiteit Leuven, Belgium >
