Because of Pluto's distance from the Sun, the Pluto system has not yet completed a revolution since its discovery, hence an uncertain heliocentric distance. In this paper, we present the fitting of our dynamical model ODIN (Orbite, Dynamique et Intégration Numérique) to observations. The small satellites P4 and P5 are not taken into account. We fitted our model to the measured absolute coordinates (RA, DEC) of Pluto, and to the measured positions of the satellites relative to Pluto. The masses we found for the bodies of the system are consistent with those of previous studies. Yet the masses of the small satellites Nix and Hydra are artificially constrained by the number of observations of Charon. The best way to improve the determination of their masses would be to use observations of P4 and P5, but there are still not enough published observations. Concerning the heliocentric distance of the system, we compared the value we obtained using ODIN and those of other models. The difference between the models far exceeds the uncertainty needed (about 1000 km) for the mission New Horizons. A new astrometric reduction of old photographic plates may be an efficient way to constrain this distance. The ephemeris for Pluto's satellites is available on the web page of the IMCCE at http://www.imcce.fr/hosted_sites/saimirror/nssreq9hf.htm. The complete version of the ephemeris is available as a SPICE kernel at http://www.imcce.fr/~beauvalet/. Appendices A and B are available in electronic form at http://www.aanda.orgTable A.2 is only available at the CDS via anonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5) or via http://cdsarc.u-strasbg.fr/viz-bin/qcat?J/A+A/553/A14