QUESTION:
I am working on trajectory analysis of a satellite that would travel from
Earth to the moon. Throughout my research I have found from several
references that for the perturbation modelling for n-body problem, mainly
two kinds of methods are in common use for such mission. One is Encke's
method and the other is Cowell's method. Actually I have written computer
simulations for both solving Encke's method and Cowell's method. The results
seem to be very close. I want to learn which method is more reliable for
such an interplanetary mission (like a Mars mission). For example, which
method do your computers use for trajectory analysis throughout the mission?

ANSWER from Pieter Kallemeyn on August 12, 1997:
For interplanetary navigation, we typically use a version of Cowell's
method, because of it's simplicity and the fact that an unlimited number of
perturbations can be included.  It is computationally more intensive than
Encke's method, but is not a problem with today's high-performance
computers.  There are a few applications of Encke's method used here at JPL,
usually all dealing with orbits near large bodies (Jupiter, Saturn, etc.)