2020-03-25, 08:54 PM
Ah, but to calculate the perturbations, you need to know the positions of all of the potentially perturbing bodies and their motions. The osculating elements are supposed to be a simplification of the orbital mechanics problem and make use of the Keplerian approximations. These approximations are 'good enough' when used within a sufficiently close time of the epoch of the osculating elements. We're not trying to hit these things with a spacecraft & even NASA has to employ midcourse corrections to pull that off 
.
Phil S.
.Phil S.