Fast Solver for J2-Perturbed Lambert Problem Using Deep Neural Network

This paper presents a novel and fast solver for the J2-perturbed Lambert problem. The consists of an intelligent initial guess generator combined with differential correction procedure. is deep neural network that trained to correct velocity vector coming from solution unperturbed module takes uses forward shooting procedure further update exactly meet terminal conditions. Eight sample forms are analyzed compared find optimum form train on accuracy performance this approach will be demonstrated representative test case: multirevolution problem in Jupiter system. proposed against classical standard method homotopy-based perturbed algorithm. It shown that, comparable level accuracy, significantly faster than other two.