Fast Evaluation of Low-Thrust Transfers via Deep Neural Networks

TL;DR

This paper introduces a deep neural network approach for rapid and accurate evaluation of low-thrust interplanetary transfers, including feasibility and fuel estimation, significantly speeding up mission design processes.

Contribution

It develops a novel DNN-based method with an efficient database generation for quick transfer feasibility and fuel consumption evaluation in interplanetary missions.

Findings

Transfer feasibility judged with >98% accuracy

Fuel consumption estimated with <0.4% error

Method outperforms traditional evaluation techniques

Abstract

The design of low-thrust-based multitarget interplanetary missions requires a method to quickly and accurately evaluate the low-thrust transfer between any two visiting targets. Complete evaluation of the low-thrust transfer includes not only the estimation of the optimal fuel consumption but also the judgment of transfer feasibility. In this paper, a deep neural network (DNN)-based method is proposed for quickly evaluating low-thrust transfer. An efficient database generation method is developed for obtaining both the infeasible and optimal transfers. A classification DNN and a regression DNN are trained based on the infeasible and optimal transfers to judge the transfer feasibility and estimate the optimal fuel consumption, respectively. The simulation results show that the well-trained DNNs are capable of quickly determining the transfer feasibility with a correct rate of greater than 98% and approximating the optimal transfer fuel consumption with a relative estimation error of less than 0.4%. The tests on two asteroid chains further show the superiority of the DNN-based method for application to the design of low-thrust-based multitarget interplanetary missions