Robust Design of Low-Thrust Gravity-Assist Trajectories via Reinforcement Learning

What is it about?

The electric propulsion spacecraft is becoming increasingly used in deep space exploration, in which the leveraging of gravity assist is an important technique for shortening the flight time and saving fuel. In the missions, there exist various uncertainties threatening the successful execution of missions. It is attractive for the spacecraft to deal with uncertainties in real-time. We develop a Reinforcement-Learning based multi-phase robust gravity-assist trajectory design method for the low-thrust exploration mission with gravity assist under uncertainties.

Featured Image