Research/ SPACE AI & AUTONOMY

Using Reinforcement Learning to Design a Low-Thrust Approach into a Periodic Orbit in a Multi-Body System

Christopher J. Sullivan, Natasha Bosanac

AIAA SciTech 2020/Jan 5, 2020/ 32 citations

Space × AI

Technique

Learning-based MethodsOpenAlex topic: Reinforcement Learning in Robotics · Guidance, Dynamics & ControlOpenAlex keyword: trajectory optimization

Indexed as

OpenAlexSpacecraft Dynamics and Control · Reinforcement Learning in Robotics · Space Satellite Systems and Control

DOI Semantic Scholar OpenAlex

Exploring multi-body systems with low-thrust-enabled small satellites opens up a new frontier of scientific opportunities. However, the limited control authority of these spacecraft introduces significant challenges during trajectory and maneuver design. In this paper, a deep reinforcement learning method is used to design a control profile for a low-thrust-enabled small satellite to approach a periodic orbit from a nearby initial condition and over a short time horizon. Multiple neural networks are trained to recover controlled paths that meet a variety of objectives while tracking to a reference trajectory on a stable manifold. Monte Carlo simulations for three distinct formulations of a reward function are performed using a control scheme derived from trained neural networks.

Using Reinforcement Learning to Design a Low-Thrust Approach into a Periodic Orbit in a Multi-Body System

Related papers

Exploring Transfers between Earth-Moon Halo Orbits via Multi-Objective Reinforcement Learning

Autonomous Guidance Between Quasiperiodic Orbits in Cislunar Space via Deep Reinforcement Learning

Reinforcement Learning for Sequential Low-Thrust Orbit Raising Problem

Deep Reinforcement Learning Policies for Underactuated Satellite Attitude Control

Exploring the Low-Thrust Transfer Design Space in an Ephemeris Model via Multi-Objective Reinforcement Learning

A Comparative Study of Reinforcement Learning in Spacecraft Orbit Control

Using Reinforcement Learning to Design Missed Thrust Resilient Trajectories

Real-Time Guidance for Low-Thrust Transfers Using Deep Neural Networks

Spacecraft Autonomous Rendezvous Control With Rotation Target on SE(3): Deep Reinforcement Learning Methods

Multi-agent reinforcement learning satellite guidance for triangulation of a moving object in a relative orbit frame

More from AIAA SciTech

Deep Learning Crater Detection for Lunar Terrain Relative Navigation

Fast Trajectory Optimization via Successive Convexification for Spacecraft Rendezvous with Integer Constraints

Optimization of Low Reynolds Number Airfoils for Martian Rotor Applications Using an Evolutionary Algorithm

CNN-Based Pose Estimation System for Close-Proximity Operations Around Uncooperative Spacecraft

Using Reinforcement Learning to Design a Low-Thrust Approach into a Periodic Orbit in a Multi-Body System

Related papers

Exploring Transfers between Earth-Moon Halo Orbits via Multi-Objective Reinforcement Learning

Autonomous Guidance Between Quasiperiodic Orbits in Cislunar Space via Deep Reinforcement Learning

Reinforcement Learning for Sequential Low-Thrust Orbit Raising Problem

Deep Reinforcement Learning Policies for Underactuated Satellite Attitude Control

Exploring the Low-Thrust Transfer Design Space in an Ephemeris Model via Multi-Objective Reinforcement Learning

A Comparative Study of Reinforcement Learning in Spacecraft Orbit Control

Using Reinforcement Learning to Design Missed Thrust Resilient Trajectories

Real-Time Guidance for Low-Thrust Transfers Using Deep Neural Networks

Spacecraft Autonomous Rendezvous Control With Rotation Target on SE(3): Deep Reinforcement Learning Methods

Multi-agent reinforcement learning satellite guidance for triangulation of a moving object in a relative orbit frame

More from AIAA SciTech

Deep Learning Crater Detection for Lunar Terrain Relative Navigation

Fast Trajectory Optimization via Successive Convexification for Spacecraft Rendezvous with Integer Constraints

Optimization of Low Reynolds Number Airfoils for Martian Rotor Applications Using an Evolutionary Algorithm

CNN-Based Pose Estimation System for Close-Proximity Operations Around Uncooperative Spacecraft