Author Information

Is this project an undergraduate, graduate, or faculty project?

Undergraduate

Project Type

individual

Campus

Daytona Beach

Authors' Class Standing

Shiloh Cuffe, Senior

Lead Presenter's Name

Shiloh Cuffe

Lead Presenter's College

DB College of Engineering

Faculty Mentor Name

Dr. Di Wu

Abstract

This project investigates the application of reinforcement learning (RL) to optimize low-thrust interplanetary trajectory design, focusing on the Earth-Venus transfer leg of the BepiColombo mission. Traditional trajectory optimization methods, such as patched conics and genetic algorithms, often require simplifying assumptions or complex optimization schemes. This work formulates the trajectory design problem as an optimal control problem (OCP) within a Markov Decision Process (MDP) framework, enabling an RL agent to learn efficient transfer strategies under realistic spacecraft constraints. The objective is to develop an autonomous guidance approach capable of replicating or improving upon established mission designs. The spacecraft is modeled as a point mass subject to heliocentric gravity, with propulsion constraints including limited thrust, variable throttle, and mass depletion due to fuel consumption. A Proximal Policy Optimization (PPO) algorithm, initialized behavior cloning from a genetic algorithm solution, is used to upgrade the agent. The observation space includes spacecraft state, target state, mass, and normalized time, while the action space consists of thrust magnitude and direction. A structured reward function guides the agent toward minimizing position and velocity errors, fuel usage, and trajectory inefficiencies. Preliminary results focus on validating the environment and comparing RL-generated trajectories against traditional methods in terms of delta-v and correction maneuvers. This work demonstrates the potential of RL to provide adaptive, fuel-efficient trajectory solutions and contributes to the advancement of autonomous spacecraft navigation and mission design.

Did this research project receive funding support (Spark, SURF, Research Abroad, Student Internal Grants, Collaborative, Climbing, or Ignite Grants) from the Office of Undergraduate Research?

No

Share

COinS
 

Interplanetary Trajectory Optimization with Reinforcement Learning

This project investigates the application of reinforcement learning (RL) to optimize low-thrust interplanetary trajectory design, focusing on the Earth-Venus transfer leg of the BepiColombo mission. Traditional trajectory optimization methods, such as patched conics and genetic algorithms, often require simplifying assumptions or complex optimization schemes. This work formulates the trajectory design problem as an optimal control problem (OCP) within a Markov Decision Process (MDP) framework, enabling an RL agent to learn efficient transfer strategies under realistic spacecraft constraints. The objective is to develop an autonomous guidance approach capable of replicating or improving upon established mission designs. The spacecraft is modeled as a point mass subject to heliocentric gravity, with propulsion constraints including limited thrust, variable throttle, and mass depletion due to fuel consumption. A Proximal Policy Optimization (PPO) algorithm, initialized behavior cloning from a genetic algorithm solution, is used to upgrade the agent. The observation space includes spacecraft state, target state, mass, and normalized time, while the action space consists of thrust magnitude and direction. A structured reward function guides the agent toward minimizing position and velocity errors, fuel usage, and trajectory inefficiencies. Preliminary results focus on validating the environment and comparing RL-generated trajectories against traditional methods in terms of delta-v and correction maneuvers. This work demonstrates the potential of RL to provide adaptive, fuel-efficient trajectory solutions and contributes to the advancement of autonomous spacecraft navigation and mission design.

 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.