Skip to content

OpenAI Gym environments for spacecraft operations problems.

License

Notifications You must be signed in to change notification settings

atharris/basilisk_env

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

33 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

basilisk_env: A Basilisk-based Astrodynamics Reinforcement Learning Problems Package

basilisk_env is a package of reference planning and decision-making problems built on both the AVS Basilisk simulation engine (https://hanspeterschaub.info/bskMain.html) and using the OpenAI gym API (https://gym.openai.com/)

Getting Started

These instructions will give you the information needed to install and use gym_orbit.

Prerequsites

gym_orbit requires the following Python packages:

  • Numerics and plotting: numpy, scipy, matplotlib, pandas
  • Spacecraft Simulation: Basilisk
  • Machine Learning + Agent Libraries: tensorflow, keras, gym, stable-baselines

Installation

To install the package, run:

pip install -e .

while in the base directory. Test the installation by opening a Python terminal and calling

import gym
import basilisk_env

Optical Navigation scenarios are run with the Basilisk Vizard visualization suite. This can be downloaded from (http://hanspeterschaub.info/basilisk/Vizard/VizardDownload.html).

To use Vizard with basilisk-env, you must set up environment variables corresponding to the vizard application directory and address/port ('viz_app','viz_address', and 'viz_port', respectively). For example:

$ export viz_app = '/mnt/c/Users/aharr/OneDrive/dev/Vizard/vizard20201211L/Vizard.exe'
$ export viz_port = '5556'
$ export viz_address = '172.20.128.1'

Repository Structure

basilisk_env is broken into two principal components: /envs, which contains environment wrappers for Basilisk, and simulators, which contains Basilisk simulation definitions for various scenarios.

Within simulators, core simulation routines representing interchangable dynamics and flight software stacks are stored in simulators/dynamics and simulators/fsw respectively.

Contributions

Authors

Maintainer: Mr. Andrew T. Harris ([email protected])

Mr. Thibaud Teil ([email protected])

License

This project is licensed under the MIT license - see the LICENSE.md file for details.

Acknowledgements

The Autonomous Vehicle Systems Laboratory

About

OpenAI Gym environments for spacecraft operations problems.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages