Repolode
Submit
AVSLab

AVSLab/bsk_rl

Python package for building Gymnasium reinforcement-learning environments for spacecraft tasking and planning, built on the Basilisk simulator.

View on GitHubWebsite
Stars
125
Forks
20
Open issues
30
Last commit
3 days ago

What is it?

BSK-RL is a Python package for constructing Gymnasium environments for spacecraft tasking problems, such as planning and scheduling which observations a satellite should collect. It sits on top of Basilisk, a modular spacecraft simulation framework, so the environments run with high-fidelity dynamics and reasonable computational cost.

The package targets spacecraft autonomy research: researchers define tasking problems as standard Gymnasium environments and train or evaluate agents against them. Installation, examples and API documentation are hosted on the project website.

The project is developed by the Autonomous Vehicle Systems (AVS) Lab at the University of Colorado Boulder.

Key features

  • 01Builds Gymnasium-compatible environments for spacecraft tasking problems.
  • 02Runs on the Basilisk spacecraft simulation framework for high-fidelity dynamics.
  • 03Focuses on satellite planning and scheduling, including earth-observation tasking.
  • 04Supports reinforcement-learning workflows for spacecraft autonomy research.

Use cases

  • Training reinforcement-learning agents to schedule satellite observations.
  • Benchmarking autonomous tasking algorithms in a realistic orbital simulation.
  • Studying onboard decision-making for earth-observation constellations.

Quick start

  1. 1

    Open the BSK-RL website and follow the installation instructions

Things to know

  • The package is built on the Basilisk simulator, so Basilisk must be set up as well.

In collections

Reinforcement learning frameworks & training environmentsRobotics & embodied AI projects

Topics

#reinforcement-learning#spacecraft#satellite-tasking#simulation#gymnasium#earth-observation#autonomy

FAQ

What is bsk_rl?
Python package for building Gymnasium reinforcement-learning environments for spacecraft tasking and planning, built on the Basilisk simulator. BSK-RL is a Python package for constructing Gymnasium environments for spacecraft tasking problems, such as planning and scheduling which observations a satellite should collect. It sits on top of Basilisk, a modular spacecraft simulation framework, so the environments run with high-fidelity dynamics and reasonable computational cost.
Is bsk_rl free to use? What license does it have?
bsk_rl is open source under the MIT license. Check the license terms before using it in commercial products.
Is bsk_rl still maintained?
Actively maintained. The last commit to bsk_rl was on Nov 4, 2026.
What can I use bsk_rl for?
Training reinforcement-learning agents to schedule satellite observations. Benchmarking autonomous tasking algorithms in a realistic orbital simulation. Studying onboard decision-making for earth-observation constellations.
How do I install bsk_rl?
Open the BSK-RL website and follow the installation instructions
Where can I get bsk_rl?
The source code and installation instructions are on GitHub: https://github.com/AVSLab/bsk_rl

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects