Skip to content

Releases: mit-ll/spacegym-od2d

v0.1.0

09 May 18:52
Compare
Choose a tag to compare

What's Changed

Added

  • An AI agent against which a human can play. Run w/
conda activate od2d_v0.1.0
cd src/orbit_defender2d/king_of_the_hill/examples   # need to have this as working directory to work properly
python playerCLI_vs_AI.py
  • Ability to create asymmetric games. See koth.py

  • Game server that can run a game that two remote users log in to play. Game server can display it's own render of the game

  • koth.py

    • Ability to create asymmetric games. All game parameters can be asymmetric between players including number of tokens, fuel, ammo, points, win score, etc...
    • Randomizer that can be used to generate random Initial game conditions for use with AI training.
    • More print function options to display info as game is played
    • Logging functions to log game to file
  • orbit_defender2d/king_of_the_hill/examples/

    • Added player command line interface against trained AI agent example. This is played on a smaller board (4 rings).

Changed

  • Updated some required packages in the environment. Added torch to play trained AI agent.
  • Changed some default names in Utils.py. seeker displays as HVA and bludger displays as Patrol. Beta changed to Bravo.
  • 'koth.py'
    • Some changes to scoring, added fuel points baed on remaining fuel. Goal Sector points remain cumulative.
    • Made many actions that are not useful illegal (cannot attack already inactive tokens for example). This helps AI training immensely
  • pettingzoo_env.py
    • Updated the renderer to have larger, easier to see tokens with numbers. Made rendered window resizable

New Contributors

  • @mpj15 made their first contribution in #2

Full Changelog: v0.0.12...v0.1.0