harman097 / RoboRugby

Training environment for ML agents in a competitive robot rugby game using OpenAI's gym standard.
MIT License
4 stars 0 forks source link

GameEnv() reset/initialize based on game state #28

Closed harman097 closed 4 years ago

harman097 commented 4 years ago

GameEnv needs the ability to reset the game to the starting state.

As part of this, it would probably make sense to allow the GameEnv to be initialized using a previous game state so we can re-run the exact same scenarios, if needed.

This would also allow us to create a "standard game state", if needed (based on the photo in the instructions).

This could also potentially tie into the "game state" that we pass back to the robot in the step function.