Skip to content

linan1109/SpaceCadetPinball-AIplayer

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

19 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

SpaceCadetPinball-AI

An AI player that can learn from the game and reach a high score, using or will use Computer Vision, Reinforcement Learning, and Genetic Algorithm.

  • Vision: Recognize the pinball's coordinates and speed vector, the score, and the states of every frame during the play.
  • Reinforcement Learning: During one game, train the AI to play the higher score using information gathered from the Vision part.
  • Genetic Algorithm: Choose the best players after each generation (game round), and use crossover (average parameters) and mutation (add random noise) to produce new generation.

To do list

  • Basic game control
  • Basic game information extraction using cv2
  • genetic algorithm design and implementation
  • [] genetic algorithm training
  • [] game control using trained model
  • [] reinforcement learning design and implementation
  • [] reinforcement learning training

Demo

Game information extraction

limitation for now

Because the window location is based on window relation, so it can not use with full screen mode.

About

An Reinforcement Learning based Space Cadet Pinball Player

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages