Skip to content

sijiaxu/Adaptive_Game_AI_Demo

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

24 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Adaptive_Game_AI_Demo

Demo is based on monster hunter's concept, and made with unity3D.

Adaptive AI is used to improve the npc's intelligence, mainly focusing on dynamically deciding the most rensonable action while interacting with the player.

Demo's feature include:

  1. use reinforcement learning to improve the npc's AI(Q-learning and UCB1), now npc's action is adapting to player's behavious(in demo it refers to change dragon's action execution condition).
  2. learning procedure converge fast and cost just a little computation, so it can execute efficiently and get the effective result in time.
  3. Adaptive AI is developing based on the traditional FSM framework. so it can quickly apply to the exsiting games and scale up easily.

Entire unity project is in the UnityPackage folder and adaptive AI code is mainly in the dragonAI.cs,dragonBattleAI.cs and KnowledgeSystem.cs files.

Under the Demo_Gmae folder is the pre-builed windows version game.

Download and have fun!

if you have any suggestions or questions, feel free to contact me. :)