Android Open Source - TicTacToe3D Artificial Inteligence

• Back to Game/tictactoe  ↑

From Project

Back to project page TicTacToe3D.

License

The source code is released under:

MIT License

Java Source Code

package com.unibh.dcet.ccc.android.tictactoe3d.ai;
/* ww  w.j  av  a 2 s.  c  om*/
import java.util.List;

public class ArtificialInteligence {

  //private Judge judge;
  Environment environment;
  private int aiPiece; 
  private int opponentPiece;
  private int dimension;
  private Dot bestMoveOpponet = null;
  private int round = 0; 
  private int roundBestOpponetMove = -1; 
  

  public ArtificialInteligence(int dim, int piece) {

    environment = new Environment(dim, piece);
    this.dimension = dim;
    this.aiPiece = piece; 
    this.opponentPiece = (piece == Environment.TTT_O ? 
        Environment.TTT_X : Environment.TTT_O );
    //this.judge = new Judge();
  }

  public int getDimension() {
    return dimension;
  }

  public Environment getEnvironment() {
    return environment;
  }

  public int getPiece() {
    return aiPiece;
  }

  public boolean isValidPosition(Dot dot) {
    if (environment.getPiece(dot) == Environment.TTT__) {
      return true;
    }
    return false;
  }  

  private Dot getBestMove(int piece) {
    int opponent = (piece == Environment.TTT_O ? 
        Environment.TTT_X : Environment.TTT_O );
    // vit?ria eminente do jogador
    Dot move = this.environment.eminentVictory(piece);
    if (move != null)
      return move;
    // vit?ria eminente do oponente 
    Dot opponetVictory = this.environment.eminentVictory(opponent);
    if (opponetVictory != null)
      return opponetVictory;
    // melhor jogada do oponente 
    int numEminentVictory = 0;
    Dot bestMoveOpponet = this.environment.maxChanceVictory(opponent);
    if (bestMoveOpponet != null) {
      for (Line l: ((DotAI)bestMoveOpponet).getPossibleLines()) {
        if (l.getHasPiece() == opponent && 
            l.getMarkedPosition() >= (dimension - 2))
          numEminentVictory++;
      }
      if (numEminentVictory >= 2)
        return bestMoveOpponet;
    }
    Dot resp = this.environment.maxChanceVictory(piece);
    if (resp == null)
      return aiDummyMove();
    return resp;
  }
  
  public Dot aiMove() {
    return getBestMove(aiPiece);
  }
  
  public Dot getBestMoveOpponet() {
    if (round != roundBestOpponetMove) {      
      bestMoveOpponet = this.getBestMove(opponentPiece);
      roundBestOpponetMove = round;
    }
    return bestMoveOpponet;
  }

  public DotAI aiDummyMove() {
    DotAI dot;
    this.environment.maxChanceVictory(aiPiece);
    for (int plane = 0; plane < dimension; plane++) 
      for (int lin = 0; lin < dimension; lin++) 
        for (int col = 0; col < dimension; col++) {
          dot = environment.getDotAI(plane, lin, col);
          if (dot.getPiece() == Environment.TTT__) {
            markAction(dot, aiPiece);
            return dot;
          }
        }
    return null;
  }

  public boolean markAction(Dot dot, int playerPiece) {
    if (isValidPosition(dot)) {
      round++;
      this.environment.markPosition(dot, playerPiece);
      return true;
    }
    return false;
  }

  public Line getVictoryLine(Dot lastChoosePosition) {
    /*
     * nao e necessario varrer toda a matriz para verificar se existe um vencedor, sera feita a analise de uma vitoria a partir do ultimo ponto
     * marcado no tabuleiro, se nenhuma das possiveis vitorias a partir do ultimo ponto marcado foi alcancada retorna null
     */
    return this.environment.hasVictory(lastChoosePosition);
  }
    
    public Dot chooseBestAction() {
      
    return null;
  }

  public int countVictoryPossibility(Dot dot) {

    return 0;
  }

  public List<Dot> findAdjacentsDots(Dot dot, int eixo) {

    return null;
  }
  
}

Java Source Code List

• Back to Game/tictactoe  ↑