Bust-The-Ghost

Description

In this project, we used Unity to implement/reproduce "Bust the Ghost" game.

• Create a 8x 20 grid
  
• Have the ghost be placed in one of the cells according to a prior distribution of Ghost over location P(Ghost).
  
  • Use a uniform distribution to start with.
    
• when clicking a cell, the player gets a color red/green/orange/yellow depending on how the far is the ghost is from the clicked cell.
  

• On the ghost: red
  
• 1 or 2 cells away: orange
  
• 3 or 4 cells away: yellow
  
• 5+ cells away: green
  

• For this, define and use a conditional probability distribution P(Colour/Distance from Ghost).
  

  • Use this probability to decide on the color to display.
    

• After each click "t", the Posterior Probability of the Ghost P(Ghost/ Colour) should be updated and displayed on the cells using Bayesian inference:
  

  • P(Ghost_t)=P(Ghost/Color_t)=P(Ghost_t-1)*P(Colour/Distance from Ghost).
    
  • P(Ghost_0)= P(Ghost/Color_0)= P(Ghost) the prior probability.
    

• Do not forget to Normalize!
  

• User can decide to "bust" a cell if ghost is in the cell; the player wins otherwise he/she looses.