F UZZY L OGIC : G AME E XAMPLE Steve Foster. BDI M ODEL The BDI model studies intention and its relation to other mental attitudes As its name implies,

Slides:



Advertisements
Similar presentations
What is expectancy theory?
Advertisements

Intermediate 1 – Methods of Practicing Skills. Skill is the learned ability to bring about predetermined results with maximum certainty, with the minimal.
Curriculum Development and Course Design
SAMPLING METHODS OR TECHNIQUES
ARCHITECTURES FOR ARTIFICIAL INTELLIGENCE SYSTEMS
DECISION THEORIES 1 Problem solving –Collaboration, GAME THEORY –Asymmetric information, AGENCY THEORY –Optimization, OPERATIONAL RESEARCH 2 Problem finding.
Formal Semantics for an Abstract Agent Programming Language K.V. Hindriks, Ch. Mayer et al. Lecture Notes In Computer Science, Vol. 1365, 1997
Tetris – Genetic Algorithm Presented by, Jeethan & Jun.
Chapter-4 Staffing.
AFC COACH EDUCATION FUTSAL SYSTEMS AND STYLES. AFC COACH EDUCATION DEFINITION.
NPC Interaction in Video Games Robert Higginbotham.
Fuzzy Logic Steve Foster.
Artificial Intelligence in Game Design Intelligent Decision Making and Decision Trees.
What’s your Policy? Steve Darn Yaşar University
Interaction and adaptation in SCORM-based SE course Todorka Glushkova, University of Plovdiv, Bulgaria
Pac-Man® Shant K. Karakashian CSE 476/876 Optional Course Project 1.
Black & White To analyze and critique the AI used Amanda Yaklin.
Artificial Intelligence and Lisp Lecture 4 LiU Course TDDC65 Autumn Semester, 2010
Incorporating Advice into Agents that Learn from Reinforcement Presented by Alp Sardağ.
Customizable Audio Kaleidoscope Agustya Mehta, Dennis Ramdass, Tony Hwang Final Project Spring 2007.
Intelligent Pac-Man Ghost AI
Design of SCS Architecture, Control and Fault Handling.
Calcul mental. Calcul n°1 2 x 7 = Calcul n°2 3x 7 =
Structures & Strategies – week 7. Question Describe a situation where your chosen structure or strategy (Fast Break – Option A, B or C) was not effective.
Norm Theory and Descriptive Translation Studies
Teachers Name : Suman Sarker Telecommunication Technology Subject Name : Computer Controller System & Robotics Subject Code : 6872 Semester :7th Department.
Marcus Gallagher and Mark Ledwich School of Information Technology and Electrical Engineering University of Queensland, Australia Sumaira Saeed Evolving.
START Help Pac-Man find his dinner by answering math questions correctly. Try to go as fast as you can. You can do it! Click the Pac Man to begin playing.
INTRODUCTION TO ARTIFICIAL INTELLIGENCE Massimo Poesio Intelligent agents.
Higher Physical Education Volleyball.  Strategy  Methods of Analysis  Non Specialised Roles (Serving and Receiving)  Strengths of this Strategy 
Artificial Intelligence in Game Design Problems and Goals.
CS 4730 Action vs. Interaction CS 4730 – Computer Game Design Credit: Several slides from Walker White (Cornell)
Artificial Intelligence in Game Design Lecture 6: Fuzzy Logic and Fuzzy State Machines.
Mobile Robot Navigation Using Fuzzy logic Controller
Ming-Feng Yeh Fuzzy Control The primary goal of control engineering is to distill and apply knowledge about how to control a process so that the.
Artificial Intelligence for Games Lecture 5 1 Minor Games Programming.
Structures, Strategies and Compositions General Data Gathering Attacking Strategy – Fast Break.
You Are What You Do In Search of the Good, chapter 2.
Enhancing multiple intelligences in story reading.
2005MEE Software Engineering Lecture 11 – Optimisation Techniques.
AP STATISTICS LESSON AP STATISTICS LESSON DESIGNING DATA.
An Axiomatic Basis for Computer Programming Robert Stewart.
 The word ‘strategy’ is derived from a Greek word ‘ strategos’, which means generalship----the actual direction of military force  Strategy is a plan.
1 Motion Fuzzy Controller Structure(1/7) In this part, we start design the fuzzy logic controller aimed at producing the velocities of the robot right.
Dr Nick Mitchell (Room CM 224)
Artificial Intelligence in Game Design Influence Maps and Decision Making.
Will Britt and Bryan Silinski
Learning goals: w Understand the use of concepts w Identify and understand methods of problem solving w Identify barriers to effective problem- solving.
1 Lecture 4 The Fuzzy Controller design. 2 By a fuzzy logic controller (FLC) we mean a control law that is described by a knowledge-based system consisting.
1 ARTIFICIAL INTELLIGENCE Gilles BÉZARD Version 3.16.
Intelligent Agents Chapter 2. How do you design an intelligent agent? Definition: An intelligent agent perceives its environment via sensors and acts.
Human features are those things created by man.
By Pouryya Hamidi. jump-the-curve-semiotics-and-culture.
GCSE Physical Education Unit: School, Sport & Participation Topic: National Healthy Schools Learning Objective To be able to describe what/who is involved.
Business System Development
CS 134 Video Game “AI”.
CSE 476/876 Optional Course Project
Taumer Anabtawi, Katie Finley, Jeff Huggins
Artificial Intelligence in Game Design
STRATEGIC MANAGEMENT AND BUSINESS POLICY THIRD EDITION AZHAR KAZMI Chapter 1 Introduction to strategic management and business policy.
Intelligent Agents Chapter 2.
Expectimax Lirong Xia. Expectimax Lirong Xia Project 2 MAX player: Pacman Question 1-3: Multiple MIN players: ghosts Extend classical minimax search.
Fuzzy System Structure
نیازهای آموزشی بیماران و طبقه بندنی این نیازها کارگاه مدیریت آموزش بیمار فاطمه حشمتی نبوی دکتری پرستاری.
Safe Path Planning for an Autonomous Agent in a Hostile Environment
An Agent that plays Pacman
Belief Desire Intention
Expectimax Lirong Xia.
Think what name would describe or market the game best
Presentation transcript:

F UZZY L OGIC : G AME E XAMPLE Steve Foster

BDI M ODEL The BDI model studies intention and its relation to other mental attitudes As its name implies, BDI features three major mental attitudes as its building blocks - belief, desire, and intention Belief is the agent’s knowledge about its environment and itself Desire describes the agent’s goal: a system state that the agent wants to achieve Intention is the course of action that the agent has chosen to achieve that goal

F UZZY P AC MAN The “beliefs” or knowledge of the intelligent ghost agents includes the location of other game entities and the performance of the player In this simple game, the primary “desire” or goal of the ghosts is to intercept the player Lastly, the “intentions” or courses of action of the ghosts include hunting Pac-Man, guarding pellets, and avoiding other ghosts.

G HOST B EHAVIOURS In order for the ghosts to act more intelligently, four different behaviours for the ghosts could be implemented Each individual ghost chooses his behaviour based upon a fuzzy logic model The first behaviour is the “hunting” approach. In this behaviour, the ghost could actively seek Pac- Man using the A* path finding algorithm

G HOST B EHAVIOURS The second behaviour is the “defence” approach. For this behaviour, the ghost will defend the area that has the most pellets, thereby ensuring that Pac-Man must come near the ghost to complete a level. The third behaviour is the “shy ghost” approach. In this behaviour, a ghost will move in a direction that will take it away from a nearby ghost. This behaviour ensures that ghosts will spread out and cover the entire level Lastly, the “random” approach is chosen when no other method is a preferred choice. In this approach, a ghost will randomly move about the level.

B EHAVIOUR C HOICES Each of these behaviours has a set of fuzzy rules that contribute to the likelihood of choosing that behaviour The inputs to the rules are a number of linguistic terms that are derived from player performance and game conditions Example: If pacman_near AND skill_good, Then hunting_behavior

S KILL Defined in terms of good skill, medium skill, and bad skill The three skill levels are actually formed from another set of fuzzy rules These intermediate rules must be calculated before the ghosts’ behaviours can be calculated The following fuzzy rules are used to define the player skill variables: If time_life_short OR pellet_rate_bad, Then skill_bad If time_life_medium OR pellet_rate_medium, Then skill_medium If time_life_long AND pellet_rate_good, Then skill_good