1. Lecture 1 CMPS 146, Winter 2012

2. UC Santa Cruz Course mechanics  The course is fully subscribed  Text Book: Artificial Intelligence for Games (2 nd edition) Ian Millington and John Funge  Syllabus review  Piazza for class discussion  DRC accommodations

3. CMPS 146, Winter 2012UC Santa Cruz Upcoming talks  Inventing The Future of Games Lectures Series  Wednesdays, 11:00 – 12:00, Simularium  It’s a two unit seminar, but you can drop in as well  First speaker : Cecil Brown, UC Berkeley, Jan. 18  Richard Hilleman  VP, Creative Director, EA  Thursday, January 19, 6:30 – 8:00pm, DARC 108  Game Telemetry and Crowd Sourcing New Game Services: How You Take Over the Game Industry  RSVP: https://games.soe.ucsc.edu/rich-hilleman-talk https://games.soe.ucsc.edu/rich-hilleman-talk

4. CMPS 146, Winter 2012UC Santa Cruz Dialog systems study

5. CMPS 146, Winter 2012UC Santa Cruz AI and Perception

6. CMPS 146, Winter 2012UC Santa Cruz What is game AI?  Tasks  Pathfinding  Bot behavior (chasing, guarding…)  Reputation  Strategic reasoning  Group behavior  Learning  Etc.  Techniques  A*  Rule systems  Neural networks  Artificial evolution  Planning  Finite state machines  Blobs of random code  Etc.

7. CMPS 146, Winter 2012UC Santa Cruz Game AI lies in perception GamePlayer Goals Beliefs Emotions

8. CMPS 146, Winter 2012UC Santa Cruz Behavior = AI + physics Behavior Code PhysicsAI

9. CMPS 146, Winter 2012UC Santa Cruz “AI” is not the same as visual representation Visual depiction of agents (ghosts) Can be read intentionally Visual depiction of agents (invaders) Can’t be read intentionally (only physics)

10. CMPS 146, Winter 2012UC Santa Cruz “AI” is not the same as complexity

11. CMPS 146, Winter 2012UC Santa Cruz Game AI: programming for interpretation GamePlayer Goals Beliefs Emotions Designer/programmer Goals Beliefs Emotions

12. CMPS 146, Winter 2012UC Santa Cruz AI is a medium Edward Ihnatowicz The Senster, 1970 Black and White, 2000 Peter Molyneux (Lionhead) Harold Cohen Aaron, 1971 -

13. CMPS 146, Winter 2012UC Santa Cruz Pac Man AI From: http://gameinternals.com/post/2072558330/understanding-pac-man-ghost-behavior

14. CMPS 146, Winter 2012UC Santa Cruz Movement modes (waves)  Three movement modes  Chase  Scatter (target squares outside corners)  Frightened (random choice at intersections)  First level phase durations  Scatter for 7, chase for 20  Scatter for 5, chase for 20  Scatter for 5, switch to chase permanently

15. CMPS 146, Winter 2012UC Santa Cruz Turn decisions given a target  Ghosts not allowed to reverse directions  Ghosts forced to reverse direction on change from chase to scatter

16. CMPS 146, Winter 2012UC Santa Cruz Scatter mode

17. CMPS 146, Winter 2012UC Santa Cruz Personalities

18. CMPS 146, Winter 2012UC Santa Cruz Red Ghost  Depending on pellets remaining  Speed increases by 5% at fixed points  Scatter target changes to chase target

19. CMPS 146, Winter 2012UC Santa Cruz Pink Ghost

20. CMPS 146, Winter 2012UC Santa Cruz Blue Ghost

21. CMPS 146, Winter 2012UC Santa Cruz Orange Ghost

22. CMPS 146, Winter 2012UC Santa Cruz The AI Model Strategy Decision Making Movement Character AI Group AI Execution Management Content Creation Scripting World Interface Animation Physics