Team 6 - McKane Andrus, Kevin Fan, Jie Ma, Vincent Politzer
Introduction
For our project, we created and implemented a game to be played by Zumys. We call it: Zumy Patrol. The game is played on a grid with obstacles forming a maze. The “player” zumy is placed at a starting point and must find its way to some predesignated ending point, while avoiding obstacles. To complicate the game, one or more additional “patrol” Zumys are added to random starting positions on the grid. These Zumys are each outfitted with a camera and try to “catch” the player zumy before it can get to the end. The game ends when the player Zumy successfully makes it to the ending point, or if a patrol Zumy sees the player Zumy.
This project has some sentimental value, as the game we constructed mimics games that many 90’s kids used to play. Watching artificial intelligence play a game can also be very entertaining, much like watching a sport, because the observer does not know what the outcome will be. The suspense is a fun feeling. From a technical perspective, the project explores dynamic pathfinding (the player Zumy has to constantly update its chosen path to respond to the movement of the patrols), as well as exploration (the patrol Zumys are not given knowledge of the whole map, but rather learn the layout bit by bit as they try to find the player).