Make the following enhancements (copy and modify the code, you do not need to re-write from scratch).
- Larger rooms (rectangular) [ very easy ] - limit your rooms size to maximum 40 by 40 cells.
- Enhanced mobility (move in 4 directions - up, down, left, right)
- More/better sensors - detect amount of dirt (integer) under vacuum and whether any of the cells you can move to is blocked (4 cells).
- sense the amount of dirt in the cell you are in (correlated to units of dirt)
- a "view" of the 4 squares next to you that can be reached in one move (up, down, left, right). Sensors can return simply "blocked" or "not-blocked", in each direction.
- Model and avoid furniture and non-moving cats
- Maintain an internal "map" of the area and where you have cleaned, to be able to report the time it takes you to clean every available square at least once. (goal-based agent)
- Read in scenario files for different sized rooms, dirt factors, furniture locations, and cats. You may assume that the files contain only numbers.
- Cats that move (about 25% of the time, sleep 75%) one cell in any of 4 directions (5 points) (can jump on furniture - should not hit walls, optional if you want them to avoid hopping in the vacuum's square)
- Cats that move in any direction up to 2 squares in each direction (10 points)
- Cats that move and shed hair (dirt) each time they leave a square (20 points)
- Do robust input/error checking (20 points)
- Battery life, recharging the battery, knowing when to return for a charge (25 points)
Follow the link below for file input format description and examples.
Due Date: Fri (midnight), 6th week of class. [email to nreed@hawaii.edu]
Turn in your code, transcripts of testing and running it, and a report on your solutions.
Back to the course homepage.
(c) N. E. Reed, 2005-2009