Designing an Artificially Intelligent Software System

Think about things you do in life, things you'd like to do, things other people do or might like to do, and design an artificially intelligent software system to help people with some specific activities. Examples (that have already been solved, and are therefore unsuitable for this project) include route planning for driving in an unfamiliar city ("Get Directions" on Google maps), the blocks world (the SHRDLU resurrection), playing chess/checkers/etc. (try playing the unbeatable checkers AI system), etc. There are many more challenges for AI, and you it's not hard to dream of one.

While it is not expected that you will be able to implement the AI system you design, you should be able to conceive of such a system. There is no need to worry about the technical details. Rather, you should think about, and describe for your project ...

the foundations of AI systems - why is search required to solve the problem;
the problem and its environment - what features do they have;
knowledge that would help guide search - what are the features of the knowledge, and how your system might acquire the knowledge;
a production system algorithm that is appropriate for your system, including how your system would choose which states and transformations to apply at each step.

A key point is to ensure that your software system will really require intelligence, i.e., that there is a search space that you need to explore to find a solution to the problem ... if there's no search then no intelligence is required.

You may work alone or in a group of two. Produce your design using a word processor, using single spacing and a 12pt Times Roman font. You can include pictures, diagrams, whatever you need to make the design clear and complete. The end product will probably be several pages long, but may not be be more than 5 sides of A4 paper. Submit your design as a PDF document. Good designs could even form the basis of MS thesis research!

The due date is 19th July 2010.

It is worth 10% of the course assessment. The marks are distributed as follows:

Description of the problem and its environment - 2%
Description of the search space - 2%
Description of the knowledge to be acquired and used - 2%
Production system algorithm, including expansion of the function that chooses the states and transformations to apply - 2%
Originality and presentation (language, terminology, formatting, etc.) - 2%

Please review the policies on assessment in the administration document.