Introduction to AI Robotics

Fall 20067
 CAP 4660 (undergrad) and
CIS 6930 (grad)
TTh 2:00-3:15p
Location: ENB 110

Using Robots Seen on CNN


Instructor: Prof. Robin R. Murphy
Office hours: TTh 12:00-1:30 and by email appointment

If you signed up for the APEX version and it is cancelled, sign up for the “regular” class as it will be podcast and all assignments can be done on-line and robot work scheduled.
 

ALL EMAIL CORRESPONDENCE MUST HAVE "CAP4660" or "CIS6930" IN THE SUBJECT LINE (ALL CAPITALS, NO SPACES) or be sent through the my.usf.edu email. OTHERWISE IT MAY NOT BE READ.

All materials and course announcements will be made through the course blackboard space at http://my.usf.edu.

This web page was updated 8/22/07

Course Objectives

This course is an introduction and survey of artificial intelligence methods for mobile robots (ground, aerial, or underwater) for science and engineering majors. It covers both the theory and the practice of mobile robots, focusing on case studies drawn from the instructor's experience in academia, work with industry, and service with the Department of Defense conducting program reviews. A bit of discussion about how the theory compares to science fiction is thrown in as well. It is a good first course in mobile robots and a great way to learn what robots can really do versus the hype in the movies. It is required for students seeking funding from or access to the laboratories at the Center for Robot-Assisted Search and Rescue (now part of the Institute for Safety Security Rescue Technology).

Programming. The undergraduate section does not require either a previous course in AI (though it will be helpful), but students taking the CS option should have had at least one object-oriented programming class to follow the discussion and conduct programming in Java or C. (The majority of programming will be modifying existing code, so don't panic.)

The graduate section shares the same lectures and tests, but diverges in the project.

Workload. The course is generally an A-B course with some C's. I have failed people, but usually most drop before that point. About 30% of the class drops the course due to the workload, we cover a large amount of theoretical material and do a lot of programming robots. In general, it can be considered an extreme class, one where you work very hard but have lots of fun!

Both sections rely on hands-on learning to amplify the course topics.

Both sections require a great deal of reading and attention to definitions and nomenclature. Tests are derived from the textbook and class lectures.

Specific Objectives. By the end of the Fall Semester, the student should be able to:

Attendance Policy

The primary method of communication will be through podcast lectures, with flexibly scheduled direct interactions with the instructor. I’ll be setting up times with you to discuss design- how to take what you’re learning and apply it to real systems- this might be during the official class period or may be set up on a day when we can run a bit longer.  You will be responsible for material presented in class and changes in assignments. Also keep in mind that I'm doing a lot of supplemental lectures that aren't in the textbook-- I'm coming out with a second edition. This means that "Oh, I can just read the book" is an out-dated and dangerous assumption.

Grading Policy


 

Deliverable

number

% total grade

On-line Quizzes

20-30

40

On-line Tests

3

30

Project

1

30


 

Final grades will be a largely plus system:

For each set of podcast lectures, there will be an on-line quiz to make sure you saw the podcast and read the material.

 

The tests will be on-line, though they will require you to submit files. You will need some sort of drawing package (.ppt, visio, omnigraffle, a scanner if you use pen and paper, whatever you should get the point) to create figures to upload.

Undergraduate projects fall into 2 categories:

1.     “No Fear” option- you will mentor local middle schools on the FIRST Robotics Lego League competition. This involves some Saturdays. You’ll be assigned a middle school and you may have a partner from class depending on the numbers. USF may (still working on this) loan you a Lego Mindstorms NXT for the semester, but you may to wish to purchase one yourself because they are really cool. (The new NXT, not the regular Mindstorms). You will have to produce materials to help teachers and help USF COE build a website of helpful resources.

2.     Programming option- this is really fun for people who really want to program. You will be programming first in a game-engine based simulator and then on the real robot-- a RWI ATRV-Jr. The purpose is to create an affect teammate- a robot that can work next to a person. Each person will have a module to write.

Graduate students will work on littoral and bridge inspection with the Sea-RAI unmanned surface vehicle, first in simulation and then the real thing. The projects will be tailored to the student’s background and research objectives. Some aspects may require a lot of programming, some less, others hardware experience.

Academic dishonesty

All aspects of the USF and CSE policies on academic dishonesty will be strictly enforced. Cheating on a test, pop test, or laboratory will result in immediate removal from the class roll and an "FF" for the course.  I have failed students for cheating.

Dates of scheduled exams, course outline including assignments and dates due

Click here for the wildly tentative syllabus and tests. Remember that my.usf.edu will be the "real" online source.

Notice of non-permission to sell notes or tapes of class lectures

All lecture and laboratory material is copyrighted. You may make notes and tapes of the lectures for your own use only.  

Required textbook or materials

The required textbook is Introduction to AI Robotics by Murphy, MIT Press. This book is under revision.

Policy on Missed Work

If you miss work: Material missed due to recognized absences (illness with doctor's excuse, death in the family) can be made up for full credit. Late material is accepted solely at the discretion of the instructor, assuming that at least 1 class period's prior notice was given.  

Reminder to student who anticipate being absent from class due to religious observances

Student who anticipate the necessity of being absent from class due to the observation of a major religious observance must provide notice of the date(s) to the instructor, in writing, by the second class meeting.