CSCI 235 - Intelligent Robotics

Fall 2014

Course Overview:

In this course, we will study the application of concepts from artificial intelligence to solving problems in robotics. We will be using the Lego Mindstorms EV3 robot platform. Each kit will be supplemented with a USB webcam, which plugs into each robot's USB port.

We will program the robots using the leJOS EV3 implementation of the Java programming language. As knowledge of Java is not a prerequsitie for the course, we will spend some time early in the semester learning the essential elements of the language for our purposes.

Java implementations of AI algorithms will be provided for your use in building controllers for your robots. We will examine these implementations in class before employing them. Consequently, our robot programming activities in this course will consist largely of configuring and combining these provided implementations to produce a target behavior.

Class periods will be a mix of lecture, discussion, student presentations, and hands-on experimentation with the robots. Robots will be kept in the lockers in the Robotics Lab (MCReynolds 318) for use outside of class time. Students should bring laptop computers to class each day, to enable experimentation.

AI algorithms and topics we will explore include:

Subsumption
Computer vision
Artificial neural networks
Localization
Mapping
Planning

This course carries an Odyssey Special Projects credit. In the last four weeks of the semester, each student will complete a final project requiring at least 30 hours of work in order to earn this credit.

At the end of this course, you will be expected to be able to:

Build a multi-layered subsumption controller for a mobile robot.
Apply (and understand the limitations) of several computer vision algorithms for robot sensing.
Create subsumption behaviors by training an artificial neural network.
Localize a robot given various types of sensor input.
Construct and employ a map using landmarks.
Understand the application of planning to robot control.

Instructor:
Dr. Gabriel Ferrer
M.C. Reynolds 312

Office Hours:
MTWRF: 1:30-2:30 pm
Feel free to make an appointment (http://drferrer.youcanbook.me), or to stop by whenever my door is open.

Class Web Page: http://ozark.hendrix.edu/~ferrer/courses/235/

Lecture Time: B4 (2:45-4:00, TR)

Final Exam Period: Thursday, December 11, 8:30-11:30 am

Grading Criteria:
Assignments: 50%
Journal: 10%
Final Project: 35%
Class Participation: 5%

Grading Scale:
Each assignment receives a letter grade. The grading criteria for each assignment will be described when it is assigned. Each letter grade has associated with it a percentage grade as follows:

Letter grade	Percentage
A	95
B	85
C	75
D	65
F	50

Missing grades will be scored zero. Any grade can have a "+" or "-" attached to it. A "+" is worth +5, and a "-" is worth -4. A grade of "A+" will only be assigned to work that in some way goes above and beyond the requirements for the assignment. For each category above, the total points earned will be divided by the total points possible to yield a percentage. These percentages will be weighted as given above. A final average of at least 90 earns an A; 80 earns a B; 70 earns a C; 60 earns a D; below 60 is failing.

Assignments: Assignments will be given throughout the semester. In a typical assignment, students will program a robot to perform a task, give a presentation and demonstration of the results, and submit a written report. Most assignments will be completed in teams of two.

Journal: Each student should maintain an independent written journal. Observations of every experiment with your robot should be recorded therein. The journal should be a composition book, not a spiral-bound notebook or loose-leaf binder. Each entry in the journal should include the following:

Date of the work.
Time the work started and ended.
Goals for the session.
Brief descriptions of activities undertaken.
Observations from activities undertaken.
Assessment of the degree to which the goals were met.

Journals will be collected and checked periodically throughout the semester.

Final Project: In the last four weeks of the semester, each student will undertake a final project. In this final project, you will build and program a robot that fulfills a contextualized purpose. A public demonstration will be made of the robot's capabilities, and a paper reflecting upon lessons learned will be submitted as well. In keeping with the Odyssey Special Project guidelines, the project will require at least 30 hours of work.

Revisions: After assignments are graded, you are welcome to revise and resubmit your work. I will grade anew each submitted revision, and average the original and revised grades together to produce a new grade for that assignment. Revisions may be submitted anytime until the start of the final exam.

No late work will be accepted. Any work not submitted on time is a zero. However, you may submit a solution after the deadline to qualify under the revision policy above. In effect, this means that late work can earn up to half credit.

Accommodations: It is the policy of Hendrix College to accommodate students with disabilities, pursuant to federal and state law. Students should contact Julie Brown in the Office of Academic Success (505-2954; brownj@hendrix.edu) to begin the accommodation process. Any student seeking accommodation in relation to a recognized disability should inform the instructor at the beginning of the course.