Elizabethtown College
CS/EGR 434 Artificial Intelligence and Robotics
Syllabus
(Fall, 2009)

 

Professor:             Joseph T. Wunderlich, Ph.D.
Office:                    Esbenshade 284E
Phone:                   717-361-1295
Email:                    wunderjt@etown.edu
Web site:               http://users.etown.edu/w/wunderjt

Office Hours:        http://users.etown.edu/w/wunderjt/schedules/Schedule Card f09 joe w.htm

Calendar:              http://users.etown.edu/w/wunderjt/schedules/CALENDAR3_f09_web.htm



Catalog Description

Robotics and machine intelligence including symbolic Artificial Intelligence (AI) and artificial neural networks. Symbolic AI uses programmed heuristics and forms of knowledge representation. Artificial neural networks are connectionist computer architectures (hardware or software) where many computational nodes are connected to solve problems requiring rapid adaptation, or where governing equations are not known or cannot be easily computed. Course includes mobile-robot and robotic-arm theory, applications, simulations, real-time control, and path-planning strategies. *Prerequisites: Computer Science 121 and Math 121. Fall semester, odd-numbered years. Prof. Wunderlich.

 

Course Credit: Four 

 

Course Readings:

o    R. Siegwart and I. Nourbakhsh, Autonomous mobile robots, Massachusetts Institute of Technology, 2004. (ISBN: 026219502X)

o    Selected papers (handed out in lecture)

o    University of Trento Ph.D. course content from “Advanced Robotics with Applications to Space Exploration”

 

Grading:

o    Homework = 10%

o    Project #1 = 30%

o    Project #2 = 30%

o    Comprehensive final exam = 30%

COURSE GRADE:
        (60-62)=D-, (63-67)=D, (68-69)=D+, (70-72)=C-, (73-77)=C, (78-79)=C+, (80-82)=B-, (83-87)=B, (88-89)=B+, (90-92)=A-, (93-100)=A
        (with any fractional part rounded to the nearest integer)

 

Attendance:

Exams will primarily cover material presented in lecture -- some of which may not be found in readings.

 

Academic Honesty:

Elizabethtown College Pledge of Integrity: "Elizabethtown College is a community engaged in a living and learning experience, the foundation of which is mutual trust and respect. Therefore, we will strive to behave toward one another with respect for the rights of others, and we promise to represent as our work only that which is indeed our own, refraining from all forms of lying, plagiarizing, and cheating."

 

Disabilities

If you have a documented disability and need reasonable accommodations to fully participate in course activities or to meet course requirements, you must: (1) Contact the Director of Disability Services, Dr. Kristin Sagun, in the Center for Student Success, BSC room 228 by calling 361-1227; and (2) Meet with me within two weeks of receiving a copy of the accommodation letter from Disability Services to discuss your accommodation needs and their implementation.

 

Course Outline

 

          I.     Traditional symbolic AI vs. neural networks

(1)    The history of traditional “AI” (Symbolic artificial intelligence)

                                          i.    Predicate calculus

                                         ii.    Heuristics

                                        iii.    Knowledge representations

                                        iv.    A.I. programming languages

1.     Overview of Prolog and LISP

                                         v.    Expert systems

                                        vi.    Past Elizabethtown College student projects (since 1999)

(2)     “Bottom-up” biological and circuit brain models vs. “Top-Down” psychological and mathematical brain models

                                          i.    J. Wunderlich research examples (1990 to present)

(3)    Evolution of neural networks

                                          i.    Perceptrons

                                         ii.    Learning rules

                                        iii.    Underlying neural network mathematics

1.     Gradient descent learning

                                        iv.    Neural network applications

                                         v.    Neural network hardware

1.     Neorovector vector-register neural network chip (J. Wunderlich, 1992)

2.     Artificial Dendritic Tree  neural network chip (J. Wunderlich, 1993)

                                        vi.    Neural network software

                                       vii.    Past Elizabethtown College student projects (since 1999)

         II.    Merging neural networks with symbolic artificial intelligence

        III.    Robotics

(1)     Mobile robots

                                          i.    Path-planning and obstacle avoidance strategies

                                         ii.    Environmental mapping

                                        iii.    Simulation models

                                        iv.    Real-time sensor fusion

1.     Vision systems

2.     Laser range finders

3.     Ultrasonic sensors

4.     GPS navigation

5.     Digital compass

6.     Motor types and control

7.     Concurrent interactive simulations and real-time control

                                         v.    Computer hardware platforms

                                        vi.    Computer software languages and environments

                                       vii.    Past Elizabethtown College student projects (since 1999)

1.     Wunderbots 0, I, II, III, IV, and iV

                                      viii.    Space exploration

(2)    Robotic arms

                                          i.    Kinematics

                                         ii.    Controls schemes

1.     Position, velocity, or acceleration control

                                        iii.    Path-planning and obstacle avoidance

                                        iv.    Hyper-redundant manipulators in constrained spaces (J. Wunderlich research 1996 to 2004)

                                         v.    Past Elizabethtown College student projects

                                        vi.    Rehabilitation robotics

1.     J. Wunderlich research examples (1993 to 1995)

                                       vii.    Industrial robotic arm designs

1.     EU designs

2.     Japanese designs

3.     USA and other designs

4.     J. Wunderlich research (2003/04)

(3)    Research Laboratories

                                          i.    United States Laboratories

1.     MIT Laboratories

2.     Elizabethtown College Robotics and Machine Intelligence Lab (since 1999)

3.     A.I. Dupont Rehabilitation Robotics Lab (U. Penn, U. Del., Oxford, and Cambridge, 1990’s)

                                         ii.    International Laboratories

1.     University of Trento, italy

2.      and Italian Institute of Technology

3.     Other EU labs

4.     Japanese labs

        IV.    Artificial humanoids

(1)    Emulating human physiology (Biomechanics, senses, control systems, etc.)

         V.    Ethical Issues involving machine intelligence

(1)    Replacing humans vs. aiding humans

        VI.    More AI programming (time permitting)

 

 

TIPS ON PRESENTATIONS:

Your visuals:

Ø  Minimize unnecessary details

Ø  Have less than 30 words per slide.

Ø  Don’t have too many slides for allotted time.

Ø  Ensure good contrast between text and background on visuals (will the lights be on?)

Ø  A picture is worth a thousand words -- and an equation or graph can be worth much more.

Ø  Consider putting an image on every page (e.g., clip-art, photo, animation) which is an abstraction of the subject matter on the slide (i.e., invoke an idea!).

Ø  Don’t read from a script.

Ø  Don’t have too many (or annoying) sound effects.

Ø  Don’t speak monotonically.

Ø  Make eye contact with audience.

You and your overall game-plan

Ø  Minimize unnecessary details

Ø  Have a clear objective (e.g., to entertain, to sell, to motivate, or to report findings).

Ø  Have a good “opener” (e.g., an agenda, a quotation, a question, or a declaration)

Ø  Be organized and logical (e.g., present problem then solution; or have priorities – least-to-most or most-to-least).

Ø  Have the audience’s expectations understood (e.g., provide meaning and/or motivation).

Ø  Have good transitions between main points.

Ø  Have a good “closing” (i.e., summarizes main ideas, restates purpose of presentation).

Ø  Be flexible (i.e., to modified on the fly if questions are allowed during presentation)