COSC 4560/5560

MODERN ROBOTS AND SOFTBOTS

Spring 2006

Time:TR 2:45-4:00pm and Location: Engineering Bldg, Room 1055

Instructor: Prof. D. Spears	Phone/Office: 766-5485 / 4087 ENG
Office Hours: TBD	Email:dspears arobase cs.uwyo.edu

Textbook: Artificial Intelligence: A Modern Approach

Author: Stuart Russell and Peter Norvig

Publisher: Prentice-Hall

ISBN: 0-13-790395-2

Course Description:

This is a combined undergraduate and graduate course on modern approaches to designing robots and softbots (i.e., the software equivalent of robots). Collectively, robots and softbots are commonly called "bots" or "agents." This course focuses on how agents reason, plan, and act in real-world situations. The syllabus begins with three popular agent paradigms, and then discusses basic representational issues for bots. Planning and control are then covered, followed by important real-world issues of how to deal with environmental uncertainty. The last part of the course covers specifics of software versus hardware (robotic) agents, and concludes with some recently popular applications, plus the topic of "safe agents."

Prerequisites:

Students are expected to know elementary logic and probability. If you lack these prerequisites, then you can still attend the class and do well, but please talk with the instructor on how to catch up.

Tentative Schedule:

This course will be taught in a seminar style. Part of the course will be lectures, and part will be discussions about papers. Many of these papers represent modern, cutting-edge agent technology. Students will take turns leading discussions on the papers. The choice of lecture material and the choice of papers to discuss will be adaptive, based on the needs and interests of the students taking the course. Relevant chapters of the textbook to read for each topic are mentioned after the topic title in the syllabus. Details of the mathematical parts of the lectures can be omitted if desired -- the intuitions behind the math will be provided by the instructor. For students who are more math-oriented, the course will allow many opportunities for math. Students who are not math-oriented will be able to focus more on the algorithms and applications, and less on the math.

GUEST LECTURES:

Guest lecture by Alina Bilt on "Knowledge representation issues in semantic intrusion detection."
Guest lecture by Dimitri Zarzhitsky on "Chemical plume tracing with a robot in a home-built test chamber at UW."
Guest lecture by Prof. Ruben Gamboa on "Automated reasoning (using logic.)"
Guest lecture by Prof. Steve Barrett on "An artificial fly eye."

PRELIMINARY MATERIAL (WHICH WILL BE COVERED AS NEEDED):

Search. Search is basic to planning. However, search will only be covered briefly on one session, and only to the extent required by the students. Read Chapter 3 R&N.
Bayesian belief networks. This material is needed for understanding reasoning under uncertainty. It will only be a brief lecture also, as needed by the students. Read Chapter 13 and 14 R&N as needed.

I. POPULAR AGENT PARADIGMS

Logic-based agents.
Biomimetic agents.
Physicomimetic agents.

II. KNOWLEDGE REPRESENTATION AND REASONING

Representation of knowledge, the frame problem, and the closed-world assumption (CWA).
Semantic networks.
Agents that reason.
The symbol grounding problem.

III. PLANNING AND CONTROL

Planning.

Plan recognition.

Control.

IV. REAL-WORLD CONDITIONS: REASONING OVER TIME UNDER UNCERTAINTY

Monitoring (e.g., Kalman filtering), prediction, learning, and planning under uncertainty.

V. ROBOTS

Mobile ground-based, air-based, aquatic, and spacecraft platforms.

Humanoid robots.

VI. SOME ROBOT APPLICATIONS

Sensor networks.

Simultaneous localization and mapping (SLAM).

Target tracking.

Chemical plume tracing (CPT).

Robotic toys.

VII. SOFTBOTS (SOFTWARE AGENTS)

Mobile agents.

Communication protocols.

Teamwork and negotiation.

VIII. SOME SOFTBOT APPLICATIONS

Entertainment.

Internet shopbots.

IX. ASSURING THE BEHAVIOR OF AGENTS

Formal verification of agent behavior.

Validation of agent behavior.

Social laws for ensuring agent behavior.

Exams:

There will be a midterm but no final exam.

Term Projects:
The presentation format will be a talk with slides, with an optional demo. For demos, fancy graphics is not required -- simple ascii graphics is acceptable. The grade will be based on the content and creativity, not the graphics sophistication. The content of the term project is expected to contain material from one or more topics covered in the lectures and/or papers. A final written report on the project is also due. Students working on a thesis or senior design project related to robots or softbots are welcome to pick a term project topic that will further their ongoing research. The level of sophistication of term projects is expected to be higher for graduate than undergraduate students.

Grading:

The various required work in this class will be counted towards your final grade as follows:

Class Participation (paper presentations and discussions) - 30%

Midterm Exam - 30%

Term Project - 40%

Grading will be on a curve. The professor reserves the right to alter the grading scheme or to take extenuating circumstances into account when assigning grades. Discussion of the course material among students is encouraged, although students are expected to write up their own assignments and programs, unless it is a team project. Copying code, homework solutions, or exam material from any source at all (unless explicitly permitted by the instructor) will most likely result in failing the course. Academic dishonesty will be treated in accordance with the strictest university standards. Students are urged to read University regulation 803, section 3 defines academic dishonesty.