Program & Exhibit Guide

EIGHTEENTH NATIONAL CONFERENCE ON

ARTIFICIAL INTELLIGENCE (AAAI-02)

FOURTEENTH CONFERENCE ON INNOVATIVE APPLICATIONS

OF ARTIFICIAL INTELLIGENCE (IAAI-02)

July 28-August 1, 2002

Shaw Conference Centre and Westin Edmonton Hotel

Edmonton, Alberta, Canada

SPONSORED BY THE AMERICAN ASSOCIATION FOR ARTIFICIAL INTELLIGENCE

Cosponsored by

ACM/SIGART, Alberta Informatics Circle of Research Excellence (iCORE),

Defense Advanced Research Projects Agency (DARPA), NASA Ames Research Center,

NSF's Directorate for Computer and Information Science and Engineering (CISE), Naval Research Laboratory

In cooperation with

The University of Alberta

ContentsAcknowledgments / 2Awards / 3Conference at a Glance / 5DC-02 / 4Exhibition / 18General Information / 33IAAI-02 Program / 13–15Intelligent Systems Demonstrations / 21Invited Talks / 10Maps / 37–39National Botball Exhibition / 31Poster Sessions / 3Registration / 32Robot Building Laboratory / 7Robot Competition and Exhibition / 24Special Events and Programs / 3–4Special Meetings / 4Sponsoring Organizations / 2Technical Program (AAAI / IAAI) / 37–39Tutorial Forum /6Workshop Program / 8

AcknowledgmentsThe American Association for Artificial Intel-ligence wishes to acknowledge and thank thefollowing individuals for their generous con-tributions of time and energy to the successfulcreation and planning of the Eighteenth Na-tional Conference on Artificial Intelligenceand the Fourteenth Conference on InnovativeApplications of Artificial Intelligence.

AAAI Conference Committee ChairJames A. Hendler, University of Maryland

AAAI-02 Program CochairsRina Dechter, University of California, Irvine,Michael Kearns, University of Pennsylvania,and Richard S. Sutton

IAAI-02 Conference ChairSteve Chien, Jet Propulsion Laboratory

IAAI-02 Conference CochairJohn Riedl, University of Minnesota

Intelligent Systems Demos ChairGeorge Ferguson, University of Rochester

Local Arrangements ChairRussell Greiner, University of Alberta

Mobile Robot Competition and Exhibition General CochairsHolly Yanco, University of Massachusetts Low-ell and Tucker Balch, Georgia Institute of Tech-nology

Robot Challenge CochairsBen Kuipers, University of Texas at Austin andAshley Stroupe, Carnegie Mellon University

Robot Rescue Competition CochairsJenn Casper, Mark Micire, and Robin Murphy,University of South Florida

Robot Host Competition CochairsDavid Gustafson, Kansas State University andFrancois Michaud, Universite de Sherbrooke

Robot Exhibition CochairsIan Horswill and Christopher Dac Le,Northwestern University

Mobile Robot Workshop ChairBill Smart, Washington University St. Louis

Robot Building Laboratory ChairDavid Miller, KISS Institute for Practical Robotics

Doctoral Consortium ChairMarie desJardins, University of Maryland,Baltimore County

Student Abstract and Poster CochairsMark Craven, University of Wisconsin,Madison and Sven Koenig, Georgia Institute of Technology

Tutorial ChairMichael L. Littman, Rutgers University

Workshop Chair and CochairBerthe Y. Choueiry, University of Nebraska-Lincoln and Janyce Wiebe, University of Pitts-burgh

A complete listing of the AAAI-02 and IAAI-02 Program Committee members appears inthe conference proceedings.

Sponsoring OrganizationsAAAI gratefully acknowledges the generouscontributions of the following organizationsto AAAI-02:■ ACM/SIGART■ Alberta Informatics Circle of Research Ex-

cellence (iCORE)■ Defense Advanced Research Projects Agen-

cy (DARPA)■ NASA Ames Research Center■ NSF’s Directorate for Computer and Infor-

mation Science and Engineering (CISE)■ Naval Research Laboratory■ University of Alberta

Conte

nts / A

ckno

wled

gmen

ts

2

Special Events and Programs AAAI Recognition AwardThe AAAI Awards will be presented by TomMitchell, AAAI President, and BruceBuchanan, AAAI Past President and AwardsCommittee Chair, on Wednesday, July 31, at8:15 AM in Exhibit Hall C of the Shaw Confer-ence Centre. Recipients will be announced inthe AAAI-02 Program Addendum. The awardswill be given in the following three categories:

Classic Paper AwardThe 2002 AAAI Classic Paper Award will begiven to the author of the most influential pa-per(s) from the Third National Conference onArtificial Intelligence, held in 1983 in Wash-ington, DC.

Distinguished Service AwardThe AAAI Distinguished Service Award recog-nizes one individual each year for extraordi-nary service to the AI community.

The AAAI Effective Expository Writing Award The 2002 AAAI Effective Expository WritingAward was established in 2000 to honor the au-thor(s) of a high quality, effective piece of writ-ing, accessible to the general public or to a broadAI audience (not just a subarea), written withinthe last two years. The contribution should bebased on sound science, interesting ideas or sys-tematic review, with non-trivial content, but theaward is primarily for the exposition.

Fellows Recognition DinnerEach year the American Association for Artifi-cial Intelligence recognizes a small number ofmembers who have made significant sustainedcontributions to the field of artificial intelli-gence, and who have attained unusual distinc-tion in the profession. AAAI is pleased to an-nounce the eight newly elected Fellows for 2002:■ Kevin D. Ashley, University of Pittsburgh■ Michael Gelfond, Texas Tech University■ Eric Horvitz, Microsoft Research■ Henry E. Kyburg, Jr., University of Rochester

and UWF/Institute for Human & MachineCognition

■ Michael I. Jordan, University of California,Berkeley

■ Sarit Kraus, Bar-Ilan University and Uni-versity of Maryland

■ Stephen H. Muggleton, Imperial College ofScience, Technology and Medicine

■ Katia P. Sycara, Carnegie Mellon UniversityThe 2002 Fellows Recognition Dinner will beheld Tuesday, July 30, from 7:30 - 10:00 PM in the

Drawing Room of the Fairmont Hotel Macdon-ald. A reception will begin at 7:30 PM, followedby dinner at 8:00 PM. (By invitation only).

Outstanding Paper AwardThis year, AAAI’s National Conference on Ar-tificial Intelligence honors one paper that ex-emplifies high standards in technical contri-bution and exposition. During the blind re-view process, members of the Program Com-mittee recommended which papers to consid-er for the Outstanding Paper Award. A subsetof the Senior Program Committee, carefullychosen to avoid conflicts of interest, reviewedall such papers and selected the winning pa-per: “On Computing All Abductive Explana-tions” by Thomas Eiter, Technische UniversitätWien and Kazuhisa Makino, Osaka UniversityProgram Cochairs Rina Dechter and MichaelKearns will present the authors with their cer-tificates on Tuesday, July 30, at 8:00 AM in Ex-hibit Hall C of the Shaw Conference Centre.

Presidential AddressTom M. Mitchell, Fredkin Professor of Com-puter Science at Carnegie Mellon University,will give the AAAI Presidential Address on “AIand the Impending Revolution in Brain Sci-ence” on Wednesday, July 31, 8:30 AM in Ex-hibit Hall C of the Shaw Conference Centre.

Technical Paper Poster SessionsThe AAAI-02 Technical Poster Sessions will beheld on Tuesday, July 30, 5:45 – 8:45 PM, andWednesday, July 31, 7:00 – 10:00 PM in the Ball-room on the second floor of the Westin Ed-monton. Sixty technical papers will be present-ed each evening, and will be organized accord-ing to subject area (maps will be available inthe ballroom). All AAAI-02 registrants are en-couraged to attend this integral portion of theconference. Refreshments will be available.

Student Abstract PostersStudents whose abstracts were chosen for in-clusion in the conference proceedings will dis-play their work at the Student Abstract PosterSession in Exhibit Hall AB, Shaw ConferenceCentre on Wednesday, July 31 from 3:00 – 5:30PM, in conjunction with the AI Festival. In ad-dition, participants in the AAAI/SIGARTDoctoral Consortium will display their posterpresentations during this session. All studentswill be available for questions.

3

Opening ReceptionThe AAAI-02 Opening Reception will be heldMonday, July 29, 7:00 – 10:00 PM, at Fort Ed-monton Park (Canada’s largest living historypark). This event will provide the traditionalopportunity for attendees to socialize in aunique setting prior to the beginning of thefirst day of technical sessions. A variety of horsd’oeuvres and a no-host bar will be available.Admittance to the reception is free to AAAI-02registrants including transportation to andfrom the event. A $40.00 CDN per person fee($15.00 CDN for children) will be charged forspouses and other nontechnical conferenceregistrants.

Shuttles will pick up attendees from theShaw Conference Centre at the south side en-trance by the upper level balcony. Shuttles willleave the Conference Centre beginning at 6:30PM and continue until 7:30 PM. Return fromFort Edmonton will begin at 8:30 PM and thelast shuttle will leave at 10:00 PM. Shuttles willdepart from Fort Edmonton at the main en-trance where attendees were dropped off, andwill return them to the conference hotels.

AI FestivalThe AI Festival will be held in Exhibit Hall ABof the Shaw Conference Centre, Wednesday,July 31 from 3:00 – 5:30 PM. This event willprovide attendees the opportunity to strollamong numerous exciting events — the Mo-bile Robot Competition and Exhibition, theIntelligent Systems Demos, the IAAI-02 En-trepreneurs Forum, the Student Posters, and acompetition between high school Botballcompetitors and the Robot Building Labora-tory participants — all enlivened by refresh-ments and conversation. Admittance to thefestival is free to AAAI-02 registrants. A $25.00CDN per person fee ($8.00 CDN for children)will be charged for spouses and other non-technical conference registrants. Guest ticketsare available in onsite registration.

Doctoral Consortium (DC-02)The Seventh AAAI/SIGART Doctoral Consor-tium program will be held on Sunday andMonday, July 28-29 from 8:30am – 5:30 PM inthe Saskatchewan Room, second floor of TheWestin Edmonton. The Doctoral Consortiumprovides an opportunity for a group of Ph.D.students to discuss and explore their researchinterests and career objectives in an interdisci-plinary workshop together with a panel of es-

tablished researchers. The thirteen studentsaccepted to participate in this program will al-so participate in the Student Poster programon Wednesday, July 31, from 3:00 – 5:30 PM

during the AI Festival. All interested AAAI-02student registrants are invited to observe thepresentations and participate in discussions atthe workshop. AAAI and ACM/SIGART grate-fully acknowledge a grant from NSF’s Direc-torate for Computer and Information Scienceand Engineering (CISE) that partially sup-ports student travel to the event.

AAAI Business MeetingThe AAAI Annual Business Meeting will beheld Thursday, August 1, from 1:00 – 1:30 PM

in Salon 6, meeting level, Shaw ConferenceCentre.

Conference Committee MeetingThe Conference Committee Meeting will beheld Thursday, August 1, from 7:30 – 8:30 AM

in Salon 1, meeting level, Shaw ConferenceCentre.

AI Journal Editorial BoardMeetingThe AI Journal Editorial Board lunch will beheld Tuesday, July 30 from 12:30 - 2:00 PM inSalon 1, Meeting Level, Shaw Conference Cen-tre.

Executive Council MeetingThe AAAI Executive Council Meeting will beheld Sunday, July 28, from 9:00 AM – 5:00 PM inTurner Valley, second floor, The Westin Ed-monton. Continental breakfast will be avail-able at 8:30 AM.

Program Committee LuncheonThe AAAI-02/IAAI-02 Program CommitteeLuncheon will be held Tuesday, July 30, 12:30– 2:00 PM in the River View Room, PedwayLevel, Shaw Conference Centre to honor thecontributions of all the members of the AAAI-02 and IAAI-02 Program Committees. (By in-vitation only).

Recep

tions

,DC-0

2,and

Meet

ings

4

Conference at a Glance

5

MORNING AFTERNOON EVENING

SUNDAY, JULY 28Registration Registration Robot Building Lab

Tutorial Forum Tutorial ForumWorkshops Workshops

AAAI/SIGART DC AAAI/SIGART DCRobot Building Lab Robot Building Lab

MONDAY, JULY 29Registration Registration Opening Reception

Tutorial Forum Tutorial ForumWorkshops Workshops

AAAI/SIGART DC AAAI/SIGART DCRobot Building Lab

TUESDAY, JULY 30Registration Registration AAAI–02 Poster Session

Keynote Address AAAI-02 / IAAI-02 AAAI Fellows DinnerAAAI-02 / IAAI-02 Program Committee Lunch

Exhibition / IS Demos Exhibition/IS DemosRobot Programs / Botball Robot Programs / Botball

WEDNESDAY, JULY 31Registration Registration AAAI–02 Poster Session

Presidential Address AAAI-02 / IAAI-02AAAI-02 / IAAI-02 Exhibition / IS Demos

Exhibition / IS Demos Robot Programs /BotballRobot Programs / Botball AI Festival

THURSDAY, August 1Registration AAAI-02 / IAAI-02

AAAI-02 / IAAI-02 Robot WorkshopRobot Workshop AAAI Business Meeting

Tutor

ial Fo

rum

6

Tutorial ForumAAAI-02 technical registration includes admission to up to four tutorials and the correspondingfour tutorial syllabi. A maximum of four consecutive tutorials may be taken due to parallelschedules. Tutorial attendees may redeem their tutorial syllabi tickets at the tutorial rooms. At-tendees who wish to obtain syllabi from other tutorials may purchase them separately for $30.00CDN per syllabus in onsite registration.

Session I: Sunday, July 28

9:00 AM - 1:00 PM

SA1: The State of the Art in Language ModelingJoshua Goodman and Eugene CharniakSalon 8, Shaw Conference Centre

SA2: AI in Space: Unique Challenges and Opportunities (Full Day) Daniel ClancySalon 12, Shaw Conference Centre

SA3: Greedy On-Line Planning Sven Koenig and Anthony StentzSalon 10, Shaw Conference Centre

SA4: Aspects of Qualitative Spatial and Temporal Reasoning Gerard Ligozat, Frank D. Anger, and Hans W. GuesgenSalon 9, Shaw Conference Centre

Session II: Sunday, July 28

2:00 - 6:00 PM

SP1: Practical Approaches to Handling Uncertainty in Planning and SchedulingJ. Christopher Beck and Thierry VidalSalon 10, Shaw Conference Centre

SP2: Collaborative Multi-Agent SystemsBarbara Grosz and Charlie OrtizSalon 8, Shaw Conference Centre

SP3: Practical Machine Learning for Software Engineering Tim Menzies and Gary D. BoetticherSalon 9, Shaw Conference Centre

Session III: Monday, July 29

9:00 AM - 1:00 PM

MA1: Information Integration on the Web Craig Knoblock and Subbarao KambhampatiSalon 12, Shaw Conference Centre

MA2: AI Techniques for Personalized Recommendation (Full Day)Anthony Jameson, Joseph Konstan and John RiedlSalon 10, Shaw Conference Centre

MA3: Algorithms for Combinatorial Auctions and Exchanges Tuomas SandholmSalon 8, Shaw Conference Centre

Tutorial Forum & Robot Building LabSession IV: Monday, July 29

2:00 - 6:00 PM

MP1: Phase Transitions and Structure in Combinatorial Problems Tad Hogg, Carla P. Gomes, Toby Walsh, and Weixiong ZhangSalon 8, Shaw Conference Centre

MP2: Rational Action in Autonomous AgentsMichael Wooldridge and Simon ParsonsSalon 12, Shaw Conference Centre

Robot Building LabThe Robot Building Laboratory will be held in Salon 2, meeting level, Shaw Conference Centreat the following times:

Sunday, July 28 9:00 AM – 9:00 PM

Monday, July 29 9:00 AM – 1:00 PM

Wednesday, July 31 3:00 PM – 5:30 PM (RBL-02 Contest/Exhibition)

Preregistration is required. The Robot Building Lab (RBL) is a chance for AI researchers to ex-periment with hardware. What happens to your favorite AI algorithm when it actually gets em-bodied? How reliable is the real world compared to a simulation? Why do roboticists always seemto be having a better time at the conference than logic theorists? These are the questions that canbest be answered by participating in the RBL. As in the past, this year’s RBL will break the par-ticipants into small groups. Each group will be given a robot kit and then will spend the next dayand a half creating a robot system to achieve this year’s task. The lab will conclude with a friend-ly competition among the different groups. The theme for this year’s lab will be ‘multi-agent co-operation’. Each robot kit will contain enough parts to create two or more independent robotsthat will work together (hopefully) to accomplish the task. Participants are encouraged (but notrequired) to bring a MacOS, Windows 98, or LINUX laptop with them so that there will be mul-tiple programming stations for each group. The results of the lab will be presented as part of therobot exhibition, during the AI Festival on Wednesday afternoon. The RBL is aimed at educa-tors, students and researchers interested in robotics. A general knowledge of programming willbe assumed. No prior robotics experience is required. The lab is being organized and taught bythe KISS Institute for Practical Robotics (KIPR) for AAAI. Instructors and assistants are fromKIPR’s trained staff. David Miller is the lead instructor.

7

Workshop ProgramAttendance at the workshops is limited, and participation is by invitation only. All workshop par-ticipants must register for the AAAI-02 technical program. Registration onsite for a workshop ispossible with the prior permission of the corresponding workshop organizer. The times for eachworkshop are listed below. Workshops will be held in the Shaw Conference Centre and The West-in Edmonton.

Sunday, July 28W1: Agent-Based B2B Electronic Commerce Technologies

Organizer: Brian BlakeConsulate Room, The Westin Edmonton8:30 AM – 6:00 PM

W5: Autonomy, Delegation, and Control: From Inter-Agent to GroupsOrganizers: Henry Hexmoor and Rino FalconeChairman Room, The Westin Edmonton8:45 AM – 6:10 PM

W7: Cognitive Robotics (CogRob2002)Organizers: Chitta Baral and Sheila McIlraithSalon 11, Shaw Conference Centre9:00 AM – 6:00 PM

W8: Game Theoretic and Decision Theoretic AgentsOrganizers: Piotr Gmytrasiewicz and Simon ParsonsSalon 3, Shaw Conference Centre9:00 AM – 5:30 PM

W9: Intelligent Integration of Information and Services on the WebOrganizers: Dean Allemang, Eleni Stroulia and John MylopoulosSalon 17/18, Shaw Conference Centre9:00 AM – 4:30 PM

W11: Meaning NegotiationOrganizer: Paolo BouquetSalon 15/16, Shaw Conference Centre8:30 AM – 6:00 PM

W15: Planning with and for Multiagent Systems (two-day)Organizers: Michael Brenner and Marie desJardins BC Room, The Westin Edmonton2:00 – 6:00 PM

W16: Preferences in AI and CP: Symbolic Approaches (two-day)Organizer: Ulrich Junker Salon 4, Shaw Conference Centre9:00 AM – 5:15 PM

W17: Probabilistic Approaches in SearchToby Walsh and Carla GomesSalon 6, Shaw Conference Centre9:00 AM – 6:00 PM

W19: Semantic Web Meets Language Resources Organizers: Nancy Ide and Chris WeltyChancellor Room, The Westin Edmonton9:00 AM – 6:00 PM

Works

hop P

rogram

8

Monday, July 29W3: Artificial Intelligence for Intelligent Business

Organizers: Daniel E. O’Leary and Alun PreeceChancellor Room, The Westin Edmonton8:30 am - 3:00 pm

W4: Automation as Caregiver: The Role of Intelligent Technology in Elder Care (W4)Organizer: Karen HaighSalon 11, Shaw Conference Centre8:30 AM – 6:00 PM

W6: Coalition Formation in Dynamic Multiagent EnvironmentsOrganizers: Leen-Kiat Soh and Charlie OrtizConsulate Room, The Westin Edmonton8:45 AM – 4:30 PM

W10: Intelligent Situation-Aware Media and Presentations (ISAMP)Organizer: Rainer Malaka and Antonio KruegerSalon 17/18, Shaw Conference Centre9:30 AM – 6:00 PM

W13: Multi-Agent Modeling and Simulation of Economic SystemsOrganizers: Koichi Kurumatani, Shu-Heng Chen and Azuma OhuchiSalon 6, Shaw Conference Centre8:55 AM – 5:00 PM

W14: Ontologies for the Semantic WebOrganizers: Adam Pease, Richard Fikes, and Jim HendlerSalon 15/16, Shaw Conference Centre9:00 AM – 5:30 PM

W15: Planning with and for Multiagent Systems (two-day)Organizers: Michael Brenner and Marie desJardinsBC Room, The Westin Edmonton9:00 AM – 6:00 PM

W16: Preferences in AI and CP: Symbolic Approaches (two-day)Organizer: Ulrich JunkerSalon 4, Shaw Conference Centre9:00 AM – 1:00 PM

W18: Real-Time Decision Support and Diagnosis Systems (Joint Workshop with KDD-02 and UAI hosted by AAAI-02)Organizers: Haipeng Guo, Eric Horvitz, William H. Hsu, and Eugene Santos Jr.Salon 3, Shaw Conference Centre8:30 AM – 6:00 PM

W20: Spatial and Temporal ReasoningOrganizers: Hans W. Guesgen, Frank D. Anger, and Gerard LigozatChairman Room, The Westin Edmonton9:00 AM – 5:00 PM

Thursday, August 1W12: Mobile Robotic Competition and Exhibition Workshop

Organizer: Bill Smart Salon 3, Shaw Conference Centre9:00 AM – 5:00 PM

Workshop Program

9

10

All AAAI-02 invited presentations will be heldin Exhibit Hall C, assembly level, Shaw Con-ference Centre. IAAI-02 invited presentationswill be held in Salon 8, meeting level, ShawConference Centre.

Tuesday, July 30

8:30 – 9:30 AM

AAAI Keynote Address

Probabilistic AI and Information RetrievalMichael I. Jordan, Department of Electrical Engi-neering and Computer Science & Department ofStatistics, University of California, Berkeley

Exhibit Hall C, Assembly Level, Shaw Conference Centre

Much progress has been made in recent yearsin the area of information retrieval, in partic-ular as embodied in Internet search enginetechnology. Much progress has also been madein probabilistic, graph-theoretic AI. What arethe possibilities for bringing these two lines ofresearch together — for viewing large-scale in-formation retrieval as a core enabling technol-ogy for AI systems, and for asking IR systemsto exhibit true inferential capabilities. Jordanwill discuss research aimed at bridging theAI/IR gap.

9:00 – 10:30 AM

IAAI Invited Talk

Robot-Assisted Urban Search and Rescueat the WTC: Where’s the AI?Robin R. Murphy, Computer Science and Engi-neering, University of South Florida

Salon 8, Meeting Level, Shaw Conference Centre

On September 11, 2001, the Center for Robot-Assisted Search and Rescue responded withinsix hours to the WTC disaster; this is the firstknown use of robots for USAR. The Universi-ty of South Florida was one of the four robotteams, and only academic institution. The USFteam participated on-site in the search effortsfrom September 12 through 22, collecting andarchiving data on the use of robots.

This talk will provide an overview of theuse of robots for USAR as well as discuss whatAI techniques were available, what was actual-ly used, and why. It will also summarize thekey lessons learned from the robotics efforts atthe WTC. The lessons learned cover the areasof platforms and mobility, sensors and sensingstrategies, control, and human-robot interac-tions. Possibly the most pervasive lessonlearned is that robots for USAR must be con-sidered from an “information technology”perspective, where platforms, sensors, controlschemes, networks, and interfaces must all be

co-evolved to ensure the information extract-ed by the robots is truly usable by the rescuecommunity.

Extensive video footage of the site and“robot’s eye” views will be shown.

Wednesday, July 31

8:30 – 9:30 AM

AAAI Presidential Address

AI and the Impending Revolution inBrain ScienceTom M. Mitchell, Carnegie Mellon University

Exhibit Hall C, Assembly Level, Shaw Conference Centre

Tom M. Mitchell is the Fredkin Professor ofComputer Science at Carnegie Mellon Univer-sity. He is author of the textbook MachineLearning, a member of the National ResearchCouncil’s Computer Science and Telecommu-nications Board, and President of the AAAI.Mitchell’s research over the years has dealtwith theoretical and practical issues in ma-chine learning. During 2000-2001 he served asChief Scientist at WhizBang! Labs, a companythat employs machine learning to extract de-tailed factual information from text. Since re-turning to Carnegie Mellon, his research hasfocused on functional Magnetic ResonanceImaging of the human brain. Mitchell is theFounding Director of CMU’s Center for Auto-mated Learning and Discovery, an interdisci-plinary research center specializing in machinelearning and data mining.

11:55 AM – 12:45 PM

AAAI Invited Panel

AAAI Mobile Robot CompetitionModerators: Holly Yanco, University of Mas-sachusetts Lowell and Tucker Balch, Georgia Insti-tute of Technology

Exhibit Hall C, Assembly Level, Shaw Conference Centre

Panelists: Peter Bonasso, Metrica; David Miller, Universityof Oklahoma and KIPR; Alan Schultz, Naval ResearchLaboratory; and Reid Simmons, Carnegie Mellon Univer-sity

11:10 AM – 12:40 PM

IAAI Invited Panel

Pioneering AI Businesses I:A 20-Year ReviewModerator: Neil Jacobstein, Chairman, President &CEO of Teknowledge Corporation

Salon 8, Meeting Level, Shaw Conference Centre

Panelists: Ed Feigenbaum, Professor of Computer Scienceand Coscientific Director of the Knowledge Systems Lab-oratory, Stanford University; Mark Fox, Professor of In-dustrial Engineering, University of Toronto and Chair-man & CEO, Novator Systems; Amos Barzilay, GeneralPartner (United States), Walden International

Several AI-based businesses started in theearly 1980s. They underwent a classic boom

Invite

d Talk

s

and bust cycle. Hype exceeded expectations,and some investors and technologists lost pa-tience. However, history shows that in cases ofdisruptive technological innovation, forecastsare usually too optimistic in the short run, andtoo conservative in the long run. Is that thecase with AI businesses? This panel of AI en-trepreneurs will review the technology baseand history of pioneering AI businesses, ex-tract lessons learned, and identify future op-portunities. Companies discussed will includeIntelliCorp, Teknowledge, Inference, Syntelli-gence, Carnegie Group, Cycorp, and others. Aninteractive question and answer session withpanel members will follow brief presentationsfrom each panelist.

2:00 – 3:30 PM

IAAI-02 Invited Panel

Pioneering AI Businesses II: Recent StartupsModerator: Craig Knoblock, Research AssociateProfessor, University of Southern California, Co-founder and Chief Scientist, Fetch Technologies

Salon 8, Meeting Level, Shaw Conference Centre

Panelists: Tom Mitchell, Fredkin Professor of ComputerScience, Carnegie Mellon University, Former Chief Scien-tist and VP at WhizBang; Yoav Shoham, Associate Profes-sor, Stanford University, Co-founder and Chairman,TradingDynamics (ARBA) and Cariocas; Daniel Weld,Professor, University of Washington, Co-founder of Net-bot, AdRelevance, Nimble Technology, and Asta Net-works, Venture Partner, Madrona Venture Group

This panel will focus on the process of startingan AI company. The challenges in creating anew company include how to apply a technol-ogy to address a specific market need and howto run a successful business. The speakers onthis panel are AI researchers that have been in-volved in recent startups. Some of the issuesthe panelists will discuss include how to gofrom a technology to a business, how to getfunding for a company, and what pitfalls towatch out for. An interactive question and an-swer session with panel members will followbrief presentations from each panelist.

2:15 – 3:15 PM

AAAI Invited Talk

Perspectives on Artificial IntelligencePlanningHector Geffner, ICREA — Universitat PompeuFabra (Barcelona)

Exhibit Hall C, Assembly Level, Shaw Conference Centre

Planning has always been a key area in Artifi-cial Intelligence. In its general form, planningis concerned with the automatic synthesis ofaction strategies (plans) from a description ofactions, sensors, and goals. Planning thus con-trasts with two other approaches to intelligentbehavior: the programming approach, whereaction strategies are defined by hand, and the

learning approach, where action strategies areinferred from experience.

Different assumptions about the nature ofactions, sensors, and costs lead to variousforms of planning: (1) Planning with completeinformation and deterministic actions, (2)planning with non-deterministic actions andsensing, and (3) planning with temporal andconcurrent actions, etc. Most work so far hasbeen devoted to “classical” planning (see point1 above), where significant changes have takenplace in the last few years. On the method-ological side, the area has become more em-pirical with experimental evaluation beingroutine; on the technical side, approachesbased on heuristic or constrained-based searchhave taken over blind-search approaches.

In this talk, Geffner will provide a coherentpicture of planning in AI while trying to con-vey some of the current excitement in the field.He’ll make emphasis on the mathematicalmodels that underlie various forms of plan-ning, and the ideas that have been found mostuseful computationally.

3:30 – 4:30 PM

Collocated with the AI Festival

IAAI-02 AI Entrepreneurs ForumExhibit Hall AB, Assembly Level, Shaw Conference Centre

This forum will provide an open and informalsetting for AI pioneers, technologists, en-trepreneurs, venture capitalists, legal, and in-tellectual property experts to network and dis-cuss issues in starting and running AI-basedcompanies.

Thursday, August 1

8:30 – 9:30 AM

AAAI-02/IAAI-02 Joint Invited Talk

Human Level “Strong” AI: The Prospectsand ImplicationsRaymond Kurzweil, KurzweilAI.net (Kurzweil Ac-celerating Intelligence Network)

Exhibit Hall C, Assembly Level, Shaw Conference Centre

Three-dimensional molecular computing willprovide the hardware for human-level“strong” AI well before 2030. The more impor-tant software insights will be gained in partfrom the reverse-engineering of the humanbrain, a process well under way. Once nonbio-logical intelligence matches the range and sub-tlety of human intelligence, it will necessarilysoar past it because of the continuing acceler-ation of information-based technologies, aswell as the ability of machines to instantlyshare their knowledge. The implication will be

Invited Talks

11

an intimate merger between the technology-creating species and the evolutionary processit spawned.

2:15 – 3:15 PM

AAAI Invited Talk

Dimension Reduction that Preserves In-formation and Neural CodingNaftali Tishby, The Hebrew University

Exhibit Hall C, Assembly Level, Shaw Conference Centre

Many cognitive functions, such as prediction,feature extraction, noise filtering, and learn-ing, can be viewed as special cases of one prin-ciple: compression while preserving informa-tion. This information theoretic principle wasturned into a computational paradigm: the in-formation bottleneck method. This variation-al method yielded several novel learning anddata analysis algorithms, with many applica-tions to information retrieval as well as to

analysis of neural coding in several neurobio-logical systems that were carried in Tishby’slab. In this talk Tishby will focus on a new ap-proach to data dimensionality reduction thatstems from this principle. Here he searches forlow dimensional (nonlinear) reduction of co-occurrence (or contingency) tables that pre-serve the (mutual) information in the table.He gives a new alternate-projection algorithmfor achieving such a reduction and shows itsconvergence to an optimal set of informationpreserving features. This approach is particu-larly useful when the data is not naturallyquantized but rather represented by low di-mension continuous features. Such a reduc-tion may have interesting biological implica-tions. (Based on joint work with Amir Glober-son and Noam Slonim.)

12

Invite

d Talk

s

Get All the AAAI and IAAI Conference Proceedings, Plus AI Magazine, on CD-ROM

SPECIAL RATES!

AAAI 2000 / IAAI 2000 Conference Proceedings CD-ROMThis CD-ROM set consists of all the papers from the 2000 National Conferences on Artificial Intelligence, as well as the 2000 Innovative Ap-plications of Artificial Intelligence Conferences. The single CD contains approximately 260 technical papers in PDF format, and is readable bymost Windows (1998 and later), Macintosh, and many UNIX platforms.

1 CD ($45.00 CDN — AAAI members; $175.00 CDN nonmembers)

AAAI 1998–99, IAAI 1998–99, KDD 1995–98 CD-ROM SetThis CD-ROM set consists of all the papers from the 1998 and 1999 National Conferences on Artificial Intelligence, as well as the 1998 and1999 Innovative Applications of Artificial Intelligence Conferences. Included also on this CD are the 1995, 1996, 1997, and 1998 KnowledgeDiscovery and Data Mining Conference papers. The single CD contains approximately 260 technical papers in PDF format, and is readable bymost Windows, Macintosh, and many UNIX platforms.

1 CD ($45.00 CDN — AAAI members; $175.00 CDN nonmembers)

AAAI 1980–97, IAAI 1993–97 CD-ROM SetThis archival CD-ROM set consists of all the papers from the 1980, 1982, 1983, 1984, 1986, 1987, 1988, 1990, 1991, 1992, 1993, 1994, 1996,and 1997 AAAI National Conferences, as well as the papers from the 1993, 1994, 1995, 1996, and 1997 Innovative Applications of Artifi-cial Intelligence Conferences. The set consists of more than 2,200 papers in PDF format, and is readable by most Windows, Macintosh,and many UNIX platforms.

1 CD ($95.00 CDN — AAAI members; $250.00 CDN nonmembers)

AI Magazine 1980–2001 CD-ROM SetThis new CD-ROM consists of all the articles from AI Magazine published between 1980 and 2001. The single CD is readable by most Win-dows (1998 and later), Macintosh, and many UNIX platforms.

1 CD ($45.00 CDN — AAAI members; $175. CDN nonmembers)

To purchase, see the AAAI Conference Registrar. You must be an AAAI member in good standing, or purchase an AAAI membership onsite to qualify for the

conference special membership rate. Offer expires August 1, 2002. Purchase must be made at AAAI–02 onsite. No additional discounts apply.

13

Tuesday, July 30—AAAI-02 Sessions(All AAAI-02 sessions will be held in Hall C.)

: - : AAAI-02 Opening Remarks and Outstanding Paper Award PresentationRina Dechter and Michael Kearns, AAAI-02 Program Cochairs

IAAI-02 Awards PresentationSteve Chien and John Riedl, IAAI-02 Chair and Cochair

: - : Keynote AddressProbabilistic AI and Information RetrievalMichael I. Jordan, University of California, Berkeley Introduction: Rina Dechter and Michael Kearns

: - : Probabilistic ReasoningSession Chair: David McAllester

A Distance Measure for Bounding Probabilistic Belief ChangeHei Chan and Adnan Darwiche

Visual Exploration and Incremental Utility ElicitationJim Blythe

A Graphical Criterion for the Identification of Causal Effects in Linear ModelsCarlos Brito and Judea Pearl

: - : Break

: - : PlanningSession Chair: Weixiong Zhang

Plan Evaluation with Incomplete Action DescriptionsAndrew Garland and Neal Lesh

Symbolic Heuristic Search for Factored Markov Decision ProcessesZhengzhu Feng and Eric A. Hansen

Algorithms for a Temporal Decoupling Problem in Multi-Agent PlanningLuke Hunsberger

: - : Lunch BreakAAAI-02/IAAI-02 Program Committee Lunch(by invitation only)

: - : Learning ISession Chair: Peter Stone

Polynomial-Time Reinforcement Learning of Near-Optimal PoliciesKarèn Pivazyan and Yoav Shoham

Minimum Majority Classification and BoostingPhilip M. Long

Data Perturbation for Escaping Local Maxima in LearningGal Elidan, Matan Ninio, Nir Friedman, and Dale Shuurmans

: - : Break

: - : Constraint Satisfaction/SearchSession Chair: Weixiong Zhang

A New Algorithm for Optimal Bin PackingRichard E. Korf

PROMPTDIFF: A Fixed-Point Algorithm for Comparing Ontology VersionsNatalya F. Noy and Mark A. Musen

Node and Arc Consistency in Weighted CSPJavier Larrosa

Model Induction: A New Source of CSP Model RedundancyY.C. Law and J.H.M. Lee

: ‒ : Poster Session IBallroom, Westin Edmonton

Tuesday, July 30—IAAI-02 Sessions(All IAAI-02 sessions will be held in Salon 8, unless otherwise noted.)

: - : AAAI-02 Opening Remarks and Outstanding Paper Award PresentationRina Dechter and Michael Kearns, AAAI-02 Program Cochairs

IAAI-02 Awards PresentationSteve Chien and John Riedl, IAAI-02 Chair and Cochair(Hall C)

: - : IAAI-02 Opening RemarksSteve Chien and John Riedl, IAAI-02 Chair and Cochair

: - : Invited TalkRobot Assisted Urban Search and Rescue at the WTC:Where’s the AI?Robin Murphy, University of South FloridaIntroduction: David Kortenkamp

: - : Break

: - : Perception and MonitoringSession Chair: David Kortenkamp

Applying Perceptually Driven Cognitive Mapping to Virtual Urban Environments (Emerging Technology)Randall W. Hill, Jr., Changhee Han, and Michael van Lent

MiTAP, Text and Audio Processing for Bio-Security: A Case Study (Deployed Application)Laurie Damianos, Jay Ponte, Steve Wohlever, Florence Reeder, David Day,George Wilson, and Lynette Hirschman

Computational Vulnerability Analysis for Information Survivability (Emerging Technology)Howard Shrobe

: - : Lunch

: - : Planning and SchedulingSession Chair: Steve Chien

Staff Scheduling for Inbound Call Centers and Customer Contact Centers(Deployed Application)Alex Fukunaga, Ed Hamilton, Jason Fama, David Andre,Ofer Matan, and Illah Nourbakhsh

Getting from Here to There: Interactive Planning and Agent Execution for Optimizing Travel (Emerging Technology)José Luis Ambite, Greg Barish, Craig A. Knoblock,Maria Muslea, Jean Oh, and Steven Minton

UTTSExam: A Campus-Wide University Exam-Timetabling System (Deployed Application)Andrew Lim, Juay-Chin Ang, Wee-Kit Ho, and Wee-Chong Oon

: - : Break

: - : Knowledge Formation and ManagementSession Chair: Bob Engelmore

Knowledge Formation and Dialogue Using the KRAKEN Toolset (Emerging Technology)Kathy Panton, Pierluigi Miraglia, Nancy Salay, Robert C. Kahlert,David Baxter, and Roland Reagan

A Web-Based Ontology Browsing and Editing System (Emerging Technology)Jérôme Thoméré, Ken Barker, Vinay Chaudhri, Peter Clark, Michael Eriksen,Sunil Mishra, Bruce Porter, and Andres Rodriguez

WhyNot: Debugging Failed Queries in Large Knowledge Bases(Emerging Technology)Hans Chalupsky and Thomas A. Russ

Wednesday, July 31—AAAI-02 Sessions

: AAAI Special Awards PresentationTom Mitchell, AAAI President and Bruce Buchanan,AAAI Past President and Awards Committee Chair

: - : AAAI Presidential AddressAI and the Impending Revolution in Brain ScienceTom M. Mitchell, Carnegie Mellon UniversityIntroduction: Bruce G. Buchanan, AAAI Past President

: - : Representation and Reasoning / Robot ApplicationSession Chair: Richard Korf

AAAI-02 Outstanding PaperOn Computing all Abductive ExplanationsThomas Eiter and Kazuhisa Makino

Measuring Inconsistency in Knowledge via Quasi-Classical ModelsAnthony Hunter

FastSLAM: A Factored Solution to the Simultaneous Localization and Mapping ProblemMichael Montemerlo, Sebastian Thrun, Daphne Koller, and Ben Wegbreit

: - : Break

: - : Robot ApplicationsSession Chair: Dieter Fox

Watch Their Moves: Applying Probabilistic Multiple Object Tracking to Autonomous Robot SoccerThorsten Schmitt, Michael Beetz, Robert Hanek, and Sebastian Buck

Experiences with a Mobile Robotic Guide for the ElderlyMichael Montemerlo, Joelle Pineau, Nicholas Roy, Sebastian Thrun,and Vandi Verma

: - : Invited PanelAAAI Mobile Robot CompetitionModerators: Holly Yanco, University of Massachusetts Lowell and Tucker Balch, Georgia Institute of TechnologyIntroduction: Michael KearnsPanelists: Peter Bonasso, Metrica; David Miller, University of Oklahoma and KIPR; Alan Schultz, Naval Research Laboratory; and Reid Simmons, Carnegie Mellon University

: - : Lunch Break

: - : Invited TalkPerspectives on Artificial Intelligence PlanningHector Geffner, ICREA - Universitat Pompeu Fabra (Barcelona) Introduction: Rina Dechter

: - : AI Festival

: - : Poster Session IIBallroom, Westin Edmonton

Wednesday, July 31—IAAI-02 Sessions

: AAAI Special Awards PresentationTom Mitchell, AAAI President and Bruce Buchanan,AAAI Past President and Awards Committee Chair

(Hall C)

: - : AAAI Presidential AddressAI and the Impending Revolution in Brain ScienceTom M. Mitchell, Carnegie Mellon UniversityIntroduction: Bruce G. Buchanan, AAAI Past President

(Hall C)

: - : Learning and DesignSession Chair: John Riedl

A Structure Based Configuration Tool: Drive Solution Designer – DSD (Deployed Application)Christoph Ranze Thorsten Scholz, Thomas Wagner, Andreas Günter,Otthein Herzog, Oliver Hollmann, Christoph Schlieder, and Volker Arlt

An Analogy Ontology for Integrating Analogical Processing and First-Principles Reasoning (Emerging Technology)Kenneth D. Forbus, Thomas Mostek, and Ron Ferguson

: - : Break

: - : Invited PanelPioneering AI Businesses I: A 20-Year ReviewModerator: Neil Jacobstein, Teknowledge Corporation

12:40 – 2:00 pmLunch

: ‒ : Invited PanelPioneering AI Businesses II: Recent StartupsModerator: Craig Knoblock, University of Southern California and Fetch Technologies

: - : AI Festival(Hall AB)

: - : IAAI-02 AI Entrepreneurs ForumCollocated with AI Festival

14

Thursday, August 1—AAAI-02 Sessions

: - : AAAI-02/IAAI-02 Joint Invited TalkHuman Level "Strong" AI: The Prospects and ImplicationsRaymond Kurzweil, KurzweilAI.net (Kurzweil Accelerating Intelligence Network) Introduction: Neil Jacobstein

: - : Learning IISession Chair: David McAllester

State Abstraction for Programmable Reinforcement Learning AgentsDavid Andre and Stuart J. Russell

Learning Temporal, Relational, Force-Dynamic Event Definitions from VideoAlan Fern, Jeffrey Mark Siskind, and Robert Givan

Bootstrap Learning for Place Recognition Benjamin Kuipers and Patrick Beeson

: - : Break

: - : Multiagent SystemsSession Chair: Peter Stone

(Im)possibility of Safe Exchange Mechanism DesignTuomas Sandholm and XiaoFeng Wang

Multi-Agent Algorithms for Solving Graphical Games David Vickrey and Daphne Koller

Reinforcement Learning of Coordination in Cooperative Multi-Agent Systems Spiros Kapetanakis and Daniel Kudenko

Complexity of Manipulating Elections with Few CandidatesVincent Conitzer and Tuomas Sandholm

: - : Lunch

: - : Invited TalkDimension Reduction that Preserves Information and Neural CodingNaftali Tishby, The Hebrew University Introduction: Michael Kearns

: - :Break

: - : SatisfiabilitySession Chair: Tuomas Sandholm

Enhancing Davis Putnam with Extended Binary Clause ReasoningFahiem Bacchus

Comparing Phase Transitions and Peak Cost in PP-Complete Satisfiability ProblemsDelbert D. Bailey, Víctor Dalmau, and Phokion G. Kolaitis

Learning for Quantified Boolean Logic SatisfiabilityEnrico Giunchiglia, Massimo Narizzano, and Armando Tacchella

Dynamic Restart PoliciesHenry Kautz, Eric Horvitz, Yongshao Ruan, Carla Gomes, and Bart Selman

Thursday, August 1—IAAI-02 Sessions

: - : AAAI-02/IAAI-02 Joint Invited TalkHuman Level "Strong" AI: The Prospects and ImplicationsRaymond Kurzweil, KurzweilAI.net (Kurzweil Accelerating Intelligence Network) Introduction: Neil Jacobstein(Hall C)

: - : Kurzweil Q&A Session(Salon 8)

: - : Break

: - : InternetSession Chair: Neil Jacobstein

The 2001 Trading Agent Competition (Emerging Technology)Michael P. Wellman, Amy Greenwald, Peter Stone, and Peter R. Wurman

RightNow eService Center: Internet Customer Service Using a Self-LearningKnowledge Base (Deployed Application)Stephen D. Durbin, Doug Warner, J. Neal Richter, and Zuzana Gedeon

A Decision-Support System for Quote Generation (Deployed Application)Richard Goodwin, Rama Akkiraju, and Fred Wu

: - : Lunch

: - : Military ApplicationsSession Chair: Russell Knight

Toward Practical Knowledge-Based Tools for Battle Planning and Scheduling(Emerging Technology)Alexander Kott, Larry Ground, Ray Budd, Lakshmi Rebbapragada,and John Langston

AI on the Battlefield: An Experimental Exploration (Emerging Technology)Robert Rasch, Alexander Kott, and Kenneth D. Forbus

Development and Deployment of a Disciple Agent for Center of GravityAnalysis (Deployed Application)Gheorghe Tecuci, Mihai Boicu, Dorin Marcu, Bogdan Stanescu, Christina Boicu,Jerry Comello, Antonio Lopez, James Donlon, and William Cleckner

Intelligent Control of Auxiliary Ship Systems (Emerging Technology)David Scheidt, Christopher McCubbin, Michael Pekala, Shon Vick,and David Alger

15

CobotDS: A Spoken Dialogue System for ChatMichael Kearns, Charles Isbell, Satinder Singh, Diane Litman, andJessica HoweExploiting Auditory Fovea in Humanoid-Human InteractionKazuhiro Nakadai, Hiroshi G. Okuno, and Hiroaki KitanoTowards CST-Enhanced SummarizationZhu Zhang, Sasha Blair-Goldensohn, and Dragomir R. Radev

PlanningPlanning with a Language for Extended GoalsUgo Dal Lago, Marco Pistore, and Paolo TraversoSymbolic Heuristic Search for Factored Markov Decision ProcessesZhengzhu Feng and Eric A. HansenPlan Evaluation with Incomplete Action DescriptionsAndrew Garland and Neal LeshAlgorithms for a Temporal Decoupling Problem in Multi-Agent PlanningLuke HunsbergerD*LiteSven Koenig and Maxim LikhachevSpeeding Up the Calculation of Heuristics for Heuristic Search-Based PlanningYaxin Liu, Sven Koenig and David FurcyIterative-Refinement for Action Timing DiscretizationTodd W. NellerA Logical Measure of Progress for PlanningAarati Parmar

Probabilistic and Causal ReasoningVisual Exploration and Incremental Utility ElicitationJim BlytheA Graphical Criterion for the Identification of Causal Effects inLinear ModelsCarlos Brito and Judea PearlA Distance Measure for Bounding Probabilistic Belief ChangeHei Chan and Adnan DarwicheStrategies for Determining Causes of EventsMark HopkinsTree Approximation for Belief UpdatingRobert Mateescu, Rina Dechter, and Kalev KaskAccuracy vs. Efficiency Trade-offs in Probabilistic DiagnosisIrina Rish, Mark Brodie, and Sheng MaA General Identification Condition for Causal EffectsJin Tian and Judea PearlA New Characterization of the Experimental Implications ofCausal Bayesian NetworksJin Tian and Judea Pearl

SatisfiabilityEnhancing Davis Putnam with Extended Binary Clause ReasoningFahiem BacchusComparing Phase Transitions and Peak Cost in PP-CompleteSatisfiability ProblemsDelbert D. Bailey, Víctor Dalmau, and Phokion G. KolaitisInference Methods for a Pseudo-Boolean Satisfiability SolverHeidi E. Dixon and Matthew L. GinsbergAutomated Discovery of Composite SAT Variable-SelectionHeuristicsAlex FukunagaLearning for Quantified Boolean Logic SatisfiabilityEnrico Giunchiglia, Massimo Narizzano, and Armando TacchellaAn Adaptive Noise Mechanism for WalkSATHolger H. Hoos

VisionDetection and Classification of Motion BoundariesRichard Mann and Allan D. JepsonRecognizing Multitasked Activities from Video Using StochasticContext-Free GrammarDarnell Moore and Irfan EssaThe OD Theory of TOD: The Use and Limits of Temporal Information for Object DiscoveryBrandon C. S. Sanders, Randal C. Nelson, and Rahul Sukthankar

Poster Session IIWednesday, July 30 (7:00 - 10:00 PM)

Constraint Satisfaction / SearchScheduling Contract Algorithms on Multiple ProcessorsDaniel S. Bernstein, Theodore J. Perkins, Shlomo Zilberstein, andLev FinkelsteinMultiple-Goal Search Algorithms and their Application to WebCrawlingDmitry Davidov and Shaul Markovitch

16

Techn

ical P

oster

Sessi

ons Poster Session I

Tuesday, July 29 (5:45 - 8:45 PM)

Constraint Satisfaction / SearchThe Yard Allocation ProblemPing Chen, Zhaohui Fu, and Andrew LimSearching for Backbones and Fat: A Limit-Crossing Approachwith ApplicationsSharlee Climer and Weixiong ZhangGenerating Random Solutions for Constraint Satisfaction ProblemsRina Dechter, Kalev Kask, Eyal Bin and Roy EmekGraph Coloring with Quantum HeuristicsAlex Fabrikant and Tad HoggPreference-Based Search and Multi-Criteria OptimizationUlrich JunkerHuman-Guided Tabu SearchGunnar W. Klau, Neal Lesh, Joe Marks, and Michael MitzenmacherNode and Arc Consistency in Weighted CSPJavier LarrosaIntegrating Local Search and Network Flow to Solve the Inventory Routing ProblemHoong Chuin Lau, Qizhang Liu, and Hirotaka OnoModel Induction: A New Source of CSP Model RedundancyY.C. Law and J.H.M. Lee

LearningContentful Mental States for Robot BabyPaul R. Cohen, Tim Oates, Carole R. Beal, and Niall AdamsData Perturbation for Escaping Local Maxima in LearningGal Elidan, Matan Ninio, Nir Friedman, and Dale ShuurmansPruning and Dynamic Scheduling of Cost-Sensitive EnsemblesWei Fan, Fang Chu, Haixun Wang, and Philip S. YuSpecific-to-General Learning for Temporal EventsAlan Fern, Robert Givan, and Jeffrey Mark SiskindStructural Extension to Logistic Regression: Discriminative Parameter Learning of Belief Net ClassifiersRussell Greiner and Wei ZhouBootstrap Learning for Place RecognitionBenjamin Kuipers and Patrick BeesonReinforcement Learning for POMDPs Based on Action Valuesand Stochastic OptimizationTheodore J. PerkinsPolynomial-Time Reinforcement Learning of Near-Optimal PoliciesKarèn Pivazyan and Yoav ShohamHierarchical Latent Class Models for Cluster AnalysisNevin L. Zhang

Markov Decision ProcessesA POMDP Formulation of Preference Elicitation ProblemsCraig BoutilierContext-Specific Multiagent Coordination and Planning withFactored MDPsCarlos Guestrin, Shobha Venkataraman, and Daphne KollerNearly Deterministic Abstractions of Markov Decision ProcessesTerran Lane and Leslie Pack KaelblingThe Size of MDP Factored PoliciesPaolo LiberatoreGreedy Linear Value-Approximation for Factored Markov Decision ProcessesRelu Patrascu, Pascal Poupart, Dale Schuurmans, Craig Boutilierand Carlos Guestrin

Multiagent Systems IComplexity of Manipulating Elections with Few CandidatesVincent Conitzer and Tuomas SandholmA Logic-Based Model of Intentions for Multi-Agent SubcontractingJohn Grant, Sarit Kraus, and Donald PerlisReinforcement Learning of Coordination in Cooperative Multi-Agent SystemsSpiros Kapetanakis and Daniel KudenkoThe Design of Collectives of Agents to Control Non-Markovian SystemsJohn W. Lawson and David H. Wolpert(Im)possibility of Safe Exchange Mechanism DesignTuomas Sandholm and XiaoFeng WangMulti-Agent Algorithms for Solving Graphical GamesDavid Vickrey and Daphne KollerDistributed Breakout RevisitedWeixiong Zhang and Lars Wittenburg

Natural Language ProcessingLearning Pattern Rules for Chinese Named Entity ExtractionTat-Seng Chua and Jimin LiuLanguage Modeling for Soft KeyboardsJoshua Goodman, Gina Venolia, Keith Steury, and Chauncey Parker

Sven Wachsmuth and Gerhard SagererValue Iteration Working with Belief SubsetWeixiong Zhang and Nevin L. Zhang

Multiagent Systems II: AuctionsSolving Concisely Expressed Combinatorial Auction ProblemsCraig BoutilierPartial-Revelation VCG Mechanism for Combinatorial AuctionsWolfram Conen and Tuomas SandholmBidding Clubs in First-Price AuctionsKevin Leyton-Brown, Yoav Shoham, and Moshe TennenholtzTruthful Approximation Mechanisms for Restricted Combinato-rial AuctionsAhuva Mu’alem and Noam NisanStructural Leverage and Fictitious Play in Sequential AuctionsWeili Zhu and Peter R. Wurman

Multiagent Systems III: Game TheoryVote Elicitation: Complexity and Strategy-ProofnessVincent Conitzer and Tuomas SandholmDispersion Games: General Definitions and Some Specific Learning Results /Trond Grenager, Rob Powers, and Yoav ShohamCompetitive Safety AnalysisMoshe Tennenholtz

Planning II: Actions & Temporal ReasoningReasoning about Actions in a Probabilistic SettingChitta Baral, Nam Tran, and Le-Chi TuanA Method for Metric Temporal ReasoningMathias BroxvallNon-Markovian Control in the Situation CalculusAlfredo Gabaldon

RoboticsRobust Global Localization Using Clustered Particle FilteringAdam Milstein, Javier Nicolás Sánchez, and Evan Tang WilliamsonExperiences with a Mobile Robotic Guide for the ElderlyMichael Montemerlo, Joelle Pineau, Nicholas Roy, Sebastian Thrun,and Vandi VermaFastSLAM: A Factored Solution to the Simultaneous Localizationand Mapping Problem Michael Montemerlo, Sebastian Thrun, Daphne Koller, and Ben WegbreitWatch Their Moves: Applying Probabilistic Multiple ObjectTracking to Autonomous Robot SoccerThorsten Schmitt, Michael Beetz, Robert Hanek, and SebastianBuckCD*: A Real-Time Resolution Optimal Re-Planner for GloballyConstrained ProblemsAnthony Stentz

SatisfiabilityA Compiler for Deterministic, Decomposable Negation NormalFormAdnan DarwicheA Mixture-Model for the Behaviour of SLS Algorithms for SATHolger H. HoosSetA*: An Efficient BDD-Based Heuristic Search AlgorithmRune M. Jensen, Randal E. Bryant, and Manuela M. VelosoDynamic Restart PoliciesHenry Kautz, Eric Horvitz, Yongshao Ruan, Carla Gomes,and Bart SelmanUsing Weighted MAX-SAT Engines to Solve MPEJames D. ParkEasy Predictions for the Easy-Hard-Easy TransitionAndrew J. ParkesThe Interface between P and NP: COL, XOR, NAE, 1-in-k, andHorn SATToby Walsh

Web and Information ExtractionA Maximum Entropy Approach to Information Extraction fromSemi-Structured and Free TextHai Leong Chieu and Hwee Tou NgReviewing the Design of DAML+OIL: An Ontology Language forthe Semantic WebIan Horrocks, Peter F. Patel-Schneider, and Frank van HarmelenStochastic Link and Group DetectionJeremy Kubica, Andrew Moore, Jeff Schneider, and Yiming Yang

Reducing Search Space in Local Search for Constraint SatisfactionH. Fang, Y. Kilani, J.H.M. Lee, and P.J. StuckeyOptimal Schedules for Parallelizing Anytime Algorithms: TheCase of Independent ProcessesLev Finkelstein, Shaul Markovich, and Ehud RivlinOptimal Depth-First Strategies for And-Or TreesRussell Greiner, Ryan Hayward, and Michael MolloyA New Algorithm for Optimal Bin PackingRichard E. KorfMemory-Efficient A* Heuristics for Multiple Sequence AlignmentMatthew McNaughton, Paul Lu, Jonathan Schaeffer, and DuaneSzafronPROMPTDIFF: A Fixed-Point Algorithm for Comparing Ontology VersionsNatalya F. Noy and Mark A. MusenOn Preference-Based Search in State Space GraphsPatrice Perny and Olivier SpanjaardAn Average-Case Analysis of Graph SearchAnup K. Sen, Amitava Bagchi, and Weixiong Zhang

Knowledge Representation IOn Computing all Abductive ExplanationsThomas Eiter and Kazuhisa MakinoMeasuring Inconsistency in Knowledge via Quasi-Classical ModelsAnthony HunterA Hoare-Style Proof System for Robot ProgramsYongmei LiuRepresenting and Reasoning about Mappings between Domain ModelsJayant Madhavan, Philip A. Bernstein, Pedro Domingos,and Alon Y. HalevyA Regression Based Adaptation Strategy for Case-Based ReasoningDavid Patterson, Niall Rooney, and Mykola GalushkaCluster Ensembles – A Knowledge Reuse Framework for Combining PartitioningsAlexander Strehl and Joydeep Ghosh

Knowledge Representation II:Logic ProgrammingLogic Programming with Ordered DisjunctionGerhard BrewkaA Three-Valued Characterization for Strong Equivalence of LogicProgramsPedro CabalarASSAT: Computing Answer Sets of a Logic Program by SATSolversFangzhen Lin and Yuting Zhao

LearningState Abstraction for Programmable Reinforcement LearningAgentsDavid Andre and Stuart J. RussellProgressive Rademacher SamplingTapio Elomaa and Matti KääriäinenLearning Temporal, Relational, Force-Dynamic Event Definitionsfrom VideoAlan Fern, Jeffrey Mark Siskind, and Robert GivanMinimum Majority Classification and BoostingPhilip M. LongContent-Boosted Collaborative Filtering for Improved RecommendationsPrem Melville, Raymond J. Mooney, and Ramadass NagarajanConstructive Adaptive User Interfaces – Composing Music Based on Human FeelingsMasayuki Numao, Shoichi Takagi, and Keisuke NakamuraConstrained Formulations and Algorithms for Stock-Price Predictions Using Recurrent FIR Neural NetworksBenjamin W. Wah and Minglun QianRule-Based Anomaly Pattern Detection for Detecting DiseaseOutbreaksWeng-Keen Wong, Andrew Moore, Gregory Cooper, and MichaelWagnerExtended Isomap for Pattern ClassificationMing-Hsuan Yang

Markov Decision ProcessesSegmenting Time Series with a Hybrid Neural Networks —Hidden Markov ModelLaura Firoiu and Paul R. CohenOn Policy Iteration as a Newton’s Method and Polynomial PolicyIteration AlgorithmsOmid MadaniEfficient Utility Functions for Ceteris Paribus PreferencesMichael McGeachie and Jon DoylePiecewise Linear Value Function Approximation for Factored MDPsPascal Poupart, Craig Boutilier, Relu Patrascu, and Dale SchuurmansBayesian Networks for Speech and Image Integration

Technical Poster Presentations

17

18

ExhibitionThe exhibition will be held in Exhibit Hall ABon the assembly level of the Shaw ConferenceCentre, Tuesday, July 30 and Wednesday, July31. Admittance is restricted to badged confer-ence attendees. Vendor-issued guest passesmust be redeemed at the Exhibitor Registra-tion Desk, in the registration area, on the as-sembly level of the Shaw Conference Centre.Further information regarding access to theExhibition can be obtained from the ExhibitorRegistration Desk. Guest tickets to the AI Fes-tival can be purchased for $25.00 CDN perperson ($8.00 CDN for children) at onsite reg-istration.

Exhibit HoursTuesday, July 30: 10:00 AM – 5:30 PM

Wednesday, July 31: 10:00 AM – 3:00 PM

3:00 PM – 5:30 PM (AI Festival)

Exhibitors■ AAAI Press■ ActivMedia Robotics■ AI Topics■ Applied AI Systems, Inc.■ Elsevier Science■ Franz Inc.■ iCORE■ iRobot Corp., Research Robots Division■ Kluwer Academic Publishers■ KurzweilAI.net■ The MIT Press■ Morgan Kaufmann Publishers■ NASA Ames Research Center■ Sony Electronics Inc., Entertainment

Robotic America■ Springer-Verlag New York, Inc.

Booth #100

AAAI Press445 Burgess DriveMenlo Park, CA 94025-3442Tel: 650-328-3123Fax: 650-321-4457Email: info@aaai.orgOnline Catalog: www.aaaipress.org/

AAAI Press is the only press headed by prac-ticing AI scientists. We are pleased to offer a20% discount to conference attendees.

Booth #208

ActivMedia Robotics44 Concord StreetPeterborough, NH 03458Tel: 603-924-9100

Email: robots@activmedia.comWeb: www.activmedia.com

ActivMedia Robots team of compatible sys-tems offers various scales, terrain capabilities,payloads and human interfaces, used by thou-sands of roboticists around the globe. Thesecross-Linux/WIN platforms feature powerfulARIA developer software plus integrated map-ping, Saphira gradient navigation, scheduling,and multitude of plug-n-play accessories. Wetake care of the basics so you can focus onwhat’s new and exciting in your research. Ac-tivMedia Robots accessories include inertialcorrection, integrated laser mapping and nav-igation, active PTZ vision, color-tracking andsurveillance, GPS, tilt-skew, bumpers, grip-pers, fixed IR, arms, active PTZ stereovision,sonar and more. Come see us in action!

Booth # 203

AI TOPICS — the AAAI PathfinderWebmaster: Jon GlickWebsite: www.aaai.org/aitopicsE-mail: aitopics@aaai.org

AI TOPICS has something for everyone. Hav-ing started as a dynamic online library spon-sored by the American Association for Artifi-cial Intelligence for students, teachers, jour-nalists, and others in need of basic, introduc-tory information about what artificial intelli-gence is and what AI scientists do, our recentexpansion of the “AI in the news” feature hasbroadened the site’s scope such that everyonewho is interested in the exciting field of artifi-cial intelligence now has a great reason to vis-it AI TOPICS. And because we are dedicatedto serving you, we hope that you take advan-tage of this opportunity to stop by Booth #203to take a guided tour of the website, to shareyour comments, concerns, ideas and sugges-tions, or to just catch up on the latest AI news.

Booth #309

Applied AI Systems, Inc.3232 Carp Road, RR 2Ontario, Canada K0A 1L0

Applied AI Systems, Inc. (AAI) will be demon-strating Khepera II, Koala, and GAIA robotsand Webots simulation software, along withmany options and accessories. Discuss newfeatures of recently released Khepera II withknowledgeable engineers who are users of therobots. Videos of other larger robots and theirapplications such as transportation for the el-derly, intelligent wheelchairs for the disabled,movement of heavy materials autonomouslyin construction or farming industry, and intel-ligent camera to detect and warn of dangerous

Exhib

it Prog

ram

19

situations (e.g., landslide or illegal behavior(e.g., garbage dumping) will be shown).

Booth #114

Elsevier Science655 Avenue of the AmericasNew York, NY 10010

Elsevier Science is the market leader of scien-tific publications. All books and journals inthe field of artificial intelligence, expert sys-tems and knowledge-based systems will be ondisplay. Also available from our booth, someof Elsevier’s electronic initiatives including theComputer Science Preprint Server (www.computersciencepreprints.com)—the perma-nent web archive and rapid distribution medi-um for research articles in the field of com-puter science—and the new to launch Com-puter Science Portal (www.computer-scienceweb.org). ComputerScienceWeb offersfull text articles from leading computer sci-ence journals and integrated services to facili-tate research in all fields of computer science.

Booth #221

Franz Inc.1995 University Avenue Ste #275Berkeley, CA 94704

Franz Inc. produces Allegro CL® 6.2, a com-plete, cross-platform development environ-ment powered by Common Lisp/CLOS. Alle-gro CL’s dynamic object-oriented technologyallows developers to create leading edge, mis-sion-critical applications that are robust,scaleable, and easy to evolve and deploy. Alle-gro CL is ideal for Dynamic Servers, Manufac-turing scheduling and control, IC design &synthesis, Knowledge Management and DataMining. Other Franz Inc. products include Al-legroServe®, a dynamic web-enabling LispWeb Server, Allegro ORBLink®, a CORBA-compliant ORB, and AllegroStore®, a persis-tent object database.

Booth #106

iCOREContact: Mary Anne Moser3608 33 Street NWCalgary, AB, T2L 2A6 CanadaTel: 403-949-3306Fax: 403-949-3320email: moser@netera.ca

The Informatics Circle of Research Excellence(iCORE) was created three years ago to devel-op and support excellent university-based re-search in information and communicationstechnology (ICT) in Alberta. Its goal is to po-

sition Alberta as a leader in information andcommunications technology by establishingand funding top research teams.

Ten new exceptional chairs and their re-search teams are well under way. iCORE in-tends to fund the establishment of over adozen more new chairs at Alberta universitiesover the next several years. iCORE also sup-ports nearly 90 ICT-related graduate studentsat universities around Alberta through a sepa-rate scholarship program.

Booth #207

iRobot Corporation,Research Robots Division37 Wilton RoadMilford, NH 03055603-654-3400www.irobot.com

Inspire. Encourage. Enable.iRobot’s team of dedicated engineers, softwaredevelopers and production specialists em-brace creativity innovative thinking and cut-ting-edge technology to design and build agrowing family of versatile, rugged, fully inte-grated mobile robot systems. The revolution-ary Mobility™ Robot Integration Softwareand rFLEX™ Robot Control Architecturedrive the entire family of research robot plat-forms, providing seamless, top-to-bottom in-tegration along with clear, intuitive migrationpaths among platforms. iRobot’s mission is toprovide robot development tools that inspire,encourage and enable advances in robotics.Stop by to see these great red machines in ac-tion.

Booth #108

Kluwer Academic Publishers101 Philip DriveNorwell, MA 02061

Kluwer Academic Publishers, a leading pub-lisher of scientific books and journals, invitesattendees to visit our display and browsethrough our latest publications. Pick up asample copy of our wide array of artificial in-telligence journals including Machine Learn-ing, Autonomous Robots, and Data Mining andKnowledge Discovery. Also, receive a 20% dis-count on all Kluwer titles on display. For moredetails on all our publications visit www.wkap.nl

Booth #120

KurzweilAI.net15 Walnut St.Suite 200Wellesley Hills, MA 02481

Exhibit Program

20

KurzweilAI.net is the web home of today’s bigthinkers and newsmakers examining the con-fluence of accelerating revolutions shapingour future world and the inside story on newtechnological and social realities. It focuses es-pecially on the exponential growth of intelli-gence, both biological and machine, and themerger of the two in a post-humanist future.Visit us on the web at KurzweilAI.net

Booth #102

The MIT PressFive Cambridge CenterCambridge, MA 02142Tel: 800-56-0343Fax: 617-253-1709Email: mitpress-orders@mit.eduWeb: mitpress.mit.edu

MIT Press is one of the leading publishers inthe field of artificial intelligence. We arepleased to offer a wide range of our recentpublications at a conference discount of 20%.

Booth #112

Morgan Kaufmann PublishersAn Imprint of Elsevier Science and TechnologyBooks 340 Pine Street, 6th FloorSan Francisco, CA 94104Tel: 415-392-2665Fax: 415-982-2665www.mkp.com

Morgan Kaufmann is dedicated to publishingdistinguished books for artificial intelligenceresearchers and students. Our list of AI publi-cations reflects the classical disciplines withinAI including representation and reasoning,machine learning, natural language process-ing, intelligent systems, computer vision,planning and uncertainty as well as emergingareas such as agents, data mining and evolu-tionary computation. We publish textbooks,surveys of areas and methods, programmingguides, and select research volumes and pro-ceedings. New, just in time for AAAI 2002, Ex-ploring Artificial Intelligence in the New Millen-nium.

Booth #200

NASA Ames Research CenterContact: Robert DuffyM/S 269-1Moffett Field, CA 94035-1000Tel: 650-604-3591

The Information Sciences Directorate ofNASA Ames Research Center is a world-classcreator of innovative, intelligent, high perfor-mance and reliable information technologiesfor NASA’s space and aeronautics missions.

The NASA booth features information on thedevelopment of autonomous technologies forrobotic spacecraft and an overview of NASA’sIntelligent Systems Program. The IS Programdirects research in automated reasoning, hu-man-centered computing and intelligent dataunderstanding that will lead to the advancedcomputer science and information technolo-gies essential to NASA’s missions of explo-ration in the 21st Century. The NASA boothwill also feature a demonstration of the Per-sonal Rover Project, a collaborative effort be-tween NASA and Carnegie Mellon University.

Booth #215

Sony Electronics Inc.,Entertainment Robotic AmericaContact: John DeCuir16450 West Bernardo DriveSan Diego, CA 92127Tel: 858-942-3818Fax: 858-942-9080

Sony is pleased to offer the OPEN-R™ SDK(Software Development Kit) for AIBO® Enter-tainment Robots. OPEN-R is the standard in-terface for Sony’s entertainment robot system.This interface greatly expands the capabilitiesof entertainment robots. The OPEN-R SDK isa cross-development environment based onGCC (C++) which you can use to make non-commercial software applications for AIBO(ERS-210, ERS-220, ERS-210A, and ERS-220A). The tools are free of charge! Please vis-it us at www.aibo.com/ openr/ to access tools,documentation, and a bulletin board whereyou can collaborate with other developers.

Booth #213

Springer-Verlag New York, Inc.175 Fifth AvenueNew York, New York 10010 USATel: 800-777-4643custserv@springer-ny.comwww.springer-ny.com

Save 20% on artificial intelligence books fromSpringer-Verlag! From the Lecture Notes inComputer Science series to the upcoming titleAgent Technology, Springer has established areputation for publishing essential books andresources in artificial intelligence and agents.

Exhib

it Prog

ram

21

Intelligent Systems DemonstrationsThe Intelligent Systems Demonstrations willbe held in Exhibit Hall AB of the Shaw Con-ference Centre and will be open to registeredconference attendees during exhibit hours.The IS demos program returns to AAAI-02 forits fourth year. Continuing advances in artifi-cial intelligence research are making it possibleto develop intelligent systems in a wide rangeof application areas. The AAAI-02 IntelligentSystems Demonstrations program showcasesstate-of-the-art AI implementations and pro-vides AI researchers with an opportunity toshow their research in action.

The program is intended to highlight inno-vative contributions to the science of AI withan emphasis on the benefits to be gained fromdeveloping and using implemented systems inAI research. Previous year’s demonstrationshave included speech- and gesture-based sys-tems, AI-based simulators and game-playingsystems, several systems using AI on the worldwide web for e-commerce and other applica-tions, and even AI pets. System builders will beon hand to present their work, and audienceinteraction with the systems is encouraged asmuch as possible.

All demonstrations will be available duringthe AI Festival on Wednesday afternoon, andthe tentative individual demonstration sched-ule for Tuesday and Wednesday is listed below.

Demonstrations Schedule

Tuesday, July 3010:00 AM: JYAG & IDEY: A Template-Based

Generator and Its Authoring Tool (BoothD314). Songsak Channarukul, Susan W.McRoy, and Syed S. Ali

11:00 AM: Constructive Adaptive User Inter-faces – Composing Music Based on HumanFeelings (Booth D417). Masayuki Numao,Shoichi Takagi, and Keisuke Nakamura

1:00 PM: An Automated Negotiator for an In-ternational Crisis (Booth D320). PeninaHoz-Weiss, Sarit Kraus, Jonathan Wilkenfeldand Tara E. Santmire

2:00 PM: SpeechWeb: A Web of Natural-Lan-guage Speech Applications (Booth D419).Richard Frost

3:00 PM: UTTSExam: A University Examina-tion Timetable Scheduler (Booth D312).Andrew Lim, Juay-Chin Ang, Wee-Kit Ho,and Wee-Chong Oon

4:00 PM: Disciple-RKF/COG: Agent Teachingby Subject Matter (Booth 413) GheorgheTecuci

Wednesday, July 3110:00 AM: Multi-ViewPoint Clustering Analy-

sis Tool (Booth D318) Mala Mehrotra Pra-gati

11:00 AM: FlexBot, Groo, Patton and Ghost-Bot: Research using Computer Games as aPlatform (Booth D316) Aaron Khoo, RobinHunicke, Greg Dunham, Nick Trienens, andMuon Van

1:00 PM: Fuzzy Neural Networks in a Palm En-vironment (Booth D421) Samuel Moyleand Michael Watts

2:00 PM: Research Applications of the MAG-NET Multi-Agent Contracting Testbed(Booth D415) John Collins and Maria Gini

3:00 PM: AI Festival (all demos available)

Demo Descriptions

Booth D320

An Automated Negotiator for an International Crisis Penina Hoz-Weiss, Sarit Kraus, Jonathan Wilkenfeldand Tara E. Santmire, Bar-Ilan University, RamatGan, Israel and University of Maryland, College Park

We present an automated agent that can nego-tiate efficiently with humans in bilateral nego-tiations with time constraints and the possi-bility of opting out. The negotiation is con-ducted using a semiformal language. Themodel used in constructing the agent is basedon a formal analysis of the scenario usinggame theoretic methods and heuristics for ar-gumentation. The agent receives messages sentby humans, analyzes them and responds. It al-so initiates discussion on one or more param-eters of an agreement. The specific scenarioconcerns a crisis between Spain and Canadaover access to a fishery.

The human players are provided with aDSS to analyze the scenario and compare theutility points associated with various out-comes, and with a language editor to facilitatethe composition of messages. In the demo, wewill present the support tools for the humansand then demonstrate a negotiation sessionbetween the agent and a human. Audiencemembers will have an opportunity to negoti-ate with the agent.

Intelligent Systems Demos

22

Booth D417

Constructive Adaptive User Inter-faces – Composing Music Based onHuman Feelings Masayuki Numao, Shoichi Takagi, and KeisukeNakamura, Department of Computer Science,Tokyo Institute of Technology

We demonstrate a method to locate relationsand constraints between a music score and itsimpressions, by which we show that machinelearning techniques may provide a powerfultool for composing music and analyzing hu-man feelings. The demonstration introducestwo user interfaces, which are capable of pre-dicting feelings and creating new objects basedon seed structures, such as spectrums and theirtransition for sounds that have been extractedand are perceived as favorable by the test sub-ject. The interfaces first collect a person’s feel-ings for some pieces, based on which they ex-tract a common musical structure causing aspecific feeling. One arranges an existing songto fit such a structure causing a specified feel-ing. Another composes a new piece.

Booth D413

Disciple-RKF/COG: Agent Teachingby Subject MatterGheorghe Tecuci, Learning Agents Lab, George MasonUniversity

Disciple-RKF/COG is a learning agent shellthat can perform many knowledge engineeringtasks, and can be used to develop knowledgebased systems by subject matter experts, withlimited assistance from knowledge engineers.The expert and the agent engage into a mixed-initiative reasoning process during which theexpert is teaching the agent his problem solvingexpertise, and the agent learns from the expert,building, verifying, and improving its knowl-edge base.

The demonstration will show how a trainedDisciple agent helps the students at the USArmy War College to learn about Center ofGravity analysis. Then, the main part of the de-mo will show how this Disciple agent wastaught the problem solving expertise of a mili-tary expert. The demonstration will concludewith a presentation of several ontology importand development tools of Disciple-RKF/COGthat are used by a knowledge engineer to per-form knowledge base development tasks thatare currently beyond the capabilities of a sub-ject matter expert.

Booth D316

FlexBot, Groo, Patton and GhostBot:Research using Computer Games as aPlatform Aaron Khoo, Robin Hunicke, Greg Dunham, NickTrienens, and Muon Van, Northwestern University

This is a demonstration of FlexBot, a researchplatform built using the Half-Life game en-gine, and three systems based on the FlexBotarchitecture. Groo, our resident Half-Life bot,uses behavior-based techniques to make criti-cal decisions in real time. Patton is a recentlyconstructed system for monitoring and con-trolling bot teams in Half-Life using remotedevices such as PDAs. GhostBot is part of a sys-tem designed to monitor a player’s behavior ina game and adjust the environment to facilitatea smooth, yet challenging play experience.

The demonstration is designed to showFlexBot in action and to exhibit the flexibility,efficiency and overall ease with which theFlexBot architecture supports a variety of AIresearch tasks. During the demo, conferenceattendees will be able to observe FlexBot agentsin action, as well as play games against them.

Booth D421

Fuzzy Neural Networks in a Palm EnvironmentSamuel Moyle and Michael Watts, Department ofInformation Science, University of Otago

Booth D314

JYAG & IDEY: A Template-BasedGenerator and Its Authoring Tool Songsak Channarukul, Susan W. McRoy, and Syed S.Ali, University of Wisconsin-Milwaukee

JYAG is the Java implementation of a real-time, general-purpose, template-based gener-ation system (YAG, Yet Another Generator).JYAG enables interactive applications to adaptnatural language output to the interactivecontext without requiring developers to writeall possible output strings ahead of time or toembed extensive knowledge of the grammarof the target language in the application. Cur-rently, designers of interactive systems whomight wish to include dynamically generatedtext face a number of barriers; for example de-signers must decide (1) How hard will it be tolink the application to the generator? (2) Willthe generator be fast enough? (3) How muchlinguistic information will the applicationneed to provide in order to get reasonablequality output? (4) How much effort will berequired to write a generation grammar thatcovers all the potential outputs of the applica-

Intell

igent

Syste

ms De

mos

23

tion? The design and implementation of ourtemplate-based generation system, JYAG, is in-tended to address each of these concerns. Atemplate-based approach to text realization re-quires an application developer to define tem-plates to be used at generation time; therefore,the tasks of authoring and testing templates areinevitable. JYAG provides predefined tem-plates. Developers may also define their tem-plates to fit the requirements of a domain-spe-cific application. Those templates might be to-tally new or they can be a variation of existingtemplates. Even though developers can authora template by manually editing its textual defi-nition in a text file, it is more convenient andefficient if they can perform such tasks in agraphical, integrated development environ-ment. The integrated development environ-ment for YAG (IDEY) provides these servicesas a tool for JYAG’s templates authoring, test-ing, and managing. IDEY’s graphical interfacereduces the amount of time needed for syntaxfamiliarization through direct manipulationand template visualization. It allows a develop-er to test newly constructed templates easily.

Booth D318

Multi-ViewPoint Clustering Analysis ToolMala Mehrotra Pragati, Synergetic Research Inc.

The multi-viewpoint clustering analysis (MVP-CA) technology uses clustering analysis tech-niques to group rules of a knowledge base thatshare significant common properties. We dem-onstrate a research prototype tool that enablesknowledge engineers and subject matter experts(SMEs) to familiarize themselves rapidly withthe terms and concepts in a knowledge base, toexploit and reuse preexisting knowledge, and tomerge and align concepts across differentknowledge bases reliably and efficiently.

Booth D415

Research Applications of the MAG-NET Multi-Agent Contracting Testbed John Collins and Maria Gini, University of Min-nesota

MAGNET is a testbed for exploring decisionprocesses and agent interactions in the domainof multi-agent contracting. Experimental re-search in this area requires a simulation envi-ronment that is sufficiently rich to be easilyadapted to a variety of experimental purposes,while being sufficiently straightforward to sup-port clear conclusions. Two different demon-strations will be available. One uses a user inter-face to help visualize the domain and the agent’s

decision processes. The other shows how theopen-source MAGNET system can be config-ured to support experimentation with delibera-tion scheduling and winner determination.

Booth D419

SpeechWeb: A Web of Natural-Lan-guage Speech ApplicationsRichard Frost, Department of Computer Science,University of Windsor, Canada

The demonstration shows an enhanced natu-ral-language speech browser navigating a webof remote hyperlinked applications. Thebrowser recognizes relatively complex spokeninput such as “which moon that was discov-ered by Hall does not orbit mars?” and sendssuch input to a remote interpreter that is ac-cessed through the Internet. Users can ask tomove from one application to another in amanner that is analogous to following a hy-perlink on a regular web page. The browserruns on a regular PC and uses off-the-shelfIBM ViaVoice speech-recognition software.The input is spoken through a wireless micro-phone giving hands-free, eyes-free, access toremote data sources. The demonstration illus-trates application of a new semantics for nat-ural-language processing that accommodatesarbitrarily-nested quantification and nega-tion, and also a new technique for improvingspeech-recognition accuracy.

Booth D312

UTTSExam: A University Examination Timetable Scheduler Andrew Lim, Juay-Chin Ang, Wee-Kit Ho, Wee-Chong Oon, School of Computing, National Uni-versity of Singapore

UTTSExam is a university examinationtimetable-scheduling program, customized forthe National University of Singapore (NUS).The display comprises informational postersand two notebooks running the Registrar andFaculty versions of UTTSExam respectively.

The demonstration begins with an expla-nation of the NUS timetabling problem, in-cluding: information on NUS; statistics of theactual data; scheduling strategy and the rea-soning behind it; the combined methodscheduling algorithm. This is followed by ademonstration of the software and schedulingprocess, including: scheduling of the facultytimetables; merging of the faculty timetables;conflict resolution. The demonstration systemswill be loaded with actual data from semester 1of the 2001/2002 academic year in NUS.

Intelligent Systems Demos

24

Robot Competition & ExhibitionThe Eleventh Robot Competition and Exhibi-tion will be held in Exhibit Hall AB on the as-sembly level of the Shaw Conference Centre,and will be open to registered conference at-tendees during exhibit hours. This series ofevents brings together teams from colleges,universities and other research laboratories tocompete, and also to demonstrate state-of-the-art research in robotics and AI. The goalsof the Competition and Exhibition are to (1)foster the sharing of research ideas and tech-nology; (2) allow research groups to showcasetheir achievements; (3) encourage students toenter the fields of robotics and AI; and (4) in-crease awareness of the field.

CompetitionThe competition allows teams to show offtheir best attempts at solving common tasks ina competitive environment. Teams competefor place awards as well as for technical inno-vation awards, which reward particularly in-teresting solutions to problems. There will bethree contest events this year: Robot Host,Robot Rescue, and the Robot Challenge.

ExhibitionThe exhibition gives researchers an opportu-nity to demonstrate state-of-the-art researchin a less structured environment. Exhibits arescheduled throughout the exhibition hallhours.

WorkshopThe robot events culminate with a workshopwhere participants describe the research be-hind their entries.

Schedule

Tuesday, July 3010:00 AM – 5:30 PM

■ Robot Challenge: Throughout Shaw Con-ference Centre

■ Robot Exhibition: Exhibit Hall AB■ Robot Rescue: Exhibit Hall AB■ Robot Host: Information Kiosk, during

breaks outside Exhibit Hall AB

Wednesday, July 3110:00 AM – 5:30 PM

■ Robot Rescue: Exhibit Hall AB

■ Robot Exhibition: Exhibit Hall AB11:55 AM - 12:45 PM

■ AAAI Invited Panel on the Robot Competition and Exhibition

3:00 PM - 5:30 PM

■ Robot Host: Robot Server, during AI Festival

5:00 PM - 5:30 PM

■ Awards Ceremony: Exhibit Hall AB,during AI Festival

Thursday, August 19:00 AM – 5:00 PM

■ Robot Workshop (by invitation only),Salon 5, Meeting Level

Robot Competition and Exhibition Teams

Joint Challenge Competitors

Carnegie Mellon University, NavalResearch Laboratory, Metrica, North-western University, and SwarthmoreCollegeRobot: GRACECMU Team: Reid Simmons, Greg Armstrong, Alli-son Bruce, Dani Goldberg, Adam Goode, IllahNourbakhsh, Nicholas Roy, Brennan Sellner, DavidSilver, Chris UrmsonNRL Team: Alan Schultz, Myriam Abramson,William Adams, Amin Atrash, Magda Bugajska,Mike Coblenz, Dennis PerzanowskiMetrica Team: David Kortenkamp, Bryn WolfeNorthwestern Team: Ian Horswill, Robert ZubekSwarthmore Team: Bruce Maxwell

GRACE (Graduate Robot Attending Confer-encE) is a multi-institutional, cooperative ef-fort consisting of Carnegie Mellon University,the Naval Research Laboratory, Metrica,Northwestern University, and SwarthmoreCollege. This year’s goal is to integrate soft-ware from the various institutions onto acommon hardware platform and attempt todo the complete AAAI Robot Challenge taskautonomously, from beginning to end. The fo-cus will be on multi-modal human robot in-teraction (speech and gesture), human-robotsocial interaction, task-level control in the faceof a dynamic and uncertain environment,map-based navigation, and vision-based in-teraction.

Interacting naturally with humans, GRACEwill find its way from the convention entranceto the registration area. It will query by-standers for directions to the registration deskand navigate there based on those directions.Along the way, it will interact with other con-

Robo

t Prog

rams &

Sche

dule

25

ferees and will ride in the elevator, using anelectronic altimeter to determine when it is onthe right floor. It will use color vision to findthe registration sign, and use laser and stereovision to queue itself and wait in line. It willinteract with the volunteer at the registrationbooth, and use map-based navigation to trav-el to the Exhibition Hall. Finally, it will presenta talk about itself (see Program addendum fortime and place).

Exhibitor

Carnegie Mellon University(Robotics Institute)Robot: The Personal Rover ProjectTeam Leader: Illah NourbakhshTeam Members: Emily Falcone, Rachel Gockley, EricPorter, Illah Nourbakhsh

The Personal Rover Project, funded byNASA/Ames Autonomy programs, aims to de-velop an affordable, highly competent mobilerover that will serve as a scientific and creativeoutlet for children. This rover is novel in itsuse of leading-edge microprocessor technolo-gy to enable very high robot competence atvery low price point. An on-board CMUcamvision system tracks colorful objects using justa Ubicom microprocessor and a CMOS imag-ing sensor. A network of on-board PIC micro-processors, called Cerebellum, provide sensorinterfaces as well as fine-grained motor con-trol using back-EMF speed sensing. An on-board StrongARM processor provides 802.11bnetworking as well as time-critical visual-mo-tor feedback loops. Finally, using a movableCenter of Mass mechanism, the PersonalRover is able to traverse steps greatly exceed-ing its wheel diameter.

During AAAI we plan to demonstrate boththe step traversal and vision-based navigationcompetencies of the Personal Rover. In addi-tion to achieving these technological compe-tencies, the Personal Rover Project also focus-es on interaction design in order to produce acompelling human-robot interface through aseries of formative evaluations. We will alsodemonstrate such interaction interfaces,painting a picture of how the Rover can be atool for conducting simple science experi-ments in and around the home.

Rescue Competitor

Carnegie Mellon University(Robotics Club)Robot: Tartan Swarm

Tartan Swarm is a low-cost multirobot ap-proach to human detection. Each robot is

based on a simple modular diff-drive plat-form, mounted with a heterogeneous array ofsensors. Sensor types include both vision andpyroelectric sensing. Successful human detec-tion is communicated through two channels: alow-bandwidth channel to ward off neighbor-ing robots; and a coded radio broadcast, indi-cating success and believed relative location.This signal is received by a rescue workstation.Individual robots accomplish their tasks au-tonomously, using distinct search strategies.Collective behavior is observed through sim-ple success-based interactions.

Tartan Swarm is a simple, low-cost, educa-tional project of the undergraduate CarnegieMellon Robotics Club

Exhibitor

Columbia UniversityRobot: RoboCupJuniorTeam Leader: Elizabeth Sklar Team Member: Simon Parsons

RoboCupJunior is a project-oriented educa-tional initiative that sponsors local, regionaland international robotic events for students.This year marks the third year of internation-al competition, with a tournament being heldin conjunction with RoboCup 2002 –RoboCupJunior 2002 will include over 60teams of high school and middle school stu-dents from more than a dozen countriesworld-wide. Teams build and program au-tonomous mobile robots to play soccer, per-form dances and simulate rescue scenarios.

We have also used the RoboCupJunior mo-tif as the theme for undergraduate classes inAI, robotics and programming. The soccerand rescue contests have been extremely moti-vating and challenging for college studentswith a variety of backgrounds and skill levels.

Our exhibition will introduce RoboCupJu-nior to the AAAI audience, in search of men-tors for teams of young students as well as ed-ucators looking for a new twist on standardundergraduate curriculum.

Exhibitor

Dartmouth UniversityRobot: Crystal RobotsTeam Leaders: Robert Fitch and Daniela Rus

A robot designed for a single purpose can per-form some specific task very well, but it willperform poorly on a different task, in a differ-ent environment. This is acceptable if the en-vironment is structured; however if the task isin an unknown environment, then a robotwith the ability to change its shape to suit theenvironment and the required functionality

Robot Teams

26

will be more likely to succeed than a fixed-ar-chitecture robot. We wish to create more ver-satile robots by using self-reconfiguration:hundreds of small modules will autonomous-ly organize and reorganize as geometric struc-tures to best fit the terrain on which the robothas to move, the shape of the object the robothas to manipulate, or the sensing needs for thegiven task. For example, the robot could syn-thesize a snake shape to travel through a nar-row tunnel, and then morph into a six-leggedinsect to navigate on rough terrain upon exit.

Self-reconfiguring robots are well-suited fortasks in hazardous and remote environments,especially when the environmental model andthe task specifications are uncertain. A collec-tion of simple, modular robots endowed withself-reconfiguration capabilities can conformto the shape of the terrain for locomotion byimplementing “water-flow” like locomotiongaits, which allow the robots to move by con-forming to the shape of the terrain. We havedesigned and built the Crystal robot—which iscapable of autonomous shape changing, loco-motion by self-reconfiguration, and self-repli-cation of a big robot into smaller robots—withthe same functionality.

Rescue Competitor

Georgia Technical UniversityRobots: Georgia Tech Yellow JacketsTeam Leader: Tucker Balch

Georgia Tech will compete in the Robot Rescuecompetition using a cooperative multi-robotsystem. The robots include an RWI ATRV Miniequipped with 8 DV cameras arranged to pro-vide omnidirectional vision, and four Sony AI-BO legged robots. The AIBOs are transportedby the ATRV, then released to explore areaswhere the ATRV cannot reach. The ATRV willprovide 3D modeling of the environment aswell as localization and tracking of the AIBOs.

Exhibitor

Idaho National Engineering and Environmental LaboratoryRobot: JuniorTeam Leader: David Bruemmer

INEEL is working to develop robots that canadjust their level of autonomy on the fly, lever-aging their own, intrinsic intelligence to meetwhatever level of control is handed downfrom the user(s). Currently, we have imple-mented a control architecture with four dif-ferent levels of autonomy: teleoperation, safe-mode, shared control and full autonomy. Eachlevel of control encompasses a different role

for the operator and makes different demandson the robot and communication infrastruc-ture. To meet this objective we are working to-wards the following technical goals: (1) slidingautonomy to support changing communica-tion, cognitive, perceptual and action capabil-ities of the user and robot; (2) self-monitoringand continuous assessment of perception andbehavior performance; (3) novel interfaces forrobust, mixed-initiative interaction betweenrobots and humans; (4) simultaneous local-ization and mapping techniques that capturean abstracted representation of the robot’s ex-perience vis-a-vis the task and environment;and (5) sensor-suites and fusion algorithms toenhance perception capabilities for sensing,interpreting, and “understanding” environ-mental features and humans.

Through these technical efforts, we will en-able remote robotic operations to be accom-plished by fewer operators with less training.This work will pave the way for a new class ofmixed-initiative robots that work with hu-mans as well as for them, accepting high-leveltasks with minimal demands on the user.

Challenge Competitor

iRobot CorporationRobot: CoWorkerTeam Members: Mark Dockser and Jim Allard

The CoWorker robot is an internet controlled,wireless, mobile, remote telepresence plat-form. CoWorker can be accessed from any PCweb browser with a high-speed connection(and secure ID and password). The user-friendly interface provides control over wherethe CoWorker goes, what it sees, what it hearsand provides an interface for speaking. Thereis even a laser pointer so that the user canhighlight what she is referring to at the robot’slocation. The platform was designed withmany available ports (power, serial, PCMCIA)for incorporation of additional hardware in-cluding sensors and/or additional cameras.

Our vision for the CoWorker is to meet theneeds of industrial users for: (1) remote ex-pert applications; (2) security; and (3) video-conferencing anywhere. By deploying Co-Workers, customers can dramatically reducetravel costs, allow for collaboration anywherein a company’s wireless network (includingthe shop floor, cafeteria or any other location),and improve worker safety (by deploying aCoWorker rather than a human to potentiallyhazardous situations). CoWorker robots arecurrently in beta applications with a numberof Fortune 500 companies.

Robo

t Tea

ms

27

Host Competitor

Kansas State UniversityRobot: BorivojTeam Members: Vojtech Derbek, Jan Kraus, TomasTichy, David A. Gustafson

Exhibitor

Kansas State UniversityRobots: Kansa, Wichita, Coronado and Pike Team Leader: Eric Matson

Robot teams have advantages over individualrobots in accomplishing goals that containlarge numbers of tasks. The advantage grows ifthey have the ability to interchange roles, shareresponsibility and have some redundant capa-bility. An organization which has the ability tocontinuously evaluate capabilities and role as-signments and reorganize to maximize effi-ciency will naturally operate at higher levels.Our research is to create a cooperativerobotics (CR) reorganization model to dy-namically evaluate and reorganize the team inthe event of a failure, or suboptimal executingcondition. We are currently building a modeland system to allow a team of heterogeneousrobots to conduct real-time reorganizationswhile working in a specific task environment.

Challenge Competitor

MITRobots: Leo and ErikTeam Leader: John Leonard

Our research addresses the problem of concur-rent mapping and localization (CML) for au-tonomous mobile robots. The problem of CMLis stated as follows: starting from an initial po-sition, a mobile robot travels through a se-quence of positions and obtains a set of sensormeasurements at each position. The goal is forthe mobile robot to process the sensor data toproduce an estimate of its position while con-currently building a map of the environment.While the problem of CML is deceptively easyto state, it presents many theoretical challenges.The problem is also of great practical impor-tance; if a robust, general-purpose solution toCML can be found, then many new applica-tions of mobile robotics will become possible.During the robot challenge, we will attempt ademonstration of an algorithm for real-timelarge-scale CML using multiple submaps

Rescue Competitor

The MITRE CorporationRobots: Moe, Larry, and CurlyTeam Members: Zach Eyler-Walker and David Smith

We are developing an approach to coordinat-

ed search using a team of robots controlled bya single human. The robots are semiau-tonomous and able to share information di-rectly with one another, and with a human viaa commander console program.

The robots will perform obstacle avoid-ance, localization, and low-level route plan-ning autonomously. Mapping, target detec-tion, and goal-directed behavior will be per-formed via coordination between robots andthe human commander.

We are currently using three ActivMediaPioneer 2-AT robots, each equipped withsonar and a single color camera. We expect toeventually integrate other platforms (e.g.iRobot Packbot) and sensors (e.g. laserrangefinder, pyrosensors, and microphones).

Rescue Competitor

New Roads High SchoolRobots: Morph Dragon, RingoTeam Members: The Scarabs, Michael Randall

In 1999, a group of high-school and junior-high students from Los Angeles took on theenormous challenge of competing againstsome of the top robotics and AI researchers inthe world in the RoboCup middle (F2000)league. After over two years of hard work,funded on a shoestring budget (mostly out-of-pocket), the Scarabs robotic team field-tested a color-tracking system at RoboCup2000 in Melbourne, Australia; designed andbuilt a prototype vehicle and omnidirectionalvision system; and successfully demonstratedthis vehicle-vision system combination in theAAAI/RoboCup Rescue Robot competition inSeattle, Washington.

The goals of the Scarabs team are to buildviable robots at minimal cost; to learn aboutmath, computer science, electronic engineer-ing, physics, AI, system integration, interna-tional relations, character development, andteamwork; to have fun (!); and to make a posi-tive difference.

We are fielding two radically differentrobots: Ringo, an updated version of the pro-totype we ran in 2001; and Morph-Dragon, asophisticated six-wheeled robot designed tocompete on the Robotica television program.Both robots will use the same vision and con-trol systems. We have upgraded our custom-built omnidirectional vision system with theAxis 2120 Network Camera (www.axis.com).The 2120 features direct connection with a10/100 MBit Ethernet network and a built-inLinux web server. This allows a single Ethernetcable for video and robot control.

Robot Teams

28

Exhibitor

Northwestern UniversityRobot: Rosey the Robot Team Leader: Christopher Dac Le

The RObot Self-Explains whY (ROSEY) sys-tem attempts to demonstrate a behavior-basedrobots ability to generate verbal explanationsin response to questions about its behavior.Specifically, the robot recognizes a class of whyquestions that seek reasons for its locomotivebehavior such as, “Why are you turning?”ROSEY the Robot will be running around theexhibit hall while fielding such questions,which will be typed in.

ROSEY the Robot represents an instance ofa class of what we call self-explanatory robots,which should be able to explain what they’redoing and why they’re doing it. To build suchrobots, we are exploring how a behavior-basedrobot can generate explanations by accessingits own internal structures and relating themto its sensory-motor state.

Rescue Competitor

Sharif University of TechnologyRobots: EMDAD II, Twins and Magellan ProTeam Leader: Dr. Amir Hossein JahangirTeam Members: S. Bayat Sarmadi, G. Asadi, S. Shar-ifi Tehrani, E. Mozaffaripour, M. Sonee, A. Farahani,H. Khaleghi, N. Honarmand

EMDAD II, Twins and Magellan Pro are therobots used in the competition. These agentsare supervised by a central system. Each of therobots searches alone, and when it encountersa suspected object, sends an alarm or a mes-sage to the central system. In this case, de-pending on the received information, the cen-tral system will realize whether a victim is pre-sent or not, and the process will be kept on tillthe end of allowed period.

Yellow Arena: A commercial robot, Magel-lan Pro is used in this arena. The main chal-lenge has been to develop software capable ofmotion planning and visual navigation.

Orange Arena: Two robots, Twins, are con-sidered for this arena. The mechanical four-wheel platform is based on a mobile toy struc-ture. An industrial small size PC does the in-formation processing. Twins use ultrasonicsensors, IR sensors and IR cameras.

Red Arena: A special robot has been de-signed for this section namely EMDAD II. Thehardware platform of EMDAD II, is a reliablesystem based on industrial PC104 boards.This hardware system is capable of operatingin hostile environments.

Host and Rescue Competitor

Swarthmore CollegeRobots: Frodo & GollumTeam Leader: Bruce Maxwell

For the Host Competition, we hope to buildsome interesting human-robot interactionsupon our name-tag reading system that wedeveloped for the 2001 competition. We alsohope to have two moving robots this year sothat they can converse when they see one an-other. Our overall goal, however, will be tosuccessfully serve hors d’oeuvres in an unob-trusive and effective manner.

In the Rescue Competition we will be ex-amining issues in semi-autonomous robotsystems. Our goal is to permit a single user tosuccessfully manage more than one robot. Wewill build upon our system from a year agothat identified victims using vision, built mapswith paths to found victims, and enabled bothteleoperation and completely autonomousrobot functioning.

Rescue Competitor

Temasek PolytechnicRobots: Temasek Search and Rescue RobotTeam Leader: Nadir Ould-Khessal

Civil Defence Search and Rescue robot is anattempt to replace human rescue personnelwith tank-like vehicle capable of searching forvictims in an unstructured terrain. The di-mension of the robot is 502 mm in width, 760mm in length and 350 mm in height. The esti-mated weight of the vehicle is 75kg. The front-mounted camera can be rotated up to 360 de-gree. Civil Defence Search and Rescue Robot isequipped with a Charged Coupled Device(CCD) camera, 24 ultrasonic sonar sensors,wireless RF receiver, wireless audio receiver,and wireless video transmitter, as well as anonboard processing unit

Exhibitor

Universita di Roma “la Sapienza”Robot: ArmHandOneTeam Members: M. Cialente, A. Finzi, I. Mentuccia,F. Pirri, M. Pirrone, M. Romano, F. Savelli, K. Vona

We have a reconfigurable maze—that is, amaze made of panels that can be suitably ar-ranged. Inside this maze we can position sev-eral road signs indicating whether the road isone-way or no-entry, arrows addressing direc-tions for the exit, and so on. We are free to putthe road signs inside the maze, that is, theagent performance should not depend on thesigns. Furthermore there is a place in the mazein which we can locate a treasure, consisting of

Robo

t Tea

ms

29

Robot Teamsa set of colored blocks, suitably arranged: e.g.forming towers etc. The task for the agent is toget into the maze, find a particular block (e.g.the red block), picking it up (to achieve thisthe robot might need to move lots of otherblocks) and direct itself toward the exit and, fi-nally, exit from the maze.

The robot we are talking about is namedArmandone (pronouced “ArmHandOne”),weighing about 4 kg and standing about 40 cmhigh. It is equipped with a grabber arm, a pan-tilt binocular head with two cameras, and oth-er sensors. Wireless control can reach 1 km.The novelty of our approach mainly relies inthe cognitive architecture we have been build-ing. The architecture is defined on three levels:(1) the cognitive level monitoring sensing andhigh level control actions, (2) the global levelmanaging the choice of tasks and actions con-trol, and (3) the reactive level managing navi-gation and localization.

Exhibitor, Challenge and Host Competitor

Universidade de Aveiro, PortugalRobot: CarlTeam Leader: Luis Seabra LopesTeam Members: Luis Seabra Lopes, Antonio Teix-eira, Mario Rodrigues, Qinghua Wang and three un-dergraduate students

CARL is an acronym for “communication, ac-tion, reasoning and learning in robotics.” Theproject is therefore concerned with the devel-opment of an integrated set of capabilities forrobot intelligence. The participation will buildupon previously implemented capabilities(multisensor navigation, spoken-language di-alog), already demonstrated at the 2001 AAAICompetition. A higher degree of interactivitywill hopefully be demonstrated.

For the Challenge competition, Carl willperform only a part of the proposed task. Ca-pabilities developed for the participation ofCarl at the Host competition will also be usedhere. In addition, we may demonstrate on-linelearning.

Exhibitor

University of ConnecticutRobots: I Comici RobotiTeam Leader: Karl R. Wurst

Combining Robotics, Puppetry, and Comedy,our troupe of three robots perform a lazzofrom the Commedia Dell’Arte. The ItalianComedies of the 16th and 17th centuries hadmany improvisational pieces called lazzi.These were comic interludes inserted by aplayer if a scene started to drag or his elo-quence gave out. I Comici Roboti performs

the Lazzo of the Statue, in which Arlecchinopretends to be a statue who moves when thebacks of the other actors are turned.

Our troupe of three robots perform a shortscript, with each executing its own plan, cue-ing off the others to keep themselves in sync. Ahuman director observing the performancecan affect the overall performance, or the per-formance of an individual robot. The robotsthemselves consist of Lego bases, carrying aHandyBoard processor, a two-way radio link,and the puppet body.

Rescue Competitor

University of Manitoba, Winnipeg,CanadaRobots: Keystone Fire BrigadeTeam Leader: Jacky BaltesTeam Members: Jacky Baltes, John Anderson

The Keystone Fire Brigade robots are based onthe 4 Stooges, a small sized RoboCup teamfrom the University of Auckland. The robotsof the 4 Stooges were designed to be robustand versatile enough to be used in a variety ofdifferent ways. This has paid off since therobots of the Keystone Fire Brigade are identi-cal to those.

The Keystone Fire Brigade use a smallCMOS camera and Thomas Braunl’s Eyebotcontroller. The Eyebot controller consists of a35 MHz 68332 processor with 2 MB of staticRAM. The design is clearly dated nowadays,but has the advantage that it is comparativelycheap and provides the possibility of directlyconnecting a CMOS camera to the processoras well as providing the necessary interface toconnect motors, servos, gyroscopes, and manyother sensors directly to the controller.

Exhibitor

University of MinnesotaRobot: MinDART (Minnesota Distributed Au-tonomous Robotic System)Team Leader: Paul E. RybskiTeam Members: Paul E. Rybski, Amy Larson, ChrisFlowers-Huebner, Maria Gini

The Minnesota Distributed AutonomousRobot Team (MinDART) is a group of simpleand low-cost robots used at the University ofMinnesota for research into reactive controlstrategies. We are interested in studying howenvironmental and control factors affect theperformance of a homogeneous multi-robotteam doing a search and retrieval task. Severalfactors which affect the performance of theteam are examined. One factor is the distribu-tion of targets. This is varied from a uniformdistribution to having all of the targets clus-

30

Robo

t Tea

mstered together into one or two small clumps.Another factor that we examine is the size ofthe team (varying from one to five robots). Fi-nally, the type of search strategy is varied be-tween a completely reactive method to a direct-ed search method that uses the robot’s ability tolocalize itself. Current work includes incorpo-rating the ability for the LEGO robots to com-municate amongst themselves using an RF datalink and determining under what environmen-tal conditions such communications is useful.

Exhibitor, Host, and Rescue Competitor

University of RochesterRobot: Mabel the Mobile TableTeam Members: Undergraduate Robot ResearchTeam

We have developed a Java application whichwe call the “Learn Server.” Its purpose is toprovide offboard graphical user interfaces andparameter adjustment modules to offboardrobotics applications. It allows communica-tion with programs running in multiple lan-guages and on multiple operating systems.Currently the program communicates withthe C and C++ languages, in Windows2000,WindowsXP, and Linux. We will be demon-strating this application in the context of ourrobot host system: Mabel the Mobile Table.

Mabel gives an appropriate multi-modalresponse to people using a combination ofspeech, food manipulation, and navigationbehaviors. We accomplish this using theSphinx Speech Recognition System developedby CMU, augmented by a digital filter. We em-ploy a directed speech recognition micro-phone, which is actively pointed towards thespeaker, using face tracking and a pan-tilt-zoom camera. To accomplish language under-standing, a specially designed grammar-basedparsing technique is under development.

The vision component’s purpose is to pro-vide the navigation component with real-timevisual percepts including estimated closest pa-tron group, best approach angle in field ofview, patron face-region detection and track-ing, nametag-region detection scene, andcharacter classification (ANN)

Autonomous navigation control involvescreating a robust model for navigatingaround a crowded room while retaining theability to return to a base station. We usesonar-based obstacle avoidance for robustnavigation. To achieve path planning and ex-ecution, we employ a trained waypoint sys-tem using wheel counters.

Exhibitor

USC/Information Sciences InstituteRobot: CONROTeam Leaders: Wei-Min Shen and Behnam Salemi

The CONRO Project has a goal of providingthe Warfighter with a miniature reconfig-urable robot that can be tasked to perform re-connaissance and search and identificationtasks in urban, seashore and other field envi-ronments. CONRO will be miniature and is tobe made from identical modules that can beprogrammed to alter its topology in order torespond to environmental challenges such asobstacles. The base topology is simply con-nected, as in a snake, but the system can re-configure itself in order to grow a set of legs orother specialized appendages. Each modulewill consist of a CPU, some memory, a battery,and a micro-motor plus a variety of other sen-sors and functionality, including vision andwireless connection and docking sensors. Ma-jor challenges include packaging, power, andcooling, as well as the major issue of program-ming and program control.

Exhibitor

University of Texas at DallasRobot: Identity Emulation (IE), Facial ExpressionRobotRobot Team Members: David Hanson, MarshallThompson, Giovanni Pioggia

Our facial expression robot uses biomimeticstructures, aesthetic design principles, and re-cent breakthroughs in elastomer material sci-ences to enact a sizable range of natural hu-manlike facial expressions. This application ofrobotics will rise in relevance as humans androbots begin to have more face-to-face en-counters in the coming years. My team and Iare also working on imbuing our robot withseveral forms of interactive intelligence, in-cluding human-form and facial-expressionrecognition, and natural language interaction.It is anticipated that an integration of me-chanics, sociable intelligence, and design aes-thetics will yield the most effective humancomputer interface robots.

Rescue Competitor

Utah State UniversityRobots: Blue Swarm 2 and Blue Swarm SentinelTeam Members: Asti Bhatt, Brandon Boldt, Scott Sk-ousen, and Dan Stormont

The Blue Swarm 2 is made up of six modifiedremote-control cars. They operate au-tonomously using a simple subsumption ar-chitecture. They will sense the location of vic-tims and send out a signal which can be re-

31

ceived by one or more Blue Swarm Sentinels.The Blue Swarm Sentinel is a modified radio-controlled tank that operates either manuallyor autonomously to locate victims, locate BlueSwarm robots that have located a victim, orlocate obstacles. The Sentinel reports infor-mation back to the rescuer GUI via a bi-direc-tional RF link. Both types of robots are con-trolled by Parallax BASIC Stamps.

Exhibitor

Utah State UniversityRobot: Blue Swarm 3Team Leader: Dan Stormont

Blue Swarm 3 is the next generation of BlueSwarm robots. The Blue Swarm 3 will be builtto compete in the Urban Search and Rescuecompetition in 2003. The current plan is tobuild robust, legged robots that communicatewith each other and with a handheld terminal(PalmPilot) using IR links. The exhibit willdemonstrate some prototypes of the robotsthat will be developed for Blue Swarm 3.

Rescue Competitor

YSC (Iran)Robot: Hanif 1, Hanif 2 (Snake Robot), Hanif 3(Tracked Robot)Team Leader: Navid GhaffarzadeganTeam Members: Mohammad Hossein Taghavi, AmirSalman Avestimehr, Sadegh Dabiri, Arezu Moteval-izadeh, Hadi Emamifar, Ali Sharifi, Peiman ShariatPanahy, Sareh Ebrahimi, Mohammad Hossein Far-dad, Ali Azimi, Sina Rastad, Mohammad Na-jmzadeh

Because of the geographical placement ofIran, every year earthquakes are both com-mon and fatal. Thus, working on robots thatcould help rescuers in detecting victims is anecessary task. Our aim in this project is todesign, construct, and control an autonomousrobot that could move around in an unstruc-tured environment and detect victims in haz-ard areas. Our team is divided into three sub-teams: the mechanical engineering studentsgroup, the software engineering studentsgroup, and the hardware engineering studentsgroup. Current research includes dynamicalanalysis of mobile plates; path planning andvictim detection using image processing; posi-tioning of mobile robots; design, fabrication,and control of snake robots; design, fabrica-tion, and control of rovs; and mechanical useof smart materials—especially SMA.

National Botball ExhibitionNo, the graduate students haven’t gottenyounger! AAAI is pleased to host the NationalBotball Exhibition, featuring top robots builtby middle and high school students fromacross the country. Botball is a game in whichrobots attempt to achieve a specified goal, inan exciting head to head, double eliminationtournament. The goal of Botball is to get mid-dle and high school students involved in thecreative side of technology - to get our up-coming workforce excited about technology,robotics, and AI. Botball involves embodiedagent computer programming (in C), me-chanical design, science, math, and teamwork.

In this year’s tournament, teams either playthe black ball or white ball side. The challengeis to score points by moving your colored pingpong balls from inside a moveable goal intothe basket or into the end-zone. Robots are re-quired to start by themselves and shut downafter 90 seconds.

These robots were completely designed,built, and programmed by students from a kitof over 2000 parts. Students first compete inone of 12 regional tournaments and then ad-vance to the National Botball Tournament,held in Norman, Oklahoma in early July. Thebest of the best from that tournament will beshowcasing their robots at AAAI this year.For more information about the Botball pro-gram, please see www.botball.org.

Event ScheduleTuesday, July 3010:00 AM - 12:00 PM: Botball Practice Rounds1:00 PM - 4:00 PM: Botball Mini-Tournament 4:00 PM: Team Programming Challenge

Wednesday, July 3110:00 AM - 12:00 PM: Botball Practice Rounds

with new programming1:00 PM - 3:00 PM: Botball Seeding Rounds3:00 PM - 5:30 PM: Botball Mini-Tournament

(Botball Students & RBL Participants)

Botball

32

Regis

tratio

n RegistrationConference registration is located on the assembly level of the Shaw Conference Centre, begin-ning Sunday, July 28. Registration hours are:

Sunday, July 28 7:30 AM – 6:00 PM

Monday, July 29 7:30 AM – 6:00 PM

Tuesday, July 30 7:30 AM – 5:30 PM

Wednesday, July 31 8:00 AM – 5:30 PM

Thursday, August 1 8:00 AM – 2:00 PM

Only checks drawn on Canadian banks, VISA, MasterCard, American Express, governmentpurchase orders, traveler’s checks, and Canadian currency will be accepted. We cannot acceptforeign currency or checks drawn on foreign banks.

Registration FeesThe AAAI-02/IAAI-02 technical program registration fee includes admission to the technicalplenary and poster sessions, the Exhibition Program, the Tutorial Forum, the Workshop Pro-gram (by invitation only), the Opening Reception, the AI Festival, and the AAAI-02/IAAI-02Conference Proceedings. Tutorial Forum attendees may register for up to four consecutive tu-torials, and will receive the corresponding syllabi. Students must present proof of full-time stu-dent status to qualify for the student rate. Onsite technical program fees are in Canadian Dol-lars and are as follows:

Regular Member: $1,070 CDNRegular Nonmember: $1,225 CDNStudent Member: $345 CDNStudent Nonmember: $455 CDN

Workshop ProgramWorkshop registration is limited to those active participants determined by the organizer pri-or to the conference. All workshop participants must register for the AAAI-02 technical pro-gram. Registration onsite for a workshop is possible with the prior permission of the corre-sponding workshop organizer.

Robot Building LabThe robot building lab registration includes admission to the robot building lab and the exhi-bition program. Fees are $230.00 CDN for members or nonmembers, and $115.00 CDN forstudents. Attendance is limited and preregistration is required.

ExhibitionAdmission to the exhibition hall programs is included in all other types of registration. For in-dividuals interested in admittance to the exhibit hall only, an exhibits only registration is avail-able in onsite registration. Exhibit hall programs include vendor exhibits, the Intelligent Sys-tems Demonstrations, the Botball Exhibition and the Robot Competition and Exhibition.High-school students are welcome and will be admitted without fee upon presentation of avalid high-school student ID. Children under 12 will also be admitted without fee, but must beaccompanied by an adult conference registrant. Please note: The AI Festival, which will be heldin the exhibit hall, is included in the technical registration fee only. All other attendees must payan additional fee.All passes are good for Tuesday and Wednesday, July 30-31

Exhibits Only Adult: $15.00 CDNExhibits Only Adult w/Festival: $38.00 CDNExhibits Only Child w/Festival: $8.00 CDN

33

AdmissionEach conference attendee will receive a namebadge upon registration. This badge is re-quired for admittance to the technical, tutori-al, exhibit, IAAI and workshop programs.Workshop attendees will also be checked off amaster registration list at individual rooms.Tutorial attendees must present syllabi ticketsto receive syllabi. Smoking, drinking and eat-ing are not allowed in any of the technical, tu-torial, workshop or IAAI sessions.

Baggage HoldingThere is no baggage holding area at the ShawConference Centre. Please check your luggagewith the bellman at your hotel after you havechecked out. Neither the AAAI, the Shaw Con-ference Centre, the Weston Edmonton, theFairmont Hotel Macdonald, the Crowne PlazaChateau Lacombe, nor the University of Al-berta accept liability for the loss or theft of anysuitcase, briefcase, or other personal belong-ings brought to the site of AAAI-02/IAAI-02.

BankingThe closest banks and automated teller ma-chines (ATM) are below. The ATM networksavailable are Discover, MasterCard, Visa, Cir-rus, Honor and Plus. The banks can also ex-change all major foreign currencies.

Canadian Western Bank10303 Jasper Avenue423-8801Monday – Wednesday and Friday, 9:30 AM - 4:30PM, Thursday, 9:30 AM - 5:00 PM

Royal Bank10107 Jasper Avenue448-6611Monday – Friday, 9:30 AM - 4:00 PM,

TD Canada Trust148 Edmonton Centre448-8000Monday – Friday, 9:30 AM - 4:00 PM,

Business CentersThe following business centers are available:

Shaw Conference CentreBanquet Office located outside of Exhibit HallA. Services include faxing and photocopyingin small quantities.

Westin Edmonton Concourse Level:Monday – Friday, 8:00 AM - 5:30 PM

414-5700. Copy center and business services.

Staples Business DepotEdmonton Core, 10330 – 101 Street Monday – Friday, 8:00 AM - 9:00 PM; Saturday,9:00 AM - 6:00 PM; Sunday, 11:00 AM - 5:00 PM

424-4114. Copy center and business services.

Career InformationA bulletin board for job opportunities in theartificial intelligence industry will be madeavailable in the registration area, on the assem-bly level of the Shaw Conference Centre. At-tendees are welcome to post job descriptions ofopenings at their company or institution.

Child Care Services For information about childcare services, youmay contact Jan Pat Management Program inEdmonton at 448-1883. They have babysittingservices and KidScenes for families travelingto the city for conferences. (This informationis provided for your convenience and does notrepresent an endorsement of this agency byAAAI. Responsibility for all child care ar-rangements must be assumed by the parents.)

Coffee Breaks Coffee will be served in the foyer on the meet-ing level of the Shaw Conference Centre, Sun-day, July 28 and Monday, July 29, & 10:30 –11:00 AM and 3:30 – 4:00 PM. Coffee will alsobe served in the ballroom foyer of the WestinEdmonton during these times.

Coffee will be served in the foyer on the as-sembly level of the Shaw Conference CentreTuesday, July 30, 10:30 – 11:15 AM and 3:15 –3:50 PM, Wednesday, July 31, 10:40 – 11:10 AM,and Thursday, August 1, 10:30 – 11:05 AM and3:00 – 3:20 PM.

Copy ServicesCopy service is available in the banquet officelocated outside of Exhibit Hall A. Also seeBusiness Centers.

CurrencyCanadian dollars will be required for onsiteregistration, exhibits only passes, and tutorialsyllabi purchases. Approximate rates of ex-change at press time were:

CDN $1.00 = 0.65 US DollarsCDN $1.00 = 0.44 British PoundsCDN $1.00 = 82 Japanese YenCDN $1.00 = 0.69 Euros

General Information

34

DiningEdmonton dining information is available inthe Edmonton Where Magazine, which hasbeen included with your registration materi-als. A concession stand (coffee, pastry items,candy, etc.) will be open on the assembly levelduring normal business hours, Sunday, July 28– Thursday, August 1. The Shaw ConferenceCentre is connected via Pedway to CanadaPlace, which has a food court, open 7:30 AM –4:30 PM, 7 days a week.

Handicapped FacilitiesThe Shaw Conference Centre, The Westin Ed-monton, the Fairmont Hotel Macdonald, andthe Crowne Plaza Chateau Lacombe are allequipped with handicapped facilities.

HousingFor information regarding hotel reservations,please contact the hotels directly. For studenthousing, please contact University of Albertaat 492-4281 or 800-615-4807 (Canada only).

Information Desk An information desk/message desk will bestaffed during registration hours, Sundaythrough Thursday, July 28 – August 1. It is lo-cated near the registration area, on the assem-bly level of the Shaw Conference Centre. Mes-sages will be posted on the message boards ad-jacent to the desk. The telephone number forleaving messages only is (780) 917-7881. Pag-ing attendees is not possible.

Internet RoomInternet access is available in Salon 5 of theShaw Conference Centre. The internet roomwill be open Sunday, July 28 - Wednesday, July31, 8:00 AM – 5:30 PM; and Thursday 8:30 AM –3:30 PM. As a courtesy, please limit your accesstime to 10 minutes if others are waiting.

List of AttendeesA list of preregistered attendees of the confer-ence will be available for review at the AAAIDesk in the registration area on the assemblylevel of the Shaw Conference Centre. Attendeelists will not be distributed.

Message CenterSee Information Desk

ParkingThe Westin Edmonton charges $16.00 CDNper day for self-parking and $21.00 for valetparking per day. The Crowne Plaza ChateauLacombe charges $7.49 CDN per day for self-parking and $10.70 CDN for valet parking perday. The Fairmont Hotel Macdonald charges$8.00 CDN for self-parking or valet for up to 4hours and $19.95 CDN for more than 4 hours.

The City of Edmonton provides over 1,600convenient parking stalls within a 5-minutewalk from the Shaw Conference Centre. TheLibrary, Canada Place and City Hall Parkadesprovide pedway connections to the Centre.Parking is also available at the City Market andon-street meters in the vicinity.■ Canada Place Parkade: 9700 Jasper Avenue.

Entrance at 97 Street North of JasperAve.■ Library Parkade: 10165 – 100 Street. En-

trance on 99 Street (Citadel) & 100 Street■ City Hall Parkade: 1 Sir Winston Churchill

Square. Entrance on 99 Street■ City Market Surface Lot: 10165 – 97 Street.

Entrance on 101 and 102 AvenueRates: GST included, $1.25 CDN per half houror part thereof.

Monday-Friday6:00 AM-6:00 PM $10.00 maximum6:00 PM-1:00 AM $2.00 maximumDaily Maximum $15.00

Saturday and SundayMidnight to 6:00 PM

first 3 hours $1.00 maximumParking thereafter $1.25/half hour6:00 PM-1:00 AM $2.00 maximumDaily Maximum $10.00

PressAll members of the media are requested to reg-ister in the Press Room, on the meeting level ofthe Shaw Conference Centre in Salon 20. Pressbadges will only be issued to individuals withapproved credentials. The Press Room will beopen during the following hours.

Sunday, July 28 9:00 AM - 5:00 PM

Monday, July 29 8:00 AM - 5:00 PM

Tuesday, July 30 8:00 AM - 5:00 PM

Wednesday, July 31 8:00 AM - 5:00 PM

Thursday, August 1 8:00 AM - 12:00 PM

A volunteer will be on duty during these hoursto assist the members of the press and media.

Printed MaterialsDisplay tables for the distribution of promo-tional and informational materials of interest

Gene

ral In

forma

tion

35

to conference attendees will be located in theregistration area on the assembly level of theShaw Conference Centre.

ProceedingsEach registrant for the AAAI-02/IAAI-02technical program will receive a ticket with theregistration materials for one copy of the con-ference proceedings. During registrationhours on Sunday, July 28, Monday, July 29 anduntil 10:00 AM on Tuesday, July 30, Proceed-ings tickets can be redeemed at the AAAI PressProceedings desk, located in AAAI onsite reg-istration on the assembly level of the ShawConference Centre. After 10:00 AM on Tues-day, the Proceedings ticket may be redeemedat the MIT Press booth #104, located in Ex-hibit Hall AB of the Shaw Conference Centre,during exhibit hours. Limited Proceedingswill be available at the AAAI Press Proceedingsdesk again on Thursday, August 1 from 8:30 –1:00 PM. Extra Proceedings may be purchasedat the conference site at the above locations.Thursday, August 1, will be the last day to pur-chase extra copies of the Proceedings on site.

We strongly recommend that you redeemyour Proceedings ticket onsite. (See “Proceed-ings Shipping” if you do not want to carry yourProceedings with you.) Your ticket can be re-deemed by mailing the ticket with your name,complete shipping address and e-mail to:

AAAI, Attn: AAAI-02 Proceedings445 Burgess Drive, Suite 100Menlo Park, California 94025 USA

Postage must be prepaid with a check drawnon a US bank or MasterCard/Visa and expira-tion date. USA: $25.00; for orders outsideUSA: $55.00 for regular air mail, $105.00 forFedEx. Please allow eight to twelve weeks toreceive your proceedings.

Proceedings on CD-ROMEach registrant for the AAAI-02/IAAI-02technical program will receive a ticket with theregistration materials for one copy of the con-ference CD-ROM. During registration hourson Sunday and Monday, July 28-29 and until10:00 AM on Tuesday, July 30, CD-ROM ticketscan be redeemed at the AAAI Press Proceed-ings desk, located in AAAI onsite registrationon the assembly level of the Shaw ConferenceCentre. After 10:00 AM on Tuesday, the CD-ROM ticket may be redeemed at the Registra-tion Desk. Extra CD-ROM tickets may be pur-chased in onsite registration. Thursday, Au-gust 1 will be the last day to purchase extra

copies of the conference CD-ROM onsite.Registrants must pick up their conferenceCD-ROM’s onsite. (AAAI cannot redeem tick-ets after the conference.)

Proceedings ShippingGoodkey Show Services at the Shaw Confer-ence Centre can arrange for shipping services.Contact: Calvin Goodkey at 426-2211.

Speaker Ready RoomThe Speaker Ready Room will be located in Sa-lon 19 on the assembly level of the Shaw Con-ference Centre. This room has audio-visualequipment to assist speakers with their prepa-rations. It is important that speakers visit thisroom to organize their materials. The room willbe open from 8:00 AM - 5:00 PM, Sunday, July 28– Tuesday, July 30, and Thursday, August 1; 8:00AM – 3:00 PM on Wednesday, July 31.

Invited speakers are asked to come to Salon19 one-day prior to their speech. Representa-tives from AV Headquarters will be availableduring the times listed above to confirm youraudiovisual needs, and assist with the prepara-tion of your materials, if necessary.

Taxes/Rebate of TaxesA federal tax (GST) of 7% is now applied tomost goods and services in Canada. A harmo-nized sales tax (HST) is added to all goods andservices. Nonresident visitors can apply for aGST/HST rebate on most goods purchased foruse outside Canada as well as on short-term ac-commodations. Rebates can be claimed in per-son at participating Canadian Duty Free Shopsat border crossings or airports, or by complet-ing and mailing the rebate form. A nonresidentvisitor qualifies for a tax refund if the followingconditions are met: you are not a resident ofCanada; you purchased eligible goods, shortterm accommodation or both; you paidGST/HST on these purchases; you have originalreceipts; the total of your purchase amounts(before taxes) for eligible goods and accommo-dations must be at least $200 CDN; each indi-vidual receipt for eligible goods shows a mini-mum total purchase (before taxes) of $50 CDN;the goods are removed from Canada within 60days of the date they were purchased. Retailershave forms available.

TelephonesPublic telephones for domestic and interna-

General Information

36

tional calls are located throughout the ShawConference Centre on all levels.

TippingIn general, a tip of 10-20 % is given to waiters,waitresses, taxi drivers, etc. Bellhops, doormen,porters, etc. at hotels, airports and railway sta-tions are generally paid $1.00 per item.

TransportationThe following information provided is thebest available at press time. Please confirmfares when making reservations.

Airlines and Rental CarsThe American Association for Artificial Intelli-gence has selected Air Canada as the officialcarrier and Avis Rent A Car as the official carrental agency for AAAI-02/IAAI-02. If you needto change your airline or car rental reservations,please call Stellar Access, our official travelagency at 858-805-6109 and reference AAAIEvent #1889. Web: www.stellaraccess.com

Airport ShuttleSky Shuttle is the official carrier to and fromEdmonton International Airport. They offerfrequent service to and from the conferencehotels. The one-way fare is $13.00 CDN;round trip is $20.00 CDN. For more informa-tion, consult www.edmontonairports.com/gtr/skyshutt.htm, or call 888-438-2342.

TaxiTaxis are available at Edmonton InternationalAirport. The approximate fare from the air-port to downtown Edmonton is $41.00 CDN.

BusGreyhound Bus: For information on fares andscheduling, call 1-800-661-8747. The down-town Edmonton Greyhound terminal is locat-ed at 10324 - 103 Street.

City Transit SystemThe Edmonton Transit System includes Buses,LRT (Light Rail Transit) and DATS (DisabledAdult Transit Service). They can be contacted at496-1611 or www.gov.edmonton.ab.ca/transit.The fares are $2.00 CDN. The station closest tothe Shaw Conference Center is Central and islocated at 100th Street and Jasper Avenue.

TrainVia Rail – The Via Rail station is located with-in a 15-minute drive of downtown Edmonton.The toll free number in Canada is, 1-800-561-8630. www.viarail.ca.

Tutorial SyllabiExtra copies of AAAI-02 tutorial syllabi will beavailable for purchase in AAAI onsite registra-tion in the Shaw Conference Centre, begin-ning Tuesday, July 30. Supplies are limited.The cost is $25.00 CDN per syllabus. Preregis-tration tutorial syllabi tickets may be re-deemed in the tutorial rooms.

Visitor InformationEdmonton Tourism welcomes you to Edmon-ton! They are located in the Shaw ConferenceCentre and are open Monday through Friday,8:00 AM – 4:30 PM.

Edmonton Tourism9797 Jasper AvenueEdmonton, Alberta, Canada T5J 1N9Telephone: 426-4715

Volunteer StationThe volunteer station will be located in the reg-istration area on the assembly level of the ShawConference Centre. All volunteers shouldcheck in at the beginning of their shifts withthe volunteer at the booth prior to their shifts.The volunteer meeting will be held Saturday,July 27 at 4:00 PM in the registration area.

DisclaimerIn offering Avis Rent A Car, Canada Air,Crowne Plaza Chateau Lacombe, The FairmontHotel Macdonald, Shaw Conference Centre,University of Alberta, The Westin Edmonton,and all other service providers (hereinafter re-ferred to as “Supplier(s)” for the National Con-ference on Artificial Intelligence and the Inno-vative Applications Conference), AAAI acts on-ly in the capacity of agent for the Supplierswhich are the providers of the service. BecauseAAAI has no control over the personnel, equip-ment or operations or providers of accommo-dations or other services included as part of theAAAI-02/IAAI-02 program, AAAI assumes noresponsibility for and will not be liable for anypersonal delay, inconveniences or other damagesuffered by conference participants which mayarise by reason of (1) any wrongful or negligentacts or omissions on the part of any Supplier orits employees, (2) any defect in or failure of anyvehicle, equipment or instrumentality owned,operated or otherwise used by any Supplier, or(3) any wrongful or negligent acts or omissionson the part of any other party not under thecontrol, direct or otherwise, of AAAI.

Gene

ral In

forma

tion

37

38

Assem

bly Le

vel M

ap (E

xhibi

t Area

)

D312

D314

D318

D316

LO

UN

GE

D320

D417

D415

D413

202

200

204

206

303

301

116

217

112

114

213

215E

XH

IBIT

ION

RO

BO

T

ROBOT WORK AREAC

OM

PE

TIT

ION

RO

BO

T

65'

110'

40'

50' 3

0'

20'

HA

LL

C

LO

BB

Y

ES

CA

LA

TO

RS

BU

SIN

ES

S

CE

NT

RE

ST

OR

E

TR

UC

K B

AY

S

LO

AD

ING

DO

CK

S

SE

RV

ICE

CO

RR

IDO

R

307

118

305

219

50'

Poste

r B

oard

s

Poste

r B

oard

s

208

309

120

221

203

102

100

201

207

106

104

205

108

209

D419

D421

65'

AR

EA

TO

UR

NA

ME

NT

BO

TB

ALL

40'

30'

BLE

AC

HE

RS

39

Meeting Level Map