Article View: comp.ai
Article #9054UNSW AI SEMINAR 18/9: Zrimec on robotics & causal qualitative modelling
From: Tim Menzies
Date: Wed, 16 Sep 1992 02:08
Date: Wed, 16 Sep 1992 02:08
64 lines
3843 bytes
3843 bytes
University of New South Wales
School of Computer Science and Engineering
Artificial Intelligence Laboratory
Seminar Series
TITLE: Learning in Robotics By Experimentation
WHO: Tatjana Zrimec
University of Ljubljana, Faculty of Electri-
cal Engineering and Computer Science,
Ljubljana, Slovenia.
WHEN: 1-2pm, Friday, 18 Sept, 1992.
ABSTRACT: Present robotic systems are still difficult
to work with and are successfully employed
only in very carefully engineered environ-
ments. One way of making robot systems more
universal is to enable robots to acquire
knowledge about their environment and to use
this knowledge.
This work investigates the use of learning
through experimentation by an autonomous
agent. A real robot is embedded in an
environment that can be sensed and acted
upon. Through learning, the robot is able to
induce laws about its environment and to
discover some basic relationships between
action and perception. The result of the
learning is a qualitative model. The result-
ing knowledge is common sense knowledge
represented through different levels of
abstraction.
With this work we show that is possible to
develop a theory of a dynamic process by
experimentation and learning. The developed
qualitative model for object pushing is com-
pletely supported by the theory of pushing
developed using classical tools.
WHERE: Rm 212, Samuel Building, University of New
South Wales. Enter off Botany Street, stop at
gate and ask grey men for day pass and
instructions. Park in the main car park on
your left. The Samuels bulding is the gray
building directly behind the gates with
strange shield-like objects over the windows.
INQUIRIES: timm@spectrum.cs.unsw.oz.au (Tim Menzies)
--------------------------------------------------------------
PLEASE NOTE:
If you have a seminar that you would like to present at this
forum, contact Tim.
Technical Details
Message-ID:
<1992Sep16.020839.27508@usage.csd.unsw.OZ.AU>Path:
arc1-ca.pugleaf.net!sparky!uunet!europa.asd.contel.com!darwin.sura.net!jvnc.net!rutgers!sun-barr!west.West.Sun.COM!sunaus!swift!pta!yarra!mel.dit.csiro.au!munnari.oz.au!manuel!sserve!usage!spectrum!timmShow Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 164a65,6d558724050f07e,start X-Google-Attributes: gid164a65,public X-Google-Thread: 108449,6d558724050f07e,start X-Google-Attributes: gid108449,public X-Google-Thread: 100110,6d558724050f07e,start X-Google-Attributes: gid100110,public X-Google-Thread: 10461e,6d558724050f07e,start X-Google-Attributes: gid10461e,public X-Google-Thread: 102449,6d558724050f07e,start X-Google-Attributes: gid102449,public X-Google-ArrivalTime: 1992-09-15 20:41:18 PST Path: arc1-ca.pugleaf.net!sparky!uunet!europa.asd.contel.com!darwin.sura.net!jvnc.net!rutgers!sun-barr!west.West.Sun.COM!sunaus!swift!pta!yarra!mel.dit.csiro.au!munnari.oz.au!manuel!sserve!usage!spectrum!timm From: timm@spectrum.cs.unsw.oz.au (Tim Menzies) Newsgroups: comp.ai,syd.seminar,aus.ai,unsw.general,comp.robotics Subject: UNSW AI SEMINAR 18/9: Zrimec on robotics & causal qualitative modelling Message-ID: <1992Sep16.020839.27508@usage.csd.unsw.OZ.AU> Date: 16 Sep 92 02:08:39 GMT Sender: news@usage.csd.unsw.OZ.AU Reply-To: timm@spectrum.cs.unsw.oz.au (Tim Menzies) Distribution: aus Organization: none Lines: 64 Nntp-Posting-Host: prussian.spectrum.cs.unsw.oz.au