[ACT-R-users] ACT-R robot (UNCLASSIFIED)

[Kelley, Troy (Civ,ARL/HRED)]

Classification:  UNCLASSIFIED 
Caveats: NONE

Bruce,

There are several people working with cognitive architectures (ACT-R and
Soar) on robotics platforms.  Lyle Long at Penn State, Paul Benjamin at
Pace University, John Laird at Michigan University, Greg Trafton and
Alan Shultz at NRL.  

At the Army we have developed a program that was inspired by ACT-R
called SS-RICS that interfaces with the ARIA object class for ActivMedia
robots. 
(www.mobilerobots.com)

Here is a link to some SS-RICS publications.

http://www.ss-rics.org/Publications/Publications.aspx 

We have found that there is still a huge "perception problem" in
robotics and using a cognitive architecture will not immediately solve
that problem, however it will help.  We have been looking at using a
cognitive architecture in conjunction with distributed connectionist
systems which can do some limited types of perception.  But perception
is a two way street and determining what kind of bottom up information
to feed to a symbolic system and what types of top down information to
feed to a connectionist system still remains a problem.

Good luck!
Troy D. Kelley
AMSRD-HR-SE
Army Research Laboratory
Human Research and Engineering Directorate (HRED)
Aberdeen Proving Ground, Aberdeen MD 21005-5425
voice: 410-278-5869
fax: 410-278-9523

-----Original Message-----
From: act-r-users-bounces at act-r.psy.cmu.edu
[mailto:act-r-users-bounces at act-r.psy.cmu.edu] On Behalf Of Bruce J
Weimer MD
Sent: Tuesday, June 10, 2008 11:37 AM
To: act-r-users at act-r.psy.cmu.edu
Subject: [ACT-R-users] ACT-R robot

I'm using LispWorks for Windows to program an open-source
laptop-on-wheels type hobby robot.  I'm looking into ACT-R as a possible
"cognitive engine".  My robot, Leaf, has emotion driven behaviors,
OpenCV face detection/recognition, SAPI5 speech recognition/tts, CU
Animate face animation, WiFi robot to robot "telepathy", internet
telepresence, X10 home automation and more.  Anyway, I'm intrigued by
ACT-R's "conflict resolution" scheme.  And I'm also interested in using
WN Lexical to interface WordNet for natural language processing.  I'm
looking then for some help interfacing ACT-R to the robot - sending data
from the robot to ACT-R regarding people the robot's seen, phrases the
robot's heard, etc and then sending data from ACT-R to the robot to give
verbal responses, actuate motor sequences, etc.
 
So, if anyone's interested in helping or can provide links to
references/papers, please contact me at:
 
bjweimer at charter.net
 
Thanks in advance!
 
Bruce.
 
Classification:  UNCLASSIFIED 
Caveats: NONE