The Dynamics of Intelligence:
A Constraint-Based Architecture for Situated Visual Agents

        Alan Mackworth
   University of British Columbia

   11am, SF1105, February 23, 1999

Abstract

We need methods for designing and building perceptual agents that are clean,
powerful and practical.  But no methodology satisfies all three criteria,
yet.  A robot is a hybrid intelligent dynamical system, consisting of a
controller coupled to its body.  The Constraint Net (CN) model is a unitary
framework for building hybrid intelligent systems as situated agents.  In
CN, the designer can specify the robot's vision, control and motor systems
uniformly as on-line systems.  We advocate an architecture for agent
perceiver/controllers consisting of multi-level constraint-satisfying
modules.  If the perceptual and control systems are designed as
constraint-satisfying devices then the total robotic system, consisting of
the robot symmetrically coupled to the environment, may be proven correct.
In some cases, a controller may be synthesized from a constraint-based
specification.  The ideas developed in this talk are illustrated by
application to the challenge of building real visually-controlled robot
soccer players.  The controllers for our softbot soccer team, UBC Dynamo98,
are modeled in CN and implemented with Java Beans technology.  This work is
joint with Ying Zhang, Yu Zhang and others in our laboratory.

Biography

Alan Mackworth is a Professor of Computer Science and the Director of the
Laboratory for Computational Intelligence at the University of British
Columbia.  He received a B.A.Sc.(Toronto) in 1966, A.M.(Harvard) in 1967 and
D.Phil.(Sussex) in 1974.  Since 1974, he has been at UBC.  He has worked
primarily on constraint-based computational intelligence with applications
in vision, robotics and situated agents.  He has published over 70
scientific papers, and co-authored the book, {\em Computational
Intelligence: A Logical Approach} by Poole, Mackworth \& Goebel (1998).  He
has received the ITAC/NSERC Award for Academic Excellence and the CSCSI
Award for Distinguished Service.  He is a Principal Investigator in the
Institute for Robotics and Intelligent Systems and a Fellow of AAAI and the
Canadian Institute for Advanced Research.