<- main page  <- Lecture 1     Lecture 3->

 

INTRODUCTION TO A.I.

 

Agents

& Perception-Action Architectures

Lecture 2

January 7, 2008

 

 

 

2.1 Outline of Topics
  Braitenberg Vehicles  
  I  
  II  
  III  
  Resources  
  Missing in Action  
  Example Agents  
     

 

 

 

 

 

 

 

 

 

 

 

 

2.2 Braitenberg Vehicles I
 
 
Braitenberg Vehicle
 
Ingredients
  
 
2 sensors
  
 
2 motors
  
 
2 Wires
  
     
 
 
"Love"
"Hate"
     

 

 

 

 

 

 

 

 

 

 

 

 

 

2.3 Braintenberg Vehicles II
 
 
"Inverse love"
"Inverse hate"
     

 

 

 

 

 

 

 

 

 

 

 

 

 

2.4 Braintenberg Vehicles III
 
 
"Curious"
  Outer light sensors

Strong positive cross-connections make vehicle steer towards light
  Inner light sensors Weak positive same-side connections make vehicle steer away from light
  Resulting behavior Vehicle will move towards light that is far away but when it comes closer it will not crash into it
  More formally W(green) < W(pink); W(red) < W(orange)
W(green) = W(red); W(pink) = W(orange)
where W is weight or transform of light energy into motor speed
     
     

 

 

 

 

 

 

 

 

 

 

 

 

 

2.5 Valentino Braitenberg Vehicles Resources
 

Valentino Braitenberg:

Vehicles - Experiments in Synthetic Psychology

MIT Press
 
Braitenberg Vehicles simulator - Downloadable Java code
 http://www.mindspring.com/~gerken/vehicles/download/index.html
 
Braitenberg Vehicles simulator - applet
 http://privatewww.essex.ac.uk/~ranewc/research/brait/javaBraitenberg.html
 
MAVRICK -
a robot architecture inspired by Braitenberg
 http://faculty.washington.edu/gmobus/AdaptiveAgents/MAVRIC-EBA.html
 

Java simulator
(Braitenberg, Subsumption, more)

 http://www.ifi.unizh.ch/~pfeifer/mitbook/braitenberg/braitenberg.html
     

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2.6 Schools of thought in AI
  Cybernetics Starting in the 1940s. Example works: Norbert Wiener's 1948 "Cybernetics, or Control and Communication in the Animal and Machine" and continued work such as WHAT THE FROG'S EYE TELLS THE FROG'S BRAIN by Lettvin, Maturana, McCullochs & Pitts, 1968.
  Emphasis
  • Effort to merge engineering/computer science and biology into a new field
  Classical AI / "good old-fashioned AI" (GOFAI) Starting 1956: The Dartmouth Conference
  Emphasis
  • Logic
  • Games
  • Isolated models
  • advanced competences
  • static structures/knowledge
  Behavior-based robotics Starting around 1985 with R. Brooks' work on subsumption architecture
  Emphasis
  • reactivity, speed
  • architectural simplicity
  • Holistic models of perception and action
  • Robotics
  • Simplicity
  Cognitive Science An effort to integrate knowledge from psychology, neurophysiology, engineering, computer science, philosophy, lingustics in the study of mind. Initially pushed forward by Carnegie-Mellon. Possibly what the Cybernetics movement intended to become.
     

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2.7
 Example A.I. Agents
  Cog
  • MIT
  • Main methodology: Subsumption architecture
  Attila / Genghis
  • MIT
  • Main methodology: Competence network; subsumption architecture
  Gandalf
  • MIT
  • Main methodology: Constructionist AI / CDM
  Others
  • AMBLER, CMU
  • Shakey, Stanford
  • Pengi/Sonja, Chapman & Agre
  • ASIMO, HONDA
  • QRIO, SONY
     

 

 

 

RODNEY BROOKS WITH COG

 

 

 

ATTILA - SIX-LEGGED ROBOT AT MIT

 

 

 

GANDALF: COMMUNICATIVE VIRTUAL HUMANOID AT MIT