Build Intelligence from the bottom up!

Slides:



Advertisements
Similar presentations
Elephants Don’t Play Chess
Advertisements

Lecture 8: Three-Level Architectures CS 344R: Robotics Benjamin Kuipers.
Artificial Intelligence Lecture 11. Computer Science Robotics & AI.
5-1 Chapter 5: REACTIVE AND HYBRID ARCHITECTURES.
1 Robotic Summer School 2009 Subsumption architecture Andrej Lúčny Department of Applied Informatics, FMFI, Comenius University, Bratislava
Embedded System Lab Kim Jong Hwi Chonbuk National University Introduction to Intelligent Robots.
A Summary of the Article “Intelligence Without Representation” by Rodney A. Brooks (1987) Presented by Dain Finn.
Robotics CSPP Artificial Intelligence March 10, 2004.
ECE 4340/7340 Exam #2 Review Winter Sensing and Perception CMUcam and image representation (RGB, YUV) Percept; logical sensors Logical redundancy.
Brent Dingle Marco A. Morales Texas A&M University, Spring 2002
Experiences with an Architecture for Intelligent Reactive Agents By R. Peter Bonasso, R. James Firby, Erann Gat, David Kortenkamp, David P Miller, Marc.
Intelligence without Reason
IofT 1910 W Fall 2006 Week 5 Plan for today:  discuss questions asked in writeup  talk about approaches to building intelligence  talk about the lab.
Jochen Triesch, UC San Diego, 1 Real Artificial Life: Robots.
IofT 1910 W Fall 2006 Week 3 Plan for today:  discuss questions asked for the writeup  talk about Brooks’ approach and compare it with other approaches.
Autonomous Mobile Robots CPE 470/670 Lecture 8 Instructor: Monica Nicolescu.
Integrating POMDP and RL for a Two Layer Simulated Robot Architecture Presented by Alp Sardağ.
Topics: Introduction to Robotics CS 491/691(X) Lecture 8 Instructor: Monica Nicolescu.
Autonomous Mobile Robots CPE 470/670 Lecture 9 Instructor: Monica Nicolescu.
Integration of Representation Into Goal- Driven Behavior-Based Robots By Dr. Maja J. Mataric` Presented by Andy Klempau.
Autonomous Mobile Robots CPE 470/670 Lecture 8 Instructor: Monica Nicolescu.
Topics: Introduction to Robotics CS 491/691(X)
Behavior- Based Approaches Behavior- Based Approaches.
Distributed Robot Agent Brent Dingle Marco A. Morales.
A Robust Layered Control System for a Mobile Robot Rodney A. Brooks Presenter: Michael Vidal.
Mobile Robot Control Architectures “A Robust Layered Control System for a Mobile Robot” -- Brooks 1986 “On Three-Layer Architectures” -- Gat 1998? Presented.
Introduction to Behavior- Based Robotics Based on the book Behavior- Based Robotics by Ronald C. Arkin.
Chapter 14: Artificial Intelligence Invitation to Computer Science, C++ Version, Third Edition.
Artificial Intelligence Chapter 2 Stimulus-Response Agents
Behavior Based Robotics: A Wall Following Behavior Arun Mahendra - Dept. of Math, Physics & Engineering, Tarleton State University Mentor: Dr. Mircea Agapie.
Nuttapon Boonpinon Advisor Dr. Attawith Sudsang Department of Computer Engineering,Chulalongkorn University Pattern Formation for Heterogeneous.
Reactive Paradigm – Overview Subsumption Architecture By Ian Jonkers Studies in Machine Learning: Intelligent Robotics.
Robotica Lecture 3. 2 Robot Control Robot control is the mean by which the sensing and action of a robot are coordinated The infinitely many possible.
4 Introduction to AI Robotics (MIT Press)Chapter 4: The Reactive Paradigm1 The Reactive Paradigm Describe the Reactive Paradigm in terms of the 3 robot.
Autonomous Mobile Robots CPE 470/670 Lecture 8 Instructor: Monica Nicolescu.
Intelligent Systems Lecture 13 Intelligent robots.
Artificial intelligence and robots Jacek Malec Department of Computer Science Lund University
Robotica Lecture 3. 2 Robot Control Robot control is the mean by which the sensing and action of a robot are coordinated The infinitely many possible.
Electrical Engineering Design Project - Fall 2002 Electrical/Computer Engineering Design Project Fall 2002 Lecture 4 – Robotics.
University of Windsor School of Computer Science Topics in Artificial Intelligence Fall 2008 Sept 11, 2008.
Behavior Control for Robotic Exploration of Planetary Surfaces Written by Erann Gat, Rajiv Desai, Robert Ivlev, John Loch and David P Miller Presented.
Behaviour-Based Control in Mobile Robotics
Robotica Lecture Review Reactive control Complete control space Action selection The subsumption architecture –Vertical vs. horizontal decomposition.
Brooks’ Subsumption Architecture EEL 6838 T. Ryan Fitz-Gibbon 1/24/2004.
Introduction to Artificial Intelligence CS 438 Spring 2008 Today –AIMA, Ch. 25 –Robotics Thursday –Robotics continued Home Work due next Tuesday –Ch. 13:
Subsumption Architecture and Nouvelle AI Arpit Maheshwari Nihit Gupta Saransh Gupta Swapnil Srivastava.
AI & Machine Learning Libraries By Logan Kearsley.
4 Introduction to AI Robotics (MIT Press)Chapter 4: The Reactive Paradigm1 The Reactive Paradigm Describe the Reactive Paradigm in terms of the 3 robot.
Rational Agency CSMC Introduction to Artificial Intelligence January 8, 2004.
ROBOTICS COE 584 Robotic Control Architecture.
The Language of Thought : Part II Joe Lau Philosophy HKU.
INTELLIGENCE WITHOUT REPRESENTAION 인지과학 협동과정 이광주.
Authors: Brooks, R.A. ; Massachusetts Institute of Technology, Cambridge, MA, USA Form : IEEE JOURNAL OF ROBOTICS AND AUTOMATION, VOL. RA-2, NO. I, MARCH.
Matt Loper / Brown University Presented for CS296-3 February 14th, 2007 On Three Layer Architectures (Erann Gat) On Three Layer Architectures (Erann Gat)
Middle East Technical University
CSPP Artificial Intelligence March 10, 2004
CMSC Artificial Intelligence March 11, 2008
Do software agents know what they talk about?
Build Intelligence from the bottom up!
Today: Classic & AI Control Wednesday: Image Processing/Vision
Introduction To Intelligent Control
CO Games Development 2 Week 19 Extensions to Finite State Machines
Non-Symbolic AI lecture 4
Artificial Intelligence Chapter 2 Stimulus-Response Agents
Build Intelligence from the bottom up!
Subsuption Architecture
Robot Intelligence Kevin Warwick.
CHAPTER I. of EVOLUTIONARY ROBOTICS Stefano Nolfi and Dario Floreano
Non-Symbolic AI lecture 4
Behavior Based Systems
Presentation transcript:

Build Intelligence from the bottom up! Behavioral Robotics Build Intelligence from the bottom up!

AI Debate #3 Is representation needed? “When we examine very simple intelligence we find that explicit representations and models of the world get in the way.” “Representation is the wrong unit of abstraction in building the bulkiest parts of intelligent systems.” Rodney A. Brooks in Intelligence without representation

Evolutionary Timeline Single cell entities arose in the soup 3.5 billion years ago Fish and vertebrates arose 550 million years ago Insects – 350 million years ago Reptiles – 370 million Primates – 120 million Man 2.5 million Agriculture 10,000 years Writing – 5000 years Expert knowledge in the last few hundred years

Traditional Robotic Architecture Perception Modeling Planning Task Execution Motor Control Actuators Sensors

Effects of Traditional Architecture Single point of failure Spends a lot of time planning! Sensor Fusion Planning is a way of avoiding deciding what to do next!

Subsumption Architecture Reason Plan Changes Monitor Changes Actuators Sensors Build Maps Explore Wander Avoid Objects

Assumptions Complex behaviors need not come from a complex control scheme. Things should be simple! Build cheap robots in natural spaces. The world is 3D! Absolute coordinates don’t work! Useful worlds are not polyhedra. Sensors and actuators fail. Build artificial beings!

Subsumption Module inhibition FSM Outputs Inputs Reset Suppression

Simple Level 0 Feel Force runaway Turn Sonar Forward Collide force heading heading Sonar Forward Collide encoders

Add level 1 Wander Avoid Sonar Feel Force runaway Turn Collide Forward heading encoders

Mapping / Navigation Since we can’t use: How can we map the world? absolute coordinates Sensor fusion How can we map the world? Distributed Maps Maja J. Mataric and R. Brooks

Landmark detector Sonar compass

Kismet

Ants

Genghis

Creature requirements Cope in a timely fashion in a dynamic environment Robust w.r.t it’s environment Maintain multiple goals Do something!

What this approach isn’t! Not neural nets / connectionism Use FSM and low connectivity Not production rules Not a blackboard Not German Philosophy Heidegger has similar ideas, but this is engineering!

Limits to Growth How many layers can be stacked up? How much complexity can happen without central representation Can learning occur in fixed networks?