introduction to special topics intelligent robotics cis480 january 16, 2007
TRANSCRIPT
Introduction to Special TopicsIntroduction to Special TopicsIntelligent RoboticsIntelligent Robotics
CIS480CIS480January 16, 2007January 16, 2007
2007 Kutztown University 2
Basic DefinitionsBasic Definitions
Artificial IntelligenceArtificial Intelligence Study intelligenceStudy intelligence Apply – design intelligent systemsApply – design intelligent systems to better serve mankindto better serve mankind
2007 Kutztown University 3
Basic DefinitionsBasic Definitions
RobotRobot EmbodiedEmbodied AutonomousAutonomous AgentAgent
Intelligent robotIntelligent robot ++ Intelligent Intelligent
2007 Kutztown University 4
EmbodiedEmbodied SituatedSituated
Located in the worldLocated in the world Sensor (receptor)Sensor (receptor)
Receive “raw” information about worldReceive “raw” information about world Effector (actuator)Effector (actuator)
Accomplish an actionAccomplish an action» with respect to selfwith respect to self
» with respect to worldwith respect to world
“ “one who brings about a result or event; one one who brings about a result or event; one who accomplishes a purpose”who accomplishes a purpose”
2007 Kutztown University 5
AutonomousAutonomous Autos :: selfAutos :: self Nomos :: lawNomos :: law DefinitionDefinition
Self-governingSelf-governing» Note: The term cybernetics stems from the Greek Note: The term cybernetics stems from the Greek
Κυβερνήτης (kybernetes, steersman, governor, pilot, Κυβερνήτης (kybernetes, steersman, governor, pilot, or rudder — the same root as government).or rudder — the same root as government).
Independent in mind or judgmentIndependent in mind or judgment Self-directingSelf-directing Not controlled by others or outside forcesNot controlled by others or outside forces
2007 Kutztown University 6
AgentAgent
DefinitionDefinition One that acts or has power/authority to actOne that acts or has power/authority to act One that represents anotherOne that represents another Root meaning – one that acts or exerts powerRoot meaning – one that acts or exerts power Means of effecting a resultMeans of effecting a result
Software agent (David Croft)Software agent (David Croft) Delegacy – discretionary authorityDelegacy – discretionary authority CompetenceCompetence Amenability – ability to adaptAmenability – ability to adapt Software residentSoftware resident
2007 Kutztown University 7
IntelligenceIntelligence Many definitionsMany definitions MSN Encarta: “general mental capability to MSN Encarta: “general mental capability to
reason, solve problems, think abstractly, learn and reason, solve problems, think abstractly, learn and understand new material, and profit from past understand new material, and profit from past experience. Intelligence can be measured by many experience. Intelligence can be measured by many different kinds of tasks . . . Intelligence draws on a different kinds of tasks . . . Intelligence draws on a variety of mental processes, including memory, variety of mental processes, including memory, learning, perception, decision-making, thinking, learning, perception, decision-making, thinking, and reasoning.”and reasoning.”
Einstein: “Imagination is more important than Einstein: “Imagination is more important than knowledge.”knowledge.”
Henk Tuten: “complex use of creativity”Henk Tuten: “complex use of creativity”
2007 Kutztown University 8
IntelligenceIntelligence Another viewAnother view Creative simplicityCreative simplicity ExamplesExamples
Every mileage divisible by 3000Every mileage divisible by 3000Cutting the Gordian knotCutting the Gordian knotEfficient use of carsEfficient use of cars
2007 Kutztown University 9
Robot “Primitives” – MurphyRobot “Primitives” – Murphy
SenseSense PlanPlan ActAct Perhaps a 4Perhaps a 4thth – Learn – Learn
2007 Kutztown University 10
Origin of Robot BasicsOrigin of Robot Basics
Serve Serve Act Act Autonomous Autonomous Sense Sense
2007 Kutztown University 11
Origin of Robot BasicsOrigin of Robot Basics
The most basic ::The most basic :: ActAct SenseSense
2007 Kutztown University 12
Origin of Robot BasicsOrigin of Robot Basics
Serve Serve Act Act Autonomous Autonomous Sense Sense Changing environment Changing environment Plan Plan Plan Plan Reason Reason Plan Plan Represent knowledge Represent knowledge Knowledge Knowledge Learning Learning
2007 Kutztown University 13
Robot BasicsRobot Basics
SenseSense ActAct Represent knowledgeRepresent knowledge ReasonReason LearnLearn
2007 Kutztown University 14
Central Issues of A.I.Central Issues of A.I.
Knowledge representation & Knowledge representation & reasoningreasoning
2007 Kutztown University 15
Chapter OneChapter OneI. Robotic ParadigmsI. Robotic Paradigms1. From Teleoperation to Autonomy1. From Teleoperation to Autonomy1.1 Overview1.1 Overview1.2 How Can a Machine Be Intelligent?1.2 How Can a Machine Be Intelligent?1.3 What Can Robots Be Used For?1.3 What Can Robots Be Used For?1.3.1 Social implications of robotics1.3.1 Social implications of robotics1.4 A Brief History of Robotics1.4 A Brief History of Robotics1.4.1 Industrial manipulators1.4.1 Industrial manipulators1.4.2 Space robotics and the AI approach1.4.2 Space robotics and the AI approach1.5 Teleoperation1.5 Teleoperation1.5.1 Telepresence1.5.1 Telepresence1.5.2 Semi-autonomous control1.5.2 Semi-autonomous control1.6 The Seven Areas of AI1.6 The Seven Areas of AI1.7 Summary1.7 Summary
2007 Kutztown University 16
ParadigmParadigm LinguisticsLinguistics
A pattern of conjugation or declension to memorize A pattern of conjugation or declension to memorize which serves as a template for a class of words. which serves as a template for a class of words.
Thomas KuhnThomas Kuhn Shared scientific theoriesShared scientific theories Common methods of solving problemsCommon methods of solving problems Common norms for scientific activityCommon norms for scientific activity Shared metaphysicsShared metaphysics
Current usageCurrent usage Example, patternExample, pattern Conjugation, declensionConjugation, declension Theoretical frameworkTheoretical framework
2007 Kutztown University 17
Paradigm – current usageParadigm – current usage Merriam-Webster:Merriam-Webster:Main Entry: Main Entry: par·a·digmpar·a·digm
Pronunciation: 'per-&-"dIm, 'pa-r&- Pronunciation: 'per-&-"dIm, 'pa-r&- alsoalso -"dim -"dimFunction: Function: nounnounEtymology: Late Latin Etymology: Late Latin paradigma,paradigma, from Greek from Greek paradeigma,paradeigma, from from paradeiknynaiparadeiknynai to show side by side, to show side by side, from from para-para- + + deiknynaideiknynai to show -- more at to show -- more at DICTIONDICTION11 : : EXAMPLEEXAMPLE, , PATTERNPATTERN; ; especiallyespecially :: an outstandingly an outstandingly clear or typical example or archetypeclear or typical example or archetype22 :: an example of a conjugation or declension showing a an example of a conjugation or declension showing a word in all its inflectional formsword in all its inflectional forms33 :: a philosophical and theoretical framework of a a philosophical and theoretical framework of a scientific school or discipline within which theories, laws, scientific school or discipline within which theories, laws, and generalizations and the experiments performed in and generalizations and the experiments performed in support of them are formulated; support of them are formulated; broadlybroadly :: a philosophical a philosophical or theoretical framework of any kind or theoretical framework of any kind
2007 Kutztown University 18
The Structure of Scientific The Structure of Scientific RevolutionsRevolutions
Stanford Encyclopedia of PhilosophyStanford Encyclopedia of Philosophy :: ::
““one of the most cited academic books of all one of the most cited academic books of all time” time”
Development of “science”Development of “science” NotNot
» Steady cumulative progressSteady cumulative progress
» Ever closer approximation to “truth”Ever closer approximation to “truth”
Normal phaseNormal phase Revolutionary (extraordinary) phaseRevolutionary (extraordinary) phase
2007 Kutztown University 19
Phases of SciencePhases of Science NormalNormal
Ruling paradigm Ruling paradigm High degree of conceptual High degree of conceptual uniformityuniformity
Steady progress in:Steady progress in:» Accretion of knowledgeAccretion of knowledge
» Solving existing “puzzles”Solving existing “puzzles”
Pre-revolutionary :: mounting set of anomaliesPre-revolutionary :: mounting set of anomalies Revolutionary :: paradigm shiftRevolutionary :: paradigm shift
Alternative paradigm(s) proposedAlternative paradigm(s) proposed Period of competitionPeriod of competition Resolution – old paradigm dies outResolution – old paradigm dies out
2007 Kutztown University 20
Norwood Russell HansonNorwood Russell Hanson Observation is theory-ladenObservation is theory-laden Observation language and theory language Observation language and theory language
deeply interwovendeeply interwoven Historical & contemporary comprehension Historical & contemporary comprehension
deeply interwovendeeply interwoven Sought logic of discoverySought logic of discovery Note :: key issues for roboticsNote :: key issues for robotics
2007 Kutztown University 21
Patterns of DiscoveryPatterns of Discovery Norwood Russell HansonNorwood Russell Hanson Perception is theory-ladenPerception is theory-laden
Rock or clump of algae?Rock or clump of algae? Optical illusionsOptical illusions
Duck or rabbit?Duck or rabbit? Wiki articleWiki article Triangle puzzleTriangle puzzle
ỆỆ theoretical incommensurability theoretical incommensurability HansonHanson KuhnKuhn
2007 Kutztown University 22
Theory-laden-nessTheory-laden-ness Gestalt shiftGestalt shift From sensor data to perceptFrom sensor data to percept Role of theoryRole of theory
Pre-process sensor dataPre-process sensor data Organize perceptsOrganize percepts Basis of discovery plansBasis of discovery plans
ReasoningReasoning Deduction (includes mathematical induction)Deduction (includes mathematical induction) InductionInduction AbductionAbduction
2007 Kutztown University 23
Theory-laden-ness ExamplesTheory-laden-ness Examples RecognitionRecognition
EmbodiedEmbodied AutonomousAutonomous AgentAgent
Critical nature of perceptologyCritical nature of perceptology Shadow vs. cliffShadow vs. cliff Rock vs. algaeRock vs. algae The Measures TakenThe Measures Taken, Brecht, Brecht
2007 Kutztown University 24
Logos-telos-teleios TriadLogos-telos-teleios Triad
DesignDesign Basic componentsBasic components TeleologyTeleology Component interactionComponent interaction
2007 Kutztown University 25
Logos-telos-teleios TriadLogos-telos-teleios Triad Design – the oft missing component in Design – the oft missing component in
softwaresoftware Basic componentsBasic components
Logos – logic, internal structure, organizing Logos – logic, internal structure, organizing principleprinciple
Telos – goal, purpose, objective, aim, function, Telos – goal, purpose, objective, aim, function, intention, reason intention, reason
Teleios – complete, finished, mature, perfectedTeleios – complete, finished, mature, perfected
2007 Kutztown University 26
TeleologyTeleology
The philosophical study of purposeThe philosophical study of purpose The triad’s central relational kernelThe triad’s central relational kernel
2007 Kutztown University 27
Component InteractionComponent Interaction ReferenceReference LogosLogos
Mind, understandingMind, understanding Comprehend needComprehend need Generate purposeGenerate purpose Create designCreate design Determine level of achievementDetermine level of achievement
TelosTelos Provides focusProvides focus Provides measure of efficacy of designProvides measure of efficacy of design
TeleiosTeleios Circumscribes designCircumscribes design Measures artifact utilityMeasures artifact utility
2007 Kutztown University 28
Knowledge Representation for Knowledge Representation for Intelligent AgentsIntelligent Agents
Fall ’06 pptFall ’06 ppt