Robots Introduction Based on the lecture by Dr. Hadi Moradi University of Southern California
Sep 25, 2015
Robots IntroductionBased on the lecture by Dr. Hadi MoradiUniversity of Southern California
OutlineControl ApproachesFeedback ControlCyberneticsBraitenberg VehiclesArtificial IntelligenceEarly robotsRobotics TodayWhy is Robotics hard
ControlSensing => ActionReactiveDont think, act: AnimalsDeliberativeThink hard, act later: ChessHybridThink and act in parallel: car racesBehavior-basedThink the way you act: human
Reactive SystemsCollection of sense-act rulesStimulus-responseAdvantages:?Disadvantages?
Reactive SystemsCollection of sense-act rulesStimulus-responseAdvantages:Inherently parallelNo/minimal stateVery fastNo memoryDisadvantagesNo planningNo learning
Deliberative Systems3 phase model:SensePlanActExample: ChessAdvantages:?Disadvantages:?
Deliberative Systems3 phase model:SensePlanActAdvantages:can planCan learnDisadvantages:Needs world modelSearching and planning are slowWorld model gets outdated
Feedback ControlReact to the sensor changesFeedback control == self-regulationQ: What type of control system is it?
Feedback types:PositiveNegative
- and + FeedbackNegative feedback:Regulates the state/outputExamples: Thermostat, bodies, Positive feedback:Amplifies the state/outputExamples: Stock marketThe first use: ancient Greek water systemRe-invented in the Renaissance for ovens
W. Grey Walters Tortoise1953 Machina SpeculatrixSensors1 photocell, 1 bump sensor2 motorsReactive control
W. Grey Walters TortoiseBehaviors: seeking light, head toward weak light, back away from bright light, turn and push (obstacle avoidance), recharge battery.Basis for creating adaptive behavior-based
Turtle PrinciplesParsimony: simple is better e.g., clever recharging strategyExploration/speculation: keeps moving except when chargingAttraction (positive tropism): motivation to approach light Aversion (negative tropism): motivation to avoid obstacles, slopes Discernment: ability to distinguish and make choices productive or unproductive behavior, adaptationDucking
Tortoise behaviorA path: a candle on top of the shell
Tortoise behaviorTwo turtles: Like dancing
New Tortoise
QuestionHow does it do the charging?Note: When the battery is low, it goes for the light.
Braitenberg VehiclesValentino Braitenberg early 1980sExtended Walters mode Based on analog circuits Direct connections between light sensors and motors Complex behaviors from very simple mechanisms
Braitenberg VehiclesComplex behaviors from very simple mechanisms
Braitenberg VehiclesBy varying the connections and their strengths, numerous behaviors result, e.g.: "fear/cowardice" - flees light "aggression" - charges into light "love" - following/hugging many others, up to memory and learning!Reactive control Later implemented on real robotsCheck: http://www.duke.edu/~mrz/braitenberg/braitenberg.htmlBots order Styrofoam cubes (16 min 30 sec)Tokyo Lecture 3 time 24:30-41:00
Brief History1750: Swiss craftsman create automatons with clockwork to play tunes1917: Word Robot appeard in Karel Capeks play1938: Issac Asimov wrote a novel about robots1958: Unimation (Universal Automation) co started making die-casting robots for GM1960: Machine vision studies started1966: First painting robot installed in Byrne, Norway.1966: U.S.A.s robotic spacecraft lands on moon.1978: First PUMA (Programmable Universal Assembly) robot developed by Unimation.1979: Japan introduces the SCARA (Selective Compliance Assembly Robot Arm).
Early Artificial Intelligence"Born" in 1955 at Dartmouth "Intelligent machine" would use internal models to search for solutions and then try them out (M. Minsky) => deliberative model! Planning became the tradition Explicit symbolic representations Hierarchical system organization Sequential execution
Artificial IntelligenceEarly AI had a strong impact on early robotics Focused on knowledge, internal models, and reasoning/planning Eventually (1980s) robotics developed more appropriate approaches => behavior-based and hybrid control AI itself has also evolved... Early robots used deliberative controlIntelligence through construction (5 min 20 sec)Tokyo Lecture 2 time 27:40-33:00