Intro to robotics

Mr. LausterMr. Lauster

NAI Technology and NAI Technology and Engineering EducationEngineering Education

Robotics IRobotics I

Introduction to Robotics

Goals:Goals:

Understand course content and Understand course content and sequence.sequence.

Know definition of Robotics???Know definition of Robotics??? Understand the 4 generations of Understand the 4 generations of

RoboticsRobotics Know various application of Robotics Know various application of Robotics

in the real world. in the real world.

Robotics I SyllabusRobotics I Syllabus

Course DescriptionCourse Description Students will acquire a basic understanding of types of Students will acquire a basic understanding of types of

robots, how they operate, and their application in the real robots, how they operate, and their application in the real world. This is a hands-on project based course introduces the world. This is a hands-on project based course introduces the student to generations of robots through a unique curriculum student to generations of robots through a unique curriculum collaboration with Carnegie Mellon University. Classroom and collaboration with Carnegie Mellon University. Classroom and lab activities will include assembling and operating robotic lab activities will include assembling and operating robotic systems, building and programming mobile robots using systems, building and programming mobile robots using LEGO NXT Mindstorm® and Vex Robotics® systems with LEGO NXT Mindstorm® and Vex Robotics® systems with Robot C. Furthermore students will design and build various Robot C. Furthermore students will design and build various robots to complete many challenges including mazes, robots to complete many challenges including mazes, obstacle courses, robot battles and other unique obstacle courses, robot battles and other unique competitions. Students will design and produce custom competitions. Students will design and produce custom robotic components utilizing a CNC router, laser engraver, robotic components utilizing a CNC router, laser engraver, and 3D printer. S.T.E.M. (Science, Technology, Engineering, and 3D printer. S.T.E.M. (Science, Technology, Engineering, and Math) concepts and TSA (Technology Student and Math) concepts and TSA (Technology Student Association) problems will be addressed throughout the Association) problems will be addressed throughout the course.course.

Famous RobotsFamous Robots

B9 Robot B9 Robot

Lost in SpaceLost in Space

R2-D2R2-D2

Star WarsStar Wars

BenderBender

FuturamaFuturama

T-1000 T-1000

TerminatorTerminator

Mecha-GodzillaMecha-Godzilla

GodzillaGodzilla

Famous Robots…Fact or Famous Robots…Fact or Fiction Fiction

Robotics as characters and content Robotics as characters and content in media.in media.

Robotic devices used in the Robotic devices used in the entertainment industry. entertainment industry.

Science fiction is turning into science Science fiction is turning into science fact with many applications of fact with many applications of robotics in industry. robotics in industry.

• Lego Mindstorms Lego Mindstorms • NXT Robotics ChallengesNXT Robotics Challenges• CMU ChallengesCMU Challenges• TSA Problem SolvingTSA Problem Solving• Battle BotsBattle Bots

• SumoSumo• MechaMecha

• Vex IQ and Cortex Robotic Vex IQ and Cortex Robotic SystemsSystems

• RC ProgramingRC Programing• Pneumatic and Motor Pneumatic and Motor

ControlControl• Robotic Virtual WorldsRobotic Virtual Worlds• TSA Problem SolvingTSA Problem Solving• Robot C ProgramingRobot C Programing

• Laser EngraverLaser Engraver• Robot Name PlatesRobot Name Plates• Custom Key ChainCustom Key Chain• End EffectorsEnd Effectors

• 3D Printer3D Printer• Individual ProjectIndividual Project• Custom Chassis PartCustom Chassis Part• Armatrol Robotic ArmArmatrol Robotic Arm

• Pick and Place ProgramPick and Place Program• Robot JengaRobot Jenga

Robotics I – Robotics I – Projects and Projects and ChallengesChallenges

Fixed RoboticsFixed Robotics Mobile RoboticsMobile Robotics

What is a Robot? What is a Robot? • A Robot is a reprogrammable, A Robot is a reprogrammable,

multifunctional device, designed to multifunctional device, designed to move materials, parts tools or move materials, parts tools or specialized devices through variable specialized devices through variable programmed motions for the programmed motions for the performance of a variety of tasks.performance of a variety of tasks.• Robot Institute of America Robot Institute of America

What is a Robot?What is a Robot?• Are the following robots?Are the following robots?

• Digital ThermostatDigital Thermostat• iPhoneiPhone• 1908 Ford Model T1908 Ford Model T• 2008 Ford Focus2008 Ford Focus

• Yes or NoYes or No• More robotic characteristicsMore robotic characteristics• Less robotic characteristicsLess robotic characteristics

Robotics and economy Robotics and economy

• Robotics and embedded systems Robotics and embedded systems make up a 500 billion dollar make up a 500 billion dollar emerging industry that will have the emerging industry that will have the same impact on the new economy as same impact on the new economy as mass production had on the mass production had on the industrial revolution and the industrial revolution and the computer had on the information computer had on the information age. age.

Robotics and embedded systems make up a 500 billion dollar emerging industry that will have the same impact on the new economy as mass production had on the industrial revolution and the computer had on the information age.

Size of Robotics Market

Genereations of RoboticsGenereations of Robotics• First Generation- Remote ControlledFirst Generation- Remote Controlled• Second Generation – AutonomousSecond Generation – Autonomous• Third Generation – Networked Third Generation – Networked • Forth Generation – Artificial IntelligenceForth Generation – Artificial Intelligence

The fourth generation of The fourth generation of robotics has robots making robotics has robots making real time decisions based on real time decisions based on their environment. Artificial their environment. Artificial Intelligence is still in the future. Intelligence is still in the future. Pictured at the left is “Boss” Pictured at the left is “Boss” Carnegie Mellon’s 2007 Carnegie Mellon’s 2007 DARPA Grand Challenge DARPA Grand Challenge entry. entry.

First Generation Robot

Remote Control

The robot pictured at the left was developed to go into Three Mile Island to inspect after the nuclear disaster. Although it is a very complex machine it is designed to be controlled remotely.

Robot ClassificationsRobot Classifications

Second Generation RobotAutonomous

The three robots pictured above are designed to work without operator intervention. Currently robots still need operator supervision!

Robot ClassificationsRobot Classifications

Autonomous LoadingAutonomous Loading Automated HarvestingAutomated Harvesting Automated MowingAutomated Mowing

Third Generation RobotNetworked Robots

If we are ever going to colonize another planet, robots, not people, will build the structures that we live in. Pictured at the left is a Carnegie Mellon research project that shows one robot acting as a supervisor and two other robots working together to assemble a group of parts.

Robot ClassificationsRobot Classifications

Fourth Generation RobotArtificial Intelligence

The fourth generation of robotics has robots making real time decisions based on their environment. Artificial Intelligence is still in the future. Pictured at the left is “Boss” Carnegie Mellon’s 2007 DARPA Grand Challenge entry.

Robot ClassificationsRobot Classifications

Three Robotic Applications

Robotic systems are everywhere. This section will highlight three robot projects that serve three different industries:• The shipping industry• Mining• Healthcare

Project #1 Ultrastrip ProjectProblem: Stripping Paint from Very Large Ship HullsShip owners now pay $400K - $800K to strip over 6 acres of painted surface with 150 sandblasting workers. They take 2 - 3 days, need a forest of scaffolding, damage the ship's surface, produce clouds of sand and 40 pounds of toxic sludge per square foot stripped, and halt other drydock work by posing a health hazard. World market: over $100,000,000 per year. Major cost: shipping time lost in drydock.

Traditional Ship Cleaning

• Hazardous to operators/others

• Slow, noisy • Contaminates steel • Huge waste stream• Inconsistent cutting• Limits adjacent operations

Robotics Ship Cleaning First Generation Testing at the

NREC

Pictured above is a robotic solution for stripping paint. This robot was rigorously tested at the NREC over a two year period. The next page shows the test-bed that was set up to simulate real world conditions.

At the NREC hundreds of thousands of dollars are dedicated to realistic environments to test in

Test Bed for Robotic Paint Stripping

¾” plates of steel welded together to simulate the hull of a ship

Testing the robot Testing the robot on a real ship!on a real ship!

More testing the More testing the robot on a real ship robot on a real ship

with a second with a second generation robot.generation robot.

Successful completion of any robotics project takes thousands of hours of effort over several years. The next two slides demonstrate the constant improvement of the technology.

Interesting fact: Robotics projects are calculated in man-years.

UltraStrip Success

3rd Generation System: Chopper

4WD, 4WS, traction control provides smooth 0-20”/s (0.5m/s)

Long-life NEMA 17 sealed motors

Belly magnets don’t wear or mar,

provide robust adhesion

High traction, non-marring wheels

Floating head and suspension for

obstacles and curves

Electric and air drives (no hydraulics)

Flexible seal allows close cut-in

Corrosion resistant stainless & aluminum

15” (0.4m) cut width

Two-position head for strip and sweep

Areas Under Improvement

BRAKES

CROSS LINK

WHEELS

CABLE MANAGEMENT

FRAME

SIDEPLATES

CHAIN TENSIONING

SEAL

CONNECTORS

PROXIMITY SENSING

FASTENERS

IMPROVEDACTUATORS

MAGNETIMPROVEMENTS

Robotic Paint Stripping

Project #2 MiningProblem: Dangerous and Unprofitable US Coal MiningThe US is a world leader in coal production, but profits keep shrinking. Utility deregulation presses prices down while mining from smaller and shorter seams increases costs. Poor visibility limits efficiency, as do the safety precautions which can't prevent thousands of yearly casualties. Only new technology can increase safety and profits and minimize dependence on foreign energy.

Longwall Mining & Conveyor Belts

• ~100 Long wall mines worldwide• $1000/minute revenue generation• Belt problems are #1 cause of downtime

Belt Economics

Mechanical Splice Failures

Each mine has ~ 10 belt systems

Each mine breaks ~ 2 belts per month

Voluntary replacement -> ~30-40 minutes downtime

Unplanned failure -> ~8 hours

Lost revenue: $30,000 to $300,000/break

Belt Inspection System - Capabilities

• Detect the belt splices and other belt defects.• Generate a digital image of the splice or defect, display on the

user interface and publish the image over the network to a web based application.

• Determine the deterioration of the splice or defect.• Alert mine personnel to the defect in the belt and location for

repair.

Belt Inspection System

Belt runs at 800 feet per minute; Camera 9000 times per second

Current Status

• 5 Belt Inspection Systems Deployed• 14 months cumulative 24 hr operation• More than 60,000 miles belt inspected• Licensing negotiations underway• Contract with DOE to extend inspection to

vulcanized splices

Problem: Developing Better Drugs Requires Costly, Subjective Behavioral Studies Developing safe drugs to alleviate human suffering requires prior behavioral study of animal subjects. These studies by skilled human observers are labor-intensive and yield data prone to variations among them. An optimized automated system for studying animal behavior would reduce the time and cost needed to develop more effective drugs. $6.2 billion per year is spent on Central Nervous System R&D in the US alone. Millions of compounds are still untested for CNS effects.

From the Mines to Pharmaceuticals

• U.S. Pharmaceutical companies spend $6.2B/yr on R&D• Manual scoring of animal behavior is the primary bottleneck to pre-clinical

drug discovery• Targeting a 100x throughput improvement of pre-clinical drug discovery

through automation and data mining.

Innovative Process Improvements

Problem

SmartCubeTM ConceptScientists are working to develop an automated system for the understanding of animal behavior and drug effects.

The SmartCube enables high-throughput, industrial-scale testing of potential new drugs and will greatly speed the discovery of new drugs for common psychological disorders.

Millions ofCandidateCNS Drug

Compounds

Thousands ofSmartCube

Systems

CNS DrugCompound

“Leads” for FurtherInvestigation

Advanced DataMining AlgorithmsConvert Behaviors

to “Drug Signatures”

BehavioralData

Image ProcessingClassifies Animal

Behavior

Robot Application

After two years of effort, the system is fully operational:• 13 SmartCube systems in daily use• 0ver 15,000 hours of video data automatically scored• Over 20 common mouse behaviors recognized• More than 1000 statistical features automatically

analyzed• Excellent agreement on mouse behaviors with human

expert• Very good classification of drug types and doses in

major, blind trials (1200+ mice)

Today’s Robotics Markets• Construction• Mining• Transportation• Agriculture• Senior Care

• Medical / Biotech• Material Handling• Utilities• Education• Military

Robots and embedded systems are ubiquitous in today’s world!