The Way Ahead New Directions at FBK ICT IRST Encounter Outstanding Visions from Across the Field Trento, December 17, 2012 CyberPhysical AI Systems for ResourceEfficient Living Wolfgang Wahlster Saarbrücken, Kaiserslautern, Bremen, Berlin, Osnabrück Phone: +49 (681) 857755252/4162 Fax: +49 (681) 857755383/5341 Email: wahlster@dfki.de WWW: http://www.dfki.de/~wahlster
After the Mathematization of all Sciences Today the Era of Informatization is in Full Swing Manufacturing Informatics Medical Informatics Energy Informatics Neuroinformatics Computational Linguistics Informatics: From Structures to Processes Media Informatics Legal Informatics Bioinformatics Automotive Informatics No Nobel Prize without ICT in Physics and Chemistry You can only talk about science, if there is a foundation in mathematics/informatics " Immanuel Kant, 1781
Foundations of Assistance Systems for ResourceEfficient Living Active Digital Product Memories 2. Evaluate 1. Sense RES COM Software Agents Cyberphysical AI Systems 3. Control
CPSS: Based on Wireless Adhoc M2M Communication of Autonomous SensorActuator Components Video Source: http://www.ted.com/talks/lang/en/david_merrill_demos_siftables_the_smart_blocks.html 1. Contextsensitive Component Behaviour 2. Dynamic Adaptation Based on Individual Role of the Component
CPS Hardware for a Digital Object Memory embedded or attached to a physical object Microsensor systems Microprocessor Radio Modules for Web Connectivity Memory Positioning Chips Actuators or Display Own Energy Supply or Energy Harvesting Unit
CPS Software for a Digital Object Memory embedded or attached to a physical object Sensor Interpretation Components State Transition and Processing Logic Components Positioning Chips Memory Management Functions Communication Interfaces User Interfaces Security Components
The German Future Project: Industry 4.0 Industrial production is the backbone of Germany s economic performance: jobs direct: 7,7 Million. indirect: 7,1 Million, every second job more than als 158 Billion trade surplus from export of industrial products (export : machine tool industry, automotive industry) Disruptive Paradigm Shift in Production based on the Future Internet 1. M2M and AllIP Factories are shifting from central MES to decentralized itemlevel production control 2. The embedded digital product memory tells the machines, which production services are needed for a particular emerging product. 3. Green and urban production based on cyberphysical production systems 4. Apps for softwaredefined products and smart product services Germany is preparing the 4th industrial revolution based on the Internet of Things, Cyberphysical Production Systems, and the Internet of Services in Real industry. PG Kommunikation DFKI der GmbH Forschungsunion
Outline of the Talk 1. From Embedded Systems to Cyberphysical AI Systems in the Smart Factory 2. The Role of Active Semantic Product Memories in Cyber Physical Production Systems 3. Semantic Web Services in a SOA Model of CyberPhysical Production Systems 4. P2P Energy Markets and Smart Micro Grids 5. Road Trains and Connecting Cars for ResourceEfficient Mobility 6. Conclusion
Towards Intelligent Environments based on the Internet of Things and Services 5) Intelligent Environments 4) Embedded Computers Smart Factory 1) Central Computer 3) Smart Phone 2) PC, Notebook Smart Card 90% of all computers are embedded 1 Computer Many Users 1 Computer 1 User Many Computers, 1 User 1941 1960 1980 2000 2020
Semantic Product Memories Covering the Complete Lifecycle of Products Smart Factory Smart Trucks The product memory prevents: tarnished chocolates Althaus Humidity Sensor melted chocolates Temperature Sensor crushed chocolates Pressure Sensor Smart Shop Private Home Smart Recycling
A Taxonomy of Digital Object Memories Internet of Things Mobile CyberPhysical Systems (CPS) Digital Object Memory (DOMe) Passive Digital Object Memory Active Digital Object Memory Digital Product Memory (DPM) Digital Maintenance Memory Semantic Product Memory Digital Transport Memory Internet of Services
Products with Integrated Dynamic Digital Storage, Sensing, and Wireless Communication Capabilities The product as an information container The product carries information across the complete supply chain and its lifecycle. The product as an agent The product affects its environment I was produced on 30 April 2010 and shipped on 3 May 2010 Grasp at the middle The product as an observer 2 mins open Please close! The product monitors itself and its environment
Future Project Industry 4.0 of German Chancellor Dr. Angela Merkel Internet of Things Vision: Internet der Dinge Intelligente Umgebungen Environments/Smart Spaces z.b. Smart Digital City 500 M for 3 Years National Program: 250 M Funding of Ministry for Research and Ministry for Economics CyberPhysical Systems Smart Factory, Smart Grid Networked Embedded Systems Intelligent Street Crossing Eingebettetes System (ES) Evolution from Embedded Systems to CyberPhysical Systems Embedded Systems Airbag National Roadmap Embedded Systems Agenda CyberPhysical Systems
The Internet of Things and Services as a Basis for the Smart Factories in the Industry 4.0 Internet of Services Smart Material App Plattform Internet der Dienste Semantic PLM, SCM, CRM, QMS and ERPServices Smart Factory App Plattform CyberPhysical Production Sytsem Smart Products App Plattform Internet of Things
The SmartFactory Shop Floor: Wireless, RFID, Sensor and Servicebased Architecture continuous flow process colored soap production discrete handling process bottling, handling, labeling, QC, packaging Live Webcam: http://www.smartfactory.de/webcam.de.html
DFKI s SmartFactory: The World s First Living Lab for CyberPhysical Production Systems
From Industry 1.0 to Industry 4.0: Towards the 4th Industrial Revolution First Mechanical Loom 1784 1. Industrial Revolution through introduction of mechanical production facilities powered by water and steam End of 18th Century 2. Industrial Revolution through introduction of mass production based on the division of labour powerde by electrical energy Start of 20th Century 3. Industrial Revolution through Introduction of electronics and IT for a further automization of production Start of 70ies 010001101 001010100 100101010 010010101 today 4. Industrial Revolution based on CyberPhysical Production Systems Industry 4.0 Industry 3.0 Industry 2.0 Industry 1.0 t Degree of Complexity
Interoperability for M2MCommunication in Industry 4.0 µ webserver 32bit ARM processor 8MB SDRAM 100Mbit Ethernet &WiFi LINUX on DIGI Connect OPCUA (Open Process Control Unified Architecture) Server for M2M The heart of an industrial CPS: XMLbased Web server or very fast binary communication based on the TCP protocols
From Bits and Bytes to Semantics driven by Electrical Engineering driven by Software Engineering Via functions To semantic services common ontology From bits and bytes SOAPprod_Core knowledge based Semantic Technologies driven by
Serviceoriented planning of plant systems Hardwareindependent planning of plant systems ERP Enterprise Resource Planning MES Manufacturing Execution System Field Layer Abstract Service hardwareindependent Service Library Device Control hardwaredependent SensorService ValveService PumpService ControlService CommunicationService Industry 4.0: AllIP Factories, no chaos of field buses, Internetbased Factory Networking based on IoS and IoT
Semantic Web Services for Industry 4.0: The Semantic SOA Model of the Smart Factory Physical Model SOAbased Smart Factory Functional Model Function Filling various drugs into individual pill boxes Function Function Fill Control quality Ultrasound Sensor Electronic Stopper Inductive Sensor Camera RFIDLSG WebService stopper Operation: hold IP 192.168.178.29 Function Stop pill box Service hold Function Detect pill box Service Directory Function Count pills Inductive Sensor Stopper Stopping Unit Services: hold release check RFIDLSG Services: read write Ultrasound Sensor Service: check Camera Service: count_pills Nach D. Zühlke, DFKI
The Structure of the Object Memory Model (OMM) Header Table of Contents Block 1 Block n Block Metadata Block Payload Block Metadata Block Link
ClosedLoop versus OpenLoop Product Memories Adhoc Data Formats Semantic Data Formats
The Smart Keyfinder with its Semantic Product Memory Chip Semantic Product Memory Chip in the backcover plastic frame with product specification Bluetooth circuit board with keyfinder logic packaged inside a plastic shell Personalized keychain with custom metal tag on the front produced by an engraving machine
The Intelligent Workpiece Carrier: a Complex CyberPhysical System
Key Components of a ServiceOriented Cyber Physical Production Systems Machine 1 Active Semantic Product Memory M2M Communication CNC Milling Machine M2M Communication Machine N Active Semantic Product Memory Production Service Discovery, Matching and Execution Workpiece Carrier 1 Active Semantic Product Memory Workpiece Carrier N Active Semantic Product Memory Production Pathplanning Basedon Semantic Product Memory Emerging Product 1 Active Semantic Product Memory Semantic Product Memory Emerging Product N Active Semantic Product Memory Top Shell Selection CircuitTop Shell Packaging RESCOM Engravature Top and Bottom Shell Assembly
RESCOM: The Use of Active Semantic Product Memories
Locationbased Industrial Assistance Systems in Smart Factories for Resource Efficiency Improvements
Industry 4.0: Smart, Green, and Urban Production Smart Production Highprecision, superior quality production of highmix, low volume smart products Green Production clean, resourceefficient, and sustainable Urban Production Smart Factories in the city close to the employees homes
Convergence of Three Networks for the Energy Transition (phasing out nuclear power > only renewable energy) Electricity Grid ICT Networks Gas Networks Future Networks for the Energy Transition
Basic Concept: Local Peer2Peer Energy Grid Trading power directly between end customers Peer Energy Cloud / Page 1
Virtual Power Plants: Cluster of Local Smart Micro Grids Locally generated energy is mainly used in the local environment: Photovoltaic Solar Thermal Energy Wind Power Bioenergy Geothermal Energy ClosedLoop Electric Ecosystem emarket, Billing Monitoring, Management Decentralized Load Management
Managing Network Capacity and Peak Loads After the Energy Transition renewable energies new loads and load management............... Voltage... new loads time / h Source: TU Dortmund RWE Deutschland AG x x Intelligent Energy Network
PeertoPeer Digital Markets for Renewable Energies in Local Smart Micro Grids Network Operators Network Service Grid Restrictions EEnergy Market Place Aggregator Demand Management Supply Management Electricity Market Energy Suppliers Energy Traders Prosumer (Active Customer) data Micro CHP electricity gas gas electricity
Collection of Individual Consumption Profiles by Machine Learning Methods with the Data of a Smart Meter Smart Meter dedicated fibre optic cable to cloud/municipal utilities
Smart Meter Disaggregation for NonIntrusive Load Monitoring Adapting Methods from Speech Recognition and Machine Learning Warm up + Espresso 2000 1800 1600 1400 1200 1000 800 600 400 200 0 Kaffeemaschine Coffee Machine Kühlschrank Refrigerator Geschirrspüler Dishwasher Bügeleisen Iron small coffee medium coffee large coffee 1 21 41 61 81 101 121 141 161 181 201 221 241 261 281 301 321 341 500 450 400 350 300 250 200 150 100 50 0 1 53 105 157 209 261 313 365 417 469 521 573 625 677 729 781 833 885 2500 2000 1500 1000 500 0 Waschzyklus 1 323 645 967 1289 1611 1933 2255 2577 2899 3221 3543 3865 4187 4509 4831 5153 2000 1800 1600 1400 1200 1000 800 600 400 200 0 1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 120 100 80 60 40 20 0 Flurbeleuchtung Beleuchtung Illumination Toaster Wasserkocher Kettle Ventilator Fan 1 15 29 43 57 71 85 99 113 127 141 155 169 183 197 211 225 239 900 800 700 600 500 400 300 200 100 0 1. Toast 2. Toast 1 12 23 34 45 56 67 78 89 100 111 122 133 144 155 166 177 188 2500 2000 1500 1000 500 0 1 Liter 1,5 Liter 1 16 31 46 61 76 91 106 121 136 151 166 181 196 211 226 241 256 50 45 40 35 30 25 20 15 10 5 0 different levels 1 28 55 82 109 136 163 190 217 244 271 298 325 352 379 406 433 460 487
DFKI s EO: ResourceEfficient and Connecting Electric Car
Relevant Book in the Springer Series Cognitive Technologies 2012
Conclusions 1. The energy transition needs smart networks and intelligent assistant systems cyberphysical AI systems are the foundation. 2. Resourceefficiency in production is the key and can be realized with cyberphysical production systems based on semantic services. 3. Local P2P agentbased energy markets can provide the necessary load balancing. 4. Active semantic product memories are a key element of solutions for resourceefficient living.
Thank you very much for your attention. Design by R.O.