Post on 28-Mar-2015
transcript
German Research Center Center for Human-Machine-Interaction for Artificial Intelligence Human-Machine-Interaction
T d F t f ThiTowards a Factory of Things
Prof. Dr.-Ing. Detlef ZuehlkeChairman of the Executive BoardTechnology Initiative SmartFactory-KLgy yKaiserslautern/Germany
www.smartfactory.eu
OutlineOutline
Introduction• Introduction
KL• The SmartFactoryKL-Initiative
• R&D Projects
• Lessons learned
© SmartFactoryKL 2009 - 3the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Our future worldOur future world
Our life is deeply affected by many new technologies which have reached a sufficient level of maturity!
WLAN, Bluetooth, UMTS…SmartPhones, PDA´s, SubNotebooks…
Speech interaction gesture controlSpeech interaction, gesture control…
The internet of thingsFrom telephone to VoIP…
Information will be available
The internet of things…
Information will be availableanywhere, anytime, with any content, for any user
using any device and any access
© SmartFactoryKL 2009 - 4the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Complexity problemComplexity problem
Customerrequirements
in
requirements
shorter
C l ithardware software
Pl i
shorter lifecycles
Complexityrequirements logistics
in inPlanning Operation
global in
global competition
Businessprocesses
© SmartFactoryKL 2009 - 5the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Smart Technology Initiatives in many Application AreasSmart Technology Initiatives in many Application Areas
© SmartFactoryKL 2009 - 6the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
OutlineOutline
Introduction• Introduction
KL• The SmartFactoryKL-Initiative
• R&D Projects
• Lessons learned
© SmartFactoryKL 2009 - 7the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Development of the SmartFactoryKLDevelopment of the SmartFactory
factory in operationdesign and build
f b
national & international R&D projects
formationof the
2004 2005 2006 2007 20092008
feasab.study
information panelsidea
20102004 2005 2006 2007 20092008 2010
© SmartFactoryKL 2009 - 8the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Membership / Sponsorship SmartFactoryKLMembership / Sponsorship SmartFactory
First multi vendor research, development and demonstration center for industrial ICT
Goal: The integration of mature ICT into factory automation
Members:
controlgroup
digitalfactorygroup
processgroup
researchgroup
Sponsors:
© SmartFactoryKL 2009 - 9the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Floorplan SmartFactoryKLFloorplan SmartFactory
demonstrationFlow Colored Color MixingFresh Flushing Soap
RFID-Floor
Unit Water Dosage MixingWater
gWater
Fillin
gng e d
continuous flow process bulk
Supply
RFID Floor
Clo
sin
Labe
ling
Bot
tleFe
ed
k goods proce
operator control stationserver
staff
Com
mis
-si
onin
g
ss
staff
development
© SmartFactoryKL 2009 - 10the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Picture of the Shop FloorPicture of the Shop Floor
continuous flow processcolored soap production
discrete handling processbottling handling labeling QC packagingcolored soap production bottling, handling, labeling, QC, packaging…
Live-Webcam: http://www.smartfactory.de/webcam.de.htm
© SmartFactoryKL 2009 - 11the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
© SmartFactoryKL 2009 - 12the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
OutlineOutline
Introduction• Introduction
K• The SmartFactoryKL -Initiative
• R&D Projects
• Lessons learned
© SmartFactoryKL 2009 - 13the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Cluster Analysis – Fields of ResearchCluster Analysis Fields of Research
location based services
human machine interaction
bli t h l i
human machine interaction
enabling technologies
virtual plant design and
management
intelligentcollaboration and
documentation
interactiondevices
remote plant control
location sensing
technologies
© SmartFactoryKL 2009 - 14the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
The wireless SmartFactoryKLy
implemented technologies:• supervisory communication > WLAN• decentralized process control > RFID
SCADA
decentralized process control > RFID• wireless device networks > ZigBee, Bluetooth• failure messaging >GPRS• localisation > UWB
Ethernet
controlparameterizationset pset-up
© SmartFactoryKL 2009 - 15the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Wireless systemsy
GPRS/UMTS high quality of service expensivet bl lvery stable slow response
DECT good quality of service few devicesstable low data ratesstable low data rates
UWB ??? quality of service few devices??? stable low data rates
WLAN high data rates „everyone´s darling“cheap fair QoS, unstable
ence
!en
ce!
Bluetooth high quality of service high power consumptionvery stable, cheap low data rates, no roaming
Zigbee high q alit of ser ice to be de elopedand
coex
iste
and
coex
iste
Zigbee high quality of service to be developedvery stable, low energy low data rates
BaBa
© SmartFactoryKL 2009 - 16the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Wireless Problems
intercept
wireless network
intrudejam
© SmartFactoryKL 2009 - 17the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Wireless application fieldspp
Class ASafety critical operation
Class AEmergency stop, hazardous tasks
Time critical operationClass B
p
Network control systems, predictable response time
Class CNon critical operation
Nomadic information retrieval, „nice to have“ operation
© SmartFactoryKL 2009 - 18the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Lessons learned
Wireless• enabler for mobility• brings flexibility• may save installation costyBut it…• should be complemented by wireless power supply• needs intensive radio planning• needs more frequency resources (on the long run)• is vulnarable by environmental changes• can not offer real-time transmission• can be jammed easily and unnoticed
Wireless systems do not yet meet basic industrial requirements !Wireless systems do not yet meet basic industrial requirements !They offer advantages in non-critical applications only !
© SmartFactoryKL 2009 - 19the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Project Group: „Digital Factory“ j p „ g y
digitalworld
realworld
Objective Demonstration of a Digital Factory“ Demonstration of a „Digital FactoryMilestones Setup of a digital model of the existing factory
Simulation of the factory
PLM Software
Simulation of the factory
Automated control system generation
Simulation support during operation
Lucian BlagaUniversitySibiu/RO
Linking the ERP and MES Systems
Creation of a research-, development- and demonstration platform
/
© SmartFactoryKL 2009 - 20the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
demonstration platform
Project Group: „Digital Factory“Project Group: „Digital Factory
SimulationPlanning Control
(Factory Link)(Factory Link)(Factory CAD, Solid Edge, NX)(Factory CAD, Solid Edge, NX) (Plant Simulation, Process Simulate)(Plant Simulation, Process Simulate)
BackboneBackbone
OPC
BackboneBackbone
WinCC Projects,Operator Panels
PLC Code
Analysis,Dynamic 3D Vis
© SmartFactoryKL 2009 - 21the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Lessons learned
Digital factory• will reduce planning effort• will reduce time to market• allows for component re-usep• may save cost (on the long run)
But it…• needs a collaborative platform• needs intensive expertise by the users• must be linked to the control level• is for big players only
A digital factory tool world will make planning and operation moreg y p g peffective!
But it needs years to develop and new standards to ease integration!
© SmartFactoryKL 2009 - 22the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Future interaction with smart production environmentsFuture interaction with smart production environments
SmartFactoryKL as a testbed: - 20 field devices can be controlled wirelessly via Bluetooth
Dosing pumps
y- users can access them according to their tasks, education and preferences- dedicated remote operation device is under development
Dosing pumps
Mass-flow meters
Valves
FlowUnit components
Th
Weighing machines
Agitators The Nomadic
User
© SmartFactoryKL 2009 - 23the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Smartphones as universal interaction devicesSmartphones as universal interaction devices
Center forHuman-Machine-Interaction
TasksUser
Human Machine Interaction
Influencing Factorson InteractionDevices Functions
Context
Technological
Organisational
Environmental
- Exceptional expertise by the Center for Human-Machine-Interaction
- Research project by SmartFactoryKL
membersmembers
© SmartFactoryKL 2009 - 24the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Smartphones as universal interaction devicesSmartphones as universal interaction devices
Development of a platform-independent software for mobile phones of different manufacturers
Data link via Bluetooth Data link via Bluetooth
Automatic detection of available field devices
Communication via Communication via heterogeneous infrastructure withdifferent access paths and gatewaysgateways
Result:20 smartphones Java2ME> 4 were o.k.
© SmartFactoryKL 2009 - 25the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
> 6 had minor problems> 10 failed completely
Universal Interaction Device by UniPO GmbHUniversal Interaction Device by UniPO GmbH
The first industrial product…
Universal Gateway
USB 2.0
Dev1Dev2 DevxApp1App2 Appx ISO OSI
67
Command-
UART
Dev1Dev2 Devx
UniBeg-Applications
App1App2 Appx ISO-OSI
345
MCAP RFCommLayer
Handler
MCAPLayer
USB 2.0
12 Transmission
-HardwareTransmission-
Hardware
© SmartFactoryKL 2009 - 26the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
USB 2.0
The future challenge of designing user interfacesThe future challenge of designing user interfaces
Yesterday…Tomorrow…
Task:
• Local access• Known hardware
Designing agooduser interface
• Fixed dialogue• Limited functions• Known location
• Nomadic access• Changing hardware
• Lifetime 20 yrs.• Multiple dialogues• Complex functions• Fuzzy locations
Solution:Abstract descriptionModel-based
© SmartFactoryKL 2009 - 27the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
• Lifetime 1 yrs. Model-basedRun-time adaption
Different layers of UI abstractionDifferent layers of UI abstraction
Final userinterface (source
Abstract userinterface
Concrete userinterface
interface (sourcecode)
Use model (useML)
interface
Analysis data(useDDL)
© SmartFactoryKL 2009 - 28the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Transformation processes for platform specific developmentTransformation processes for platform specific development
PDAPDA--PDAPDAtransformertransformer PCPC--
transformertransformeruseMLuseML
SmartphoneSmartphone--t ft f
UIMLUIMLDISLDISL
Development environmentDevelopment environment
transformertransformer
© SmartFactoryKL 2009 - 29the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Lessons learned
User-friendly operation• supports the user to do it right• brings more flexibility• is important for customer satisfactionp
But it…• must be developed with methodologyp gy• must focus on the users tasks• should be hardware independant• should care for cultural differences• should be treated as important as hard- and software
Help the users to perform the tasks by a good usability!Help the users to perform the tasks by a good usability!Be prepared for rapid hardware changes and nomadic access!
© SmartFactoryKL 2009 - 30the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Location Based Services in Industrial Applicationspp
control room
processcontrol
locationcomponents logistics products
location locationlocation
location and contextb d i
context model
location
based services location model
task modellocationpeople
task model
service model
xxx model
© SmartFactoryKL 2009 - 31the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Indoor Positioning Systems installed in the SmartFactoryKLIndoor Positioning Systems installed in the SmartFactory
• Ubisense UWB-Realtime Positioning System
• RFID Grid for Mobile Workshop Navigation
• Cricket Ultrasonic Indoor L ti S tLocation System
© SmartFactoryKL 2009 - 32the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Location Based Services in Industrial Applicationspp
World levelWorld level Accuracy:
5-15mSystems: GPS (US)
GLONASS (russ.)GALILEO (europ from `12)
3 axesGALILEO (europ. from 12)cell phone localization
Accuracy: Systems: UbiSenseBuilding levelBuilding levely
0.5 – 1.5m / 3 -5°
4 axes
Systems: UbiSenseDOLPHINCricketWLAN localization
Accuracy:
1 – 10cm / 1.0°
Room levelRoom levelSystems: iGPS-Laser
UbiSenseUWB localization
Device levelDevice levelAccuracy:
4 - 5 axes
Systems: ??? 1 – 5mm / 0.1°
5 – 6 axes
© SmartFactoryKL 2009 - 33the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Lessons learned
Location-based services• will be essential in wireless applications• enable context and position dependancy• enable a more efficient use of resources
But it…• we need indoor location sensing systemsg y• we need location standards• needs a careful handling of sensitive information (privacy)
We still need a convincing Indoor GPS“!We still need a convincing „Indoor-GPS !We should seriously discuss the very sensitive privacy issues!
© SmartFactoryKL 2009 - 34the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Project: SoA-Architecture for the SmartFactoryKLProject: SoA Architecture for the SmartFactory
Structure Service repository mStructureISA95
B2MML ProcessBPEL
Service repositoryUDDI Orders
BPEL-Engine trol
sys
tem
BPELWebservices
WSDL
BPEL Engine
Java-invoker
HTTPS-Server
cont
rver
Process deployment
PDD
HTTPS Server
ACPLT/KS-Client
web
ser
sACPLT/KS-ServerPDD
SIEMENS-PCS7
S7 PLC´s
ield
dev
ices
fisensors
actors
© SmartFactoryKL 2009 - 35the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
SoA prerequisitesSoA prerequisites
planningaccounting PC´s
supervisorysupervisorycontrol PC´s
machinecontrol
PC´sPLC´s??
device µWS µWS µWS µWS µWS
µWS
devicecontrol
webserversµWS
© SmartFactoryKL 2009 - 36the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
SoA structure and problemsSoA structure and problems
Problems:enterprise services
Problems:• no experience• no trust• no experts• no experts
domain services
A successful implementation ofSoA requires the company-wide definition of tasks and services
local services first !!Tools and software come second !
© SmartFactoryKL 2009 - 37the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Lessons learned
SoA architectures• will decouple hardware from control engineering• will bring more agility• will focus on a semantic control level
But it…• may have an unpredicable behavioury p• needs new expertise of users• needs smart webservers in each device• needs well defined service standards• revolutionizes control engineering
SoA enables the abstract design of more agile systems!g g ySoA is a new engineering paradigm and needs a different way of
thinking!
© SmartFactoryKL 2009 - 38the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
OutlineOutline
Introduction• Introduction
K• The SmartFactoryKL -Initiative
• R&D Projects
• Lessons learned
© SmartFactoryKL 2009 - 39the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Physical Device RepresentationModel Based Device Representationy p
Physical Device Model
p
DeviceService-Model- elementary services- working logic- . . .
Communication-Model- layer- protocol
. . .
Product-Data- part number
part description
- . . .
CAD-Data- mech. data
- part description- . . .
mech. data- electr. data- . . .
© SmartFactoryKL 2009 - 41the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
The Pyramid of Automation yesterday y y yWLAN
ERP-LevelEnterprise Resource Planningdata sets
MES L l
control room
MES-LevelManufacturing Execution System
words
Control-LevelMachine controllers
bytes
power Device-bi
processes logisticsmanufacturing
i t
LevelSensor-Actor-Machine
bits
© SmartFactoryKL 2009 - 42the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
gmaintenance
The Pyramid of Automation today y y
To services
Service repository
UDDI
WLAN
ERP-LevelEnterprise Resource Planning
MES L l
To servicesCommunication layerEthernet
IP 183.77.19.0
control room
MES-LevelManufacturing Execution SystemVia functions
IP 183.77.xx.x
Via data Control-LevelMachine controllers
Communication layerProfinet
powerUMTS Device-
From
Communication layerASI, CAN, Interbus…
processes logisticsmanufacturing
i t
UMTS
LevelSensor-Actor-Machine
From signals
© SmartFactoryKL 2009 - 43the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
gmaintenance
The Pyramid Network of Automation Services tomorrow
customer-services
IP 103.511.10.0
P t lP t lfrom services
y
WLANbluetooth
bluetooth
services
IP 183.77.19.0 IP 183.70…..
PortalPortalfrom services
control room
MES-servicesWLAN
ERP-services
logistics
UMTS
global servicesIP 183.77.19.0 IP 183.77…..
Service repository
UDDI
WLAN
maintenance
UMTSIP 183.77.18.. IP 171.77…..
IP 183.57…..
domain services
ProcessProcessto services
???
ZigBeebluetooth
ZigBeemanufacturing
maintenance
local services
IP 185.68….
IP 183.77…..
processesdevice services
IP 183.77…..
IP 183.77…..ResourceResource
© SmartFactoryKL 2009 - 44the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Key changes in the factory of the futureKey changes in the factory of the future
eyerything is described by described by
semantic servicesread_temp<valve_in4>
th f t f thieyerything eyerything
IP 183.77.18..
the factory of thingsy y gis mobile
y y gis networked
virtualityintelligentcollaboration and
documentation
interactiondevices
virtualitymerges with
reality
© SmartFactoryKL 2009 - 45the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
reality
Requirements and challengesRequirements and challenges
higher agility higher mobility shorterproduct lifecycles improved quality
Market Requirements
Wireless networks simulation Low-power electronics
Enabling TechnologiesPlug ´n playSmart devices SoA Location sensing
Semantic web
authenticationWireless networks Low power electronics
SoA Location sensing
worker fast product networkSolutions
intelligent
digital factories workermobility
fast productchanges
networkproduction
Complexity Education Safety Standardizationgcollaboration and
documentationHealthrisks
Privacyconcerns
Definitionof work
Quality of service
Complexity Trainingy
SecurityStandardization
© SmartFactoryKL 2009 - 46the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
We need new engineering paradigmsg g p g
the 00´sUseware
Modeling interactionobjects and systems
Placing buttons, switchesand windows
fromUseware
engineeringobjects and systemsand windows
ALARM 81
<action>
<comm_obj>:alarm
tfrom
Drawings Modeling mechatronic objects and systems
Mechanicalengineering
tothe 90´sWriting code
Modeling objects and systemsSoftware
engineeringIF $B5 EQ #0A
El t i lD i i t i t
the 80´s
engineeringTHEN GOTO… UML
Electricalengineering
Modeling functions and systemsDesigning transistors, gates and circuits
ARCHITECTURE xx67 OF nor_gate IS BEGIN y <= a NOR b; END xx67;
VHDL
© SmartFactoryKL 2009 - 47the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Towards Future ProductionTowards Future Production
Productivity
Smart Agile Factories
C t i t t d M f t i
Global Production Systems
Computer-integrated Manufacturing
Fordism & Taylorism
Mass Production
© SmartFactoryKL 2009 - 48the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Paradigms of the Factory of ThingsParadigms of the Factory of Things
Transparent and
Increase ofIncrease of
Highly ReconfigurableProcesses
pintegrated Processes
Context-sensitive Assistance
Explicit Process and Product Description
By means of Semantics
Increase of ProductivityIncrease of Productivity
Global, standardisedCommunication Architecture
by means of SOA
Optimization of Process QualityOptimization of Process Quality
Structural Flexibilityby means of Modularity
Efficient Use of Resources
Efficient Use of Resources
© SmartFactoryKL 2009 - 50the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Horizontal and Vertical IntegrationHorizontal and Vertical Integration
grat
ion
Business P
Quantity
cal I
nteg
Production
Process
Vert
ic PlanningHorizontal Integration Flexibility
FoTProductPlanning
ProductDevelopment
ProductIn Use
ProductRecycling
Automation & Control
Costs
Smart Devices
Quality Time
© SmartFactoryKL 2009 - 51the intelligent factory of the futureGerman Research Center
for Artificial Intelligence
Devices
shaping the futuref f t i ICTof manufacturing ICT
© SmartFactoryKL 2009 - 52the intelligent factory of the futureGerman Research Center
for Artificial Intelligence