Excerpt of Market Intelligence Report Robot Vision 2012

Market

Robot Vision 2011/2012

Excerpt of the Market Report„Robot Vision 2011/2012

All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the author.This prohibition includes copying by a firm for its internal use. © 201

For further information please contact us un

AMC HOFMANNPO BOX 115664629 HEPPENHEIM, GERMANYTELEPHONE: +49 (0) 1577 530 696EMAIL: [email protected]: WWW.AMC-HOFMANN.COM

arket Intelligence Report (


Excerpt of the Market Report„Robot Vision 2011/2012“

All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the author.This prohibition includes copying by a firm for its internal use. © 2012 by AMC Hofmann

For further information please contact us under [email protected] .

RMANY+49 (0) 1577 530 6969

HOFMANN.COMHOFMANN.COM

eport (MIR)


Excerpt of the Market Report

All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the author.

mailto:info:@amc-hofmann.com

mailto:HOFMANN:@AMC-HOFMANN.COM

http://WWW.AMC-HOFMANN.COM/

Market Intelligence Report Robot Vision 2012

©AMC Hofmann Automation – Market – Competence Page - 2 -

TABLE OF CONTENTS

Introduction ............................................................................................................................................ 3Why this Report? .......................................................................................................................................... 3

Goals of the market report ............................................................................................................................ 3

Industries involved ........................................................................................................................................ 4

Companies involved by Core Competence .................................................................................................. 4

Regions represented .................................................................................................................................... 5

Analysis in this report ................................................................................................................................... 6

Generation of data - Research ..................................................................................................................... 6

Definition of terms* ....................................................................................................................................... 8

1.0 Robot Vision- Technologies ........................................................................................................... 91.1 Detection Technologies .......................................................................................................................... 9

1.2 Algorithms ............................................................................................................................................ 11

1.3 Handling ............................................................................................................................................... 12

1.4 Robot Vision Turnkey- Concepts .......................................................................................................... 14

Introduction........................................................................................................................................ 14

Basic concepts for system set-up: ..................................................................................................... 14

2.0 Robot Vision Applications Overview ........................................................................................... 152.1 Introduction ........................................................................................................................................... 15

2.2 Applications Overview .......................................................................................................................... 15

2.3 Assembly- 2D & 3D & Best Fit .............................................................................................................. 17

3.0 Robot Vision Market Overview .................................................................................................... 193.1 Introduction ........................................................................................................................................... 19

3.2 Industries Overview .............................................................................................................................. 19

3.3 Aerospace ............................................................................................................................................ 21

4.0 Future Trends in Robot Vision and Machine Vision ................................................................. 224.1 Introduction ........................................................................................................................................... 23

4.1.1 Technologies ............................................................................................................................ 25

4.1.2 User Requirements .................................................................................................................. 26

4.1.3 Applications .............................................................................................................................. 27

4.1.4 Outlook ..................................................................................................................................... 28

5.0 Vendors Overviews ....................................................................................................................... 295.1 Content of The Survey ......................................................................................................................... 29

5.2 Core Competence by vendors .............................................................................................................. 30

5.3 Business Region: distribution by vendors ............................................................................................ 31

5.4 Industries- distribution by vendors ....................................................................................................... 32

5.5 No. of Robot Vision systems installed - overall ..................................................................................... 34

Copyright .................................................................................................................................................... 35



INTRODUCTION

WHY THIS REPORT?

Discussions with vendors and users of automation technologies about their needs for information has shown thatthe technology of robot vision already plays an important role for automation. New interesting markets as well asthe boost that came with 3D have influenced and grown the world of robot vision and its future development. Tocollect, show and analyze all dimensions of robot vision is the goal of this report.

The content topics selected for the report is a direct result of many discussions with end users and vendors ofMachine Vision and vision based automation equipment.

The Market Intelligence Report (MIR) “Robot Vision” therefore addresses all robot vision related topics andtrends.

GOALS OF THE MARKET REPORT

Provide an overview and show a “big picture” of the report´s topics as it is not

available from other sources.

Provide market intelligence, analysis and suggestions, based on data provided

by a large number of key market players.

Provide an overview and outlook on trends in Robot Vision and related

industries and applications where “robot vision” is used.

Keep you up to date on the current status and inform you on new interesting

topics for the world of robot vision.

Provide information and contact data and overviews on vendors, suppliers and

other sources for further information.



INDUSTRIES INVOLVEDIn order to get a complete overview of possible trends, applications and needs, the following industries have beeninvolved in the survey.

Aerospace

Agricultural & Forestry

Automotiveo BIW & Press shopo Assemblyo Logisticso Powertraino Paint shopo Tier 0.5 & Tier 1

Beverage

Ceramics

Consumer Goods

Electronics & Semiconductor

Entertainment

Food

Foundry Glass

Lab Automation

Life Science

Logistics & Transportation

Medical Engineering

Medical & Health Care

Metal Processing

Military & Defense

Mining

Packaging

Paper

Pharmaceuticals Photovoltaic

Plastics

Recycling

Service Robotics

Security & Surveillance

Textile

Wind Energy

Wood

COMPANIES INVOLVED BY CORE COMPETENCE

In order to give broad overviews, provide new information and see trends coming down the road from manydirections, the companies involved in the report represent several different core competences emerging indifferent regions and many industries as well.

In addition to the company types shown in the chart below, the views of end users are of high importance for ourreports. Before starting a new project the content is first discussed with end users relevant for the topic. Within thesurvey end users are questioned as well. Due to the fact that the end users typically do not want to be listedwithin the reports we respect that but do use and count their suggestions and input.

The chart below shows the fields of business named as core competence by the vendors involved.

8 8 ,9 %

5 2 ,2 % 5 3 ,3 %

1 5 ,6 %

4 1 ,1 %

7 ,8 %

0 %

1 0 %2 0 %

3 0 %4 0 %

5 0 %6 0 %

7 0 %8 0 %

9 0 %1 0 0 %

M ac hineVision

Robotic s &Position ing

Syste mInte gratio n

M ac hineBu ild ing

A utomation L ine Build ing

C ore com pe t e nce of the Com panie s

As already recognized in other reports we see again the strong commitment of “automationvendors” to Machine Vision. More and more companies having their roots outside MachineVision have now made it one of their core competences.



REGIONS REPRESENTED

The chart below shows the regions where the vendors are active and present. The numbers do not differentiatebetween legal entities of a company and representatives or partners.

0% 20% 40% 60% 80% 100%

Germany, Austria, Switzerland

Northern Europe

W estern Europe

Central Europe

Southern Europe

North America

South America

Middle East / Africa

China

Japan

India

Russia

Australia / Oceania

77,9%

67,5%

63,4%

59,3%

57,8%

63,1%

37,8%

21,3%

36,6%

29,1%

32,1%

25,0%

22,1%

Regions where vendors are present and active

25% of the companies claim to be already active in Russia, or are from Russia.21% of the companies claim to be already active in Middle East / Africa



ANALYSIS IN THIS REPORT

Analysis of the data requires a clear understanding of the sources and their technological roots. Therefore eachcompany was classified by three parameters.

Industry, where active

Core Competence (Technology)

Regions, where active

In order to have the information represented best, multiple selections were possible and allowed.

Questioning companies being active in many different industries brings the benefit to get a very broad informationbase, but makes it necessary to take a deeper look at the sources of the information to get a clear understandingof the results and conclusions.

Counting votes:

In several statistics company votes can be counted several times when theyare active in several industries, regions or markets. Same with the corecompetence of a company.

When having core competence in robotics, Machine Vision and systemintegration their vote is counted in each section. Overall statistics on such atopic where all attending companies are counted would show only one vote ofthis company.

GENERATION OF DATA - RESEARCH

Data used: All information used is either public or provided and submitted by the companies included in thesurvey. Therefore the correctness of the information is obliged to these sources. Data comprised in the report hasbeen mainly generated by direct interviews conducted with the leaders in this field of automation. Additionally, theresults of further research with a variety of sources such as internet, trade shows, presentations, publications andconferences have been included in order to complete a very thorough and comprehensive overview.

The survey:

Companies attending the survey have been asked to rate all topics shown inthis report, based on expected future potential for growth. Available ratingshave been:

NoneLowAverageHigh

Multiple selections have not been allowed.



Data handling and analysis within the tables and charts:

Due to the fact that not all information or statements were always givenby all of the companies, the numbers in the charts do not necessarily addup to 100%

Note:

In order to give a broad and complete overview information in this reportare not limited to the information provided by of the questionnaires.

Material available in form of presentations or from websites is added aswell. All sources used are listed in the appendix.

Methods: Data generation of information was done by the following methods

Web research

Literature research

Direct personal communication with vendors and end users

Questionnaires sent out to vendors

Questionnaires sent out to end users

Geography: In order to get an international overview, companies from the following regions have been involvedin this survey:

Europe

North America

Asia

India

Russia

Australia & New Zealand

Confidentiality: All information / data published in this report has either been confirmed or publicly released bythe providing companies for publishing in this report or derived from public sources.



DEFINITION OF TERMS*

Many times terms are used in different ways. To avoid misunderstanding please read the following definitions.

Detection unit- For this report “detection unit” is defined as the unit generating image information. The detectionunit can consist of a full set- up of illumination and optics, combined with a single sensor, several sensors or aset-up of a scanner and single cameras.

OEM- Abbreviation for „original equipment manufacturer“. However, car manufacturers refer also to themselvesas OEM. In this report the term OEM is used to describe automation equipment manufacturers, including machinebuilders, with a high degree of standardization. A robot manufacturer for example would be an OEM from themachine vision point of view.

Sensor- is generically used in this report to describe the various kinds of optical units integrated within onehousing. This includes the entire range from simple standard sensors to stereo-optic and triangulation sensors.

T0.5 & T1- Tier 0.5 & Tier 1 are terms specifying key automotive suppliers, classified into 0.5, 1 and 2 dependingon their scope of delivery and the extent of involvement in the processes of the car manufacturers.

Powertrain- Powertrain is the term for automotive manufacturing of parts that are used in underbody frame andengine and transmission. Powertrain includes the car manufacturers´ facilities as well as the Tier Supplier.

Random bin picking- Random bin picking many times is used instead of bin picking to indicate that the goods ina bin are without any known or defined orientation.

UAV- Unmanned Aerial Vehicle guided either by on board sensors and processors or guided remotely through aman-machine interface.

UGV- Unmanned Ground Vehicle, guided either by on board sensors and processors or guided remotely througha man-machine interface.

General Industries- Especially companies supplying equipment to “industry” classify and structure theindustries. Due to the large amount of industries many vendors structure is based on their business and theirfocus. In many cases the term General Industries is used to describe and summarize many industries that are notcovered by the other industries classified. So General Industries includes different content depending on thecompany you are talking to.

Robot Vision- The term Robot Vision depends on the terms Robot and Vision (for Machine Vision). This bringsus to the question - what is the definition of a robot? Pictures below show some vehicles and mechanicaldevices described as robots.

To come to a common understanding and conclusion on what is a robot here is the explanation from Wikipedia:“A robot is a mechanical contraption which can perform tasks on its own, or with guidance.”

So the definition of Robot Vision, within our reports, is based on this definition which remains a little fuzzy. RobotVision is the automated guidance of a robot, handling device or moving vehicle- guidance can be by MachineVision technology but does not have to be limited to that. Other technologies for example radar or ultrasonictechnology can be used as well.



1.0 ROBOT VISION- TECHNOLOGIES

The more challenging an application is, the more important the clear understanding of all involved sub-systems becomes. Robot Vision applications consist of many technologies and components that aremore or less important depending on the specific application.

Therefore this report explains the following topics, with a strong focus on the machine vision topics.

Detection (Machine Vision / Scanning)

Algorithms (Machine Vision)

Handling & Robotics

Basic concepts for system integration

1.1 DETECTION TECHNOLOGIES

Starting in the semiconductor and electronics business in the late 1980´s machine vision was focusedon 2D position measurement mainly in combination with axes handling systems and more and moreSCARA robots. In the late 1990´s when machine vision companies introduced the first 3D visiontechnologies. Applications where machine vision systems guided 6- axes robots started to be popular.

At this time, companies like Perceptron (www.perceptron.com ), LMI Technologies (www.lmi.com ),Isra Vision (www.isravision.com ), VMT (www.vmt-gmbh.com ), HGV Vosseler (www.hgv.de) andVitronic (www.vitronic.com ) successfully developed the first 3D vision technologies for robot guidanceusing Photogrammetry, Laser triangulation and Stereometry. This was done primarily to meet theautomation needs of the automotive industry.

At the same time, sensor companies such as Sick, Leuze Electronics, Micro-Epsilon and Keyencestarted to introduce Laser scanning mainly in the field of displacement sensors.

Time of Flight (ToF) cameras have recently become available and are getting more and more popularfor machine vision applications from vendors such as Baumer, PMD, Mesa and Canesta.

In many business fields different technologies and concepts compete, just look for example atcommunication interfaces. In 3D machine vision technology this is different. All of these technologieshave their strengths and weaknesses and therefore are needed. Sometimes the right combination ofseveral of these technologies can lead to the best solution.

http://www.perceptron.com/

http://www.lmi.com/

http://www.isravision.com/

http://www.vmt-gmbh.com/

http://www.hgv.de/

http://www.vitronic.com/



Detection Technologies:

Machine Vision (2D)

Machine Vision (2 ½ D)

(Laser-) Triangulation (3D)

Photogrammetry (3D)

Stereometry (3D)

Laser Scanning (3D)

Time of Flight (3D)

Combining several of these technologies and different solution concepts offers lots ofoptions enabling a suitable solution for an application.

0,0% 20,0% 40,0% 60,0%

3D: Photogrammetry

3D: Triangulation (Sheet of light)

3D: Stereometry

3D: Scanning

3D with one camera

Time of Flight (ToF)

Visual Servo

12,2%

48,9%

26,7%

28,9%

24,4%

20,0%

14,4%

3D Detection Technology overall

The chart shows the distribution of the 3D technologies over all the companiesparticipating the survey. This includes line builders, machine builders and robotmanufacturers as well.

The full report provides more topics, statistical data, analysis, charts, and overviews, as well asarticles and links.



1.2 ALGORITHMS

Machine Vision algorithms are one of the keys to a successful implementation of robot guidanceprojects. On the one hand they assure, based on the information provided by the detection unit, thatpart position and dimension data is accurate and reliable, on the other hand algorithms canenormously influence the user comfort for system set-up. In the 90´s most algorithms, sometimesworking on binary images, have been based on blob (binary large object) detection, or edge detection.More sophisticated was grey scale correlation, however, this tool hasn´t been capable to detect a partin all possible rotational variations.

At the end of the 1990´s the first algorithms being able to work at complete 360° rotation, eithercorrelation or contour based, became available. At this time Cognex Corporation developed algorithmswith the ability to measure the scale of a repeatable object and in this way determine rough “z”-information (distance to the top of the part and by deduction the height of the part). In the followingyears these algorithms became faster, easier to use, control/parameterize and available from manymachine vision software suppliers.

The chart shows the distribution of the algorithms over all participants of the survey.Vendors were free to choose more than one option.

45,6%

38,9%42,2%

32,2%

13,3%

1,1%0,0%

20,0%

40,0%

60,0%

Contourbased

Correlationbased

Featureextraction

Cloud ofpoints

Companyown Tools

Others

Detection Algorithms




1.3 HANDLING

Highly standardized robots can be chosen based on application and product needs. Depending onmarket and application needs the different robot types do have more or less market shares.

Some of the robot manufacturers have focused early on the combination of robot and vision systems.One pioneer here definitely was the company Adept Technology. Among the leading 6- axis robotmanufacturers Fanuc Robotics is most focused on this topic.

In addition to the robots described below cartesian systems are used in many applications and insidemachines as well.

6- axes robots

In 1973 Kuka develops the first industrial robotwith six electromechanically driven axes.

Today six axes robots are available from manyvendors, for different applications, with differentaccuracies and payloads of up to 1000 kg(Kuka 1000 titan).

6- Axes robot, Kuka Robotics

SCARA robots

The first SCARA (Selective CompliantAssembly Robot Arm) robot was developed in1981 at the University of Yamanashi under theguidance of Hiroshi Makino.

SCARA robots are mainly used in assemblyprocesses of light and small objects.

SCARA Robot, Epson

Delta robots

The delta robot was invented in the early 1980sby Reymond Clavel at the Ecole PolytechniqueFederale de Lausanne (EPFL, Switzerland).

Delta robots are mainly used to handle light andsmall objects at very high speed.

Delta Robot, Adept Quattro



Cartesian coordinate robot

A linear robot uses three axes. They move in astraight line rather than rotate and are at rightangles to each other.

One of the advantages of this mechanicalarrangement is to simplify the robot control armsolution.

Cartesian robots sometimes are also calledGantry robots and they often are quite large.

Popular applications for cartesian robots arecomputer numerical control machines (CNCmachine), cutting machines and fast pick andplace systems.. Gantry handling, Güdel

The chart below shows the distribution of robot use among the participants of the survey.

39%12%

8%11%

13%36%

6%43%

11%29%

4%3%

23%13%

0% 10% 20% 30% 40% 50%

ABBAdept

ComauDensoEpsonFanucGüdelKuka

MitsubishiMotoman

NachiReis

StäubliKawasaki

Handling used / supported



1.4 ROBOT VISION TURNKEY- CONCEPTS

INTRODUCTION

Beyond the specific part to determine the position of an object or a feature of an object, varioustechnical processes provide many differences for the implementation of a Robot Vision system. Theseare both customer specific as well as process specific.

Site conditions as well as other requirements vary widely and make it sometimes impossible toimplement an application by using one unique concept.

As we know from many applications such as palletizing, glueing or robot guided assembly, differentcell layouts enable the realization of projects with different site conditions based on the samecomponents and subsystems.

For customers looking for solutions of several applications with different parts, the portfolio of availabletechnologies and concepts can be a key factor for decision making.

Many of the vendors already support stationary, robot mounted and hybrid concepts. Those conceptsand some variants are described later in this chapter.

BASIC CONCEPTS FOR SYSTEM SET-UP:

There are three basic concepts to set-up a robot vision cell:

Fixed mounted (stationary)

Robot / Handling mounted (moving)

Hybrid (combination of stationary and moving)

The following explanations and description are based on handling applications such as de-palletizing,pick and place or bin picking.




2.0 ROBOT VISION APPLICATIONS OVERVIEW

2.1 INTRODUCTION

Robot Vision as basic technology as well as an application, can be used across many markets,industrial and non- industrial. Sometimes it becomes difficult to define where to use the term industrial.However, our goal is to cover as much fields of application where Robot Vision technology is in use orcan be used in the near future as possible. The chapter "Future Trends" shows the expectation of thevendors for the applications addressed in this report.

For some case examples, especially when talking about new R&D projects and other prototypes onlylimited information can be distributed due to confidentiality. In these instances this report refers to thecompanies working on the project.

Many times turnkey suppliers, either system integrators or OEM´s (original equipment manufacturers),have focused on specific applications and technologies, for example on handling, assembly, glueingand many more. Technology, product and component suppliers also always have to look for othermarkets, applications and regions where their technologies and products can be used. Therefore toget the most comprehensive overview of applications for Robot Vision integrators, line builders andtheir suppliers have been interviewed. For verification of the results as well as to complete the picture,end users in several industries have also been interviewed directly.

The pre-selection of industries offered in the questionnaire was focused on markets where the use ofRobot Vision technology is already confirmed or at least planned for near term use. Multiple selectionshave been allowed.

Suppliers were asked to mark the industries where they are already active in. This was not limited toindustries where they are active in with Robot Vision. The option was given to the vendors to addapplications relevant to Robot Vision.

Chapter 2.2 shows an overview of the applications and the distribution by activities of the companies.Starting from chapter 2.3., applications are shown and described with reference to robot visiontechnology.

2.2 APPLICATIONS OVERVIEW

In order to get a complete overview of possible trends, applications and needs, the followingapplications have been included in the survey.

Assembly 2D

Assembly 3D

Assembly 3D Best Fit

Bin Picking

Baggage Handling

Palletizing / De- Palletizing

Racking / De- Racking

Outdoor Guidance

Pick & Place 2D

Pick & Place 3D

RV & Inspection 2D

RV & Inspection 3D

RV 2D & Identification


Seam Tracking

Service Robotics

UAV

UGV

Visual Servo



Chart below shows the applications covered by the survey and the distribution of activities of thecompanies included in the survey.

0% 10% 20% 30% 40% 50% 60%

Assembly 2D

Assembly 3D

Assembly 3D, Best Fit

Bin Picking

Baggage Handling

Palletizing / De- Palletizing

Racking / De-Racking

Outdoor Guidance

Pick & Place (2D)

Pick & Place (3D)

RV & Inspection(2D)

RV & Inspection(3D)

RV(2D) & Identification


Seam Tracking

Service Robotics

UAV

UGV

Visual Servo

40,0%

40,0%

22,2%

36,7%

8,9%

42,2%

25,6%

6,7%

54,4%

44,4%

34,4%

31,1%

28,9%

25,6%

18,9%

14,4%

5,6%

6,7%

22,2%

Applications where vendors are active in

Today the 2D Robot Vision applications still have a stronger representation in themarkets. However, the chapter "Future Trends" shows the expectation of the companiesand the forecasted changes.



2.3 ASSEMBLY- 2D & 3D & BEST FIT

Assembly is a huge field of application formachine vision and robotics. Great manyindustries and markets provide potentialfor these applications.Robot Vision for 2D Assembly is availablefrom almost all vendors, Vision vendorsas well as Robot manufacturers. 3Dassembly applications are much morechallenging. Therefore most of theseapplications have required experiencedmachine vision system integrators withstrong competence in vision and robotics.

Picture by Inos Automationssoftware, Best Fit for Panels, GlassModules, Cockpit and Front End Assembly.

Inos Automationssoftware is one of those innovative companies having introduced the HPFITtechnology (high precision fitting). The picture shows the assembly of glass modules, a typicalapplication in the automotive industry where the 3D Best Fit technology is required.Best Fit is the idea of assembling two imperfect parts and get an optimized result. For manyautomotive assembly applications perfect means equalized gaps around the fitted part. However, thedefinition of "perfect" is different from one manufacturer to the other.Especially the automotive industry is a huge field for 3D assembly applications. One will find roofassembly, door assembly, glass assembly, cockpit assembly, seat assembly and wheel assembly toname some of them.

Below please find information provided by the companies within the survey

All companies (100%) active in best fit assembly see machine vision as their corecompetence.

The strongest representation of companies with their core competence in robotics is found in2D assembly (66%).

94,4%

66,7%

52,8%

27,8%

44,4%

13,9%

97,2%

58,3% 58,3%

19,4%

47,2%

11,1%

100,0%

50,0%60,0%

30,0%

50,0%

15,0%

0,0%

20,0%

40,0%

60,0%

80,0%

100,0%

Machine VisionCompanies

Robotics System Integration Machine building Automation Line building

Assembly by Core CompetenceAssembly 2D Assembly 3D Assembly 3D, Best Fit

Future trends: 43% of all companies participating in the survey expect a high potential forgrowth for the 3D assembly application, 53% expect a high potential for 3D assembly Best Fit.

Future trends: companies that rate best fit assembly "high" are mainly active in Automotive,Foundry and Aerospace.



Companies in this survey active in 2D assembly:

ABB Robotics Adept Technology Advenovation AIC S.A. AMT Avalon BMH Cognex Comau Compar Edixia Epson Robotics

Fanuc Robotics Fatronik Grenzebach Hahn Automation HGV IBS Ilorya ImagingLab Kawasaki Robotics KeySysTech LMI MVTec

Motoman Robotics Octum Profactor RobotWorx SAC Scholz SuE Sick IVP SIR Visionic VisionTools ViTec

19%36%

3%8%

39%17%

58%11%

44%22%

19%14%

22%56%

0% 20% 40% 60% 80%

KawasakiStäubli

ReisNachi

MotomanMitsubishi

KukaGüdelFanucEpsonDensoComauAdept

ABB

Assembly 2D: handling

17%

28%

56%

33%

25%

47%

17%

78%

67%

92%

0% 20% 40% 60% 80% 100%

Visual Servo

Time of Light

3D Triangulation

3D Stereometry

3D Scanning

3D with one camera

3D Photogrammetry

3D Robot Vision

2 1/2D Robot Vision

2D Robot Vision

Assembly 2D: detection

20%

4%

37%

22%26%

14%

36%

19%

30%

18%14%14%

21%

14%

1%

27%

16%19%

11%7%

16%12%13%

23%19%

11%

2%

30%

8%

22%17%

20%

4%

11%6% 6%

9%10%

0%5%

10%15%20%25%30%35%40%

Assembly 2D: industries overview




3.0 ROBOT VISION MARKET OVERVIEW

3.1 INTRODUCTION

Robot vision is a basic technology as well as an application and can be used across many industries.Consequently there is a great number of markets and industries to be discovered and conquered bymachine vision. In order to give a broad overview on the industries and markets already discovered,the survey results depicted in the graph on the next page are drawn from European and NorthAmerican turnkey suppliers as well as product and component suppliers.

Turnkey suppliers, both system integrators or OEM´s (original equipment manufacturers), often arefocused on specific industries; product and component suppliers on the other hand always have tolook for other industries and regions where their existing products can be used. Therefore, to get amost comprehensive overview of the potentials for robot vision integrators, line builders and theirsuppliers have been interviewed. For verification of the results as well as to complete the overview,end users in several industries have also been interviewed directly.

The pre-selection of industries offered in the questionnaire was focused on industries where the use ofrobot vision technology is already confirmed or at least planned for near term use. Multiple selectionshave been allowed.

Suppliers were asked to mark the industries where they are already active in. This was not limited toindustries where they are active with robot vision. The option was given to the vendors to addindustries relevant to robot vision.

Chapter 3.2 shows an overview of the industries and the distribution by activities of the companies.Starting from chapter 3.3., industries and markets are shown and described with reference to robotvision technology.

3.2 INDUSTRIES OVERVIEW

In order to get a complete overview of existing and future trends, applications and needs, the followingindustries have been involved in the survey.

Aerospace Agriculture & Forestry Automotiveo BIW & Press shopo Assemblyo Logisticso Powertraino Paint shopo Tier 0.5 & Tier 1

Beverage Ceramics Consumer Goods

Electronics & Semiconductor Entertainment Food Foundry Glass Lab Automation Life Science Logistics & Transportation Medical Engineering Medical & Health Care Metal Processing Military & Defense

Mining Packaging Paper Pharmaceuticals Photovoltaic Plastics Recycling Service Robotics Security & Surveillance Textile Wind Energy Wood



The chart below shows all industries & markets covered by the survey and the distribution of activitiesof the companies included in the survey.

0,0% 20,0% 40,0% 60,0% 80,0%

AerospaceAgriculture & Forestry

AutomotiveAutomotive Press shop

Automotive Body in whiteAutomotive Powertrain

Automotive LogisticsAutomotive Assembly

Automotive Paint shopAutomotive T0,5 & T1

BeverageCeramics

Consumer GoodsElectronics & Semiconductor

EntertainmentFood

FoundryGlass

Lab AutomationLife Science

Logistic & TransportationMedical Engineering

Medical & Health CareMetal Processing

Metal & SteelMilitary & Defense

MiningPackaging

PaperPharmaceutical

PhotovoltaicPlastics

RecyclingService Robotics

Security & SurveillanceTextile

Wind EnergyWood

41,1%10,0%

80,0%34,4%

42,2%55,6%

33,3%57,8%

23,3%48,9%

20,0%21,1%

32,6%23,3%

12,2%40,0%

32,2%27,8%

12,2%7,8%

34,4%25,6%26,7%

42,2%27,8%

15,6%4,4%

48,9%13,3%

36,7%24,4%

31,1%7,8%

21,1%16,7%

10,0%14,4%15,6%

Industries where participants are active

The chart shows that 80% of all companies covered in the report are active in theAutomotive industry.

Outside the Automotive industry, Packaging (49%), Aerospace (41%), Food (40%) andMetal Processing (42 %) have the strongest representation.

Electronics & Semiconductor have been selected by only 23% of the companies. Thiscan be explained by the clear focus of the study on robot vision and a strongrepresentation of system integrators. Semiconductor and Electronics mainly usestandard machines with integrated vision functionality.



3.3 AEROSPACE

Compared to other industries such as Automotive,Packaging, Electronics, Photovoltaics orPharmaceuticals, the automation grade in aerospace ismuch lower. However, the number of companiesparticipating this survey being active in Aerospace isrelatively high. This probably is caused by the strongrepresentation of 3D robot vision in this survey.

One of the reasons for the lower grade of automation isthe much lower production rate of airplanes comparedfor example to cars. And the reason is the enormoussize of an airplane, making many automation tasksdifficult. However, the need for repeatable inspection,high accuracy in production and the goal to ease theheavy work are driving factors for automation.

Picture by Brötje Automation, vision guided rivetingmachine for aircraft and helicopter production.

The placement and inspection of rivets is one example where the automation already is standard since manyyears. However, so far specialized machines and not robots do the work.

The automated painting is another potential robot vision application considered in the aerospace industry. Thisapplication is already standard in the automotive industry, but also in wind energy for the automated painting ofrotors of wind mills (see the chapter wind energy). There are also many typical powertrain applications that areused for aerospace parts as well, especially when suppliers work for both industries, aerospace and automotive.

Additional opportunities can be found in related markets "outside" aerospace manufacturing, such as theguidance of the airplane for parking assistance, but also in related industries such as UAV, Military & Defense andService Robotics.

Below please find information provided by the companies within the survey

Compared with other industries andapplications aerospace shows strongactivities of the line builders (71%) andmachine builders (57%).

Machine vision core competence is onlyrepresented with 44%.

43,8%51,1%

45,8%57,1%

45,9%

71,4%

0,0%

10,0%

20,0%

30,0%

40,0%

50,0%

60,0%

70,0%

80,0%

MachineVision

Companies

Robotics SystemIntegration

Machinebuilding

Automation Linebuilding

Aerospace by Core Competence

Applications with strongest representation:

Pick & Place 2D Pick & Place 3D Palletizing / De- Palletizing

Handling: Kuka (59%), Fanuc (59%) and ABB (50%) have the strongest representation among thesystem integrators active in Aerospace.

Vision Systems: Sick (27%), Dalsa and Fanuc (both 22%) have the strongest representationamong the companies active in Aerospace.

Smart Cameras: Cognex Insight (32%) is the Smart Camera with the strongest representation atthe system integrators active in Aerospace.

MV Software: The vision software products most commonly used by companies active inAerospace are Cognex (22%) and National Instruments (19%).

MV Sensors: The vision sensors most used by system integrators active in Aerospace are bySick, Baumer and Automation Technology.



Companies in this survey which stated to be active directly in Aerospace:

ABB Robotics Advenovation AIC S.A. AMT Aqsense AV&R BiBa Bremen Blumenbecker Prag Comau Dalsa Fanuc Robotics Fatronik Fraunhofer

Grenzebach Automation HGV ibea ImagingLab Inos ISRA Kuka Robotics MVTec Motoman Robotics National Instruments Orus Integration PPT RobotWorx

Scholz SuE SIR Tattile Tec Automation ThyssenKrupp Krause Tordivel Vision++ Visionic Visio Nerf VisionTools VRSI

20%

24%

14%

21%

8%

22%

18%

3%

26%

24%

18%

18%

18%

18%

14%

11%

0% 10% 20% 30%

Assembly 2D

Assembly 3D


Bin Picking

Baggage Handling

Pelletizing / De- Palletizing


Outdoor Guidance

Pick & Place 2D

Pick & Place 3D

RV & Inspection 2D

RV & Inspection 3D


RV3D & Identification

Seam Tracking

Visual Servo

Aerospace: Applications

50%14%

18%9%

14%59%

9%59%

14%45%

9%5%

27%23%

0% 20% 40% 60% 80%

ABBAdept

ComauDensoEpsonFanucGüdelKuka

MitsubishiMotoman

NachiReis

StäubliKawasaki

Handling: by Integrators active in Aerospace

37,8%

40,5%

37,8%

62,2%

24,3%

0% 10% 20% 30% 40% 50% 60% 70%

3D with one Camera

Scanning

Stereometry

Triangulation

Photogrammetry

Aerospace: 3D Technologies

9%

23%

5% 5%

14%

5% 5%

14%9%

14%

5%9%

27%

5% 5% 5% 5% 5% 5% 5%

36%

5% 5% 5%0%

5%

18%

0%

5%

10%

15%

20%

25%

30%

35%

40%Vision Systems used by Integrators active in Aerospace




4.0 FUTURE TRENDS IN ROBOT VISION AND MACHINE VISION

4.1 INTRODUCTION

When I started in Machine Vision in 1996 at Cognex Corporation, Machine Vision was a great newworld of interesting visions and opportunities.

Trying to explain what Machine Vision is, I used to summarize that it is the optical, camera basedtechnology for automated inspection and guidance. My explanation was always to describe that nearlyeverything around us has come into contact with Machine Vision during its production. The tooth brushand the shampoo you are using in the morning, the cookies, eggs and all the packaging you arehaving on your breakfast table. The car you are using to go to work, even the ticket you are getting fordriving too fast is connected to Machine Vision in one or the other way.

Today, almost 15 years later, looking back to all the“Visions” we had in the 90´s, many of them becamereal – even standard. In addition many new “visions”have been born and grew into new dimensions.

Looking for a nice picture for “Future User Interface” forthis report I found many named as “Minority ReportInterface”- the one used by Mr. Tom Cruise. A movie –still a vision?

Outside the world of so called “Industrial MachineVision” drones are patrolling in the sky, UGV´s areused to explore the deep sea and repair oil lecks and amowing robot has become our favorite Christmaspresent.

Miniaturized components are used for surgeries insideour body. Advanced algorithms are used to classifycancer and identify human faces in Facebook.

So you see the explanation for “Machine Vision” stilldescribes great many different applications around us –however, the Machine Vision technologies alreadyhave conquered many fields outside the “typical”industrial fields.

Picture, from movie Minority Report

In the last 15 years prices of components went down, but labor cost either increased or stayed thesame. Therefore it was necessary also to work on the “ease of use” for the standardization ofcomponents, interfaces and MMI. Vendors, who were not typical Machine Vision companies enteredinto the Machine Vision fields. Those companies came up with user friendly, easy to use andinexpensive products for many applications, enabling automation companies and end- users to installand set- up these solutions by themselves.

Machine Vision is influenced increasingly by trends and technologies becoming available fromconsumer markets. This “evolution” is on-going. In the early times special hardware was needed and



image information needed to be reduced in order to achieve acceptable processing times. Todaypowerful hardware and sophisticated algorithms are available at relatively low cost.

Many changes that were expected did happen – others didn´t or we are still waiting for breakthroughs.

Looking from today back on the last 20 years but also ahead into the near future, it is obvious that theworld of Machine Vision offers enough space and opportunities for many different technologies,products and vendors.

This report is about several of these hot topics being discussed today especially when related torobotic automation. Interestingly two of those topics “Cloud Computing for Machine Vision “ and“Computational Cameras” have been the center of key level speeches at the EMVA conference inAmsterdam in May 2011.

During this conference Ramesh Raskar, Associate Professor at MIT Media Lab, was talking about the10th dimension and looking around a corner to explain computational cameras. Others are looking forthe computational power for processing large number of data for full 3D in “the Cloud” (the late Mr.Jobs had declared as our future for data storage and processing).

Reason enough to create this Report and present useful information for everybody active or interestedin Future Trends in Machine Vision, but has never been at MIT.



4.1.1 TECHNOLOGIES

Within the survey companies where asked about their expectation of several topics for the near futureof Machine Vision / Robot Vision. The topics that have been addressed were collected fromdiscussions with users and vendors of Machine Vision and automation equipment and solutions.

Technology Topics that have been addressed within the survey :Increase in camera dynamic range

Computational cameras

Real- time detection, moving objects

Use of color information

Analysis of point clouds

Increase in camera resolution

Cloud Computing for Machine Vision

Vision system integrated into the “robot controller”

Participating companies rated the topics in one of the four categories:

None Low Average High

Multiple selections have not been allowed

The analysis of point clouds as a topic strongly dedicated to 3D imaging gets one of thehighest rankings with 46% of the votes.

The Integration of vision systems into robot controllers is a topic for many years andalready utilized by several companies, but it remains a very important topic for thefuture (55%).

Real- time detection / tracking of moving parts is ranked high by 50% of theparticipants.

55%

19%

24%

46%

39%

43%

50%

31%

31%

0% 10% 20% 30% 40% 50% 60%

Vision system integrated into the "Robot Controller"

Cloud computing for Machine Vision

Increase in camera resolution

Analysis of point clouds

Use of color information

Further system / product price reduction

Real- time detection, moving objects

Computational cameras

Increase in camera dynamic range

"High" rating for Technologies



4.1.2 USER REQUIREMENTS

Within the survey companies where asked about their expectation of several topics related to userrequirements for the near future of Machine Vision / Robot Vision. The topics that have beenaddressed were collected from discussions with users and vendors of Machine Vision and automationequipment and solutions.

User Requirements that have been addressed :User Interface / MMI

Processing speed

Data mining SPC

System accuracy

System reliability and robustness

Miniaturization of components

Price / performance ratio

Responsiveness of supplier to application specific requests

Responsiveness of supplier to service and support requests




System reliability and robustness, the user interface, and the price / performance ratiogot the highest rating for “High Potential” overall.

Most of the topics addressed do have relatively high ratings. Only data mining and theminiaturization of components are rated relatively low.

66%

51%

10%

58%

74%

13%

62%

59%

42%

0% 10% 20% 30% 40% 50% 60% 70% 80%

User Interface

Processing speed

Data mining & SPC

System accuracy

System reliability / robustness

Miniaturization of components

Price / Performance ratio

Responsiveness of supplier to application specific requests

Responsiveness of supplier to service and support requests

"High" rating of User Requirements



4.1.3 APPLICATIONS

Within the survey companies where asked about their expectation of several topics related to userrequirements for the near future of Machine Vision / Robot Vision. The topics that have beenaddressed were collected from discussions with users and vendors of Machine Vision and automationequipment and solutions.

Applications that have been addressed :Assembly 2DAssembly 3DAssembly 3D Best FitBin PickingBaggage HandlingPalletizing / De- PalletizingRacking / De- RackingOutdoor GuidancePick & Place 2DPick & Place 3D

RV & Inspection 2DRV & Inspection 3DRV 2D & IdentificationRV 3D & IdentificationSeam TrackingService RoboticsUAVUGVVisual Servo




The combination of 3D Inspection and Robot Guidance got the highest rating for thefuture.

Overall the expectation shows a clear shift to applications that use 3D technology.

18%

46%

50%

46%

5%

33%

18%

32%

27%

50%

41%

53%

24%

46%

14%

26%

11%

25%

7%

0% 10% 20% 30% 40% 50% 60%

Assembly 2D

Assembly 3D


Bin Picking

Baggage Handling

Pelletizing / De- Palletizing


Outdoor Guidance

Pick & Place (2D)

Pick & Place (3D)

RV & Inspection(2D)

RV & Inspection(3D)



Seam Tracking

Service Robotics

UAV

UGV

Visual Servo

"High" rating for Applications



4.1.4 OUTLOOK

Machine Vision traditionally was focused on the industrial markets. Looking back to the 90´s marketleading vendors used this message in their naming such as “Cognex - Vision for Industry”. In the lastyears the Machine Vision business changed. It is developing more and more into other markets andapplications outside the traditional industrial business.

At the same time technologies and trends from other industries, especially consumer electronicmarkets influenced the Machine Vision business and helped to open new areas in the traditionalindustrial markets.

In the moment we see the world become 3- dimensional – a trend that is pushed from severaldirections: Machine Vision, industrial measurement and consumer markets such as gaming and videoentertainment as well. This is only one example showing how things come together, influencing thewhole business, changing markets and their vendors.

Therefore it is of high importance to monitor all these developments and changes in technologies andmarkets. Market Intelligence Reports help you to keep track, be prepared and stay on top of thesechanges and developments.




5.0 VENDORS OVERVIEWS

5.1 CONTENT OF THE SURVEY

Data in this report have been generated by a questionnaire sent to and returned by the companiesand by additional research. Research methods have been already listed in “Chapter 1, Introduction”.Companies not having responded to the survey and not providing public information related to thetopics are not included in these overviews.

The following chapters show the information provided by the participants or collected within researchand provide different analysis of the data. The Report summary gives you some selected charts. Thefull report provides more extensive overviews and information.

Topics that have been addressed in the survey :

Core competence of vendors

Business activities of the vendors

Regions where vendors are active

Industries where vendors are active

No. of Robot Vision systems installed, overall / per year

Detection Technologies, available or supported

Detection Algorithms

Applications where vendors are active

Application / Customer references

Robots & Handling systems used and supported

Vision systems used and supported

Smart Cameras used and supported

Machine vision software used and supported

Machine vision sensors used and supported

Future Trends- Applications

Future Trends- Technologies

Future Trends- User Requirements

Man Machine Interface (MMI)

Robot Vision system architecture

Multiple selections have been allowed for almost all questions



5.2 CORE COMPETENCE BY VENDORS

Information requested:

Which of the following fields of business does a vendor define as core competence?

Machine VisionRobotics & PositioningSystem Integration

Machine BuildingAutomationLine Building

Multiple selections have been allowed

89 % of the companies naming system integration as core competence also namedmachine vision.

81 % of the companies naming robotics & positioning as core competence also namedmachine vision.

88,9%

52,2% 53,3%

15,6%

41,1%

7,8%

0%10%20%30%40%50%60%70%80%90%

100%

MachineVision

Robotics &Positioning

SystemIntegration

MachineBuilding

Automation Line Building

Core competence of the Companies



5.3 BUSINESS REGION: DISTRIBUTION BY VENDORSInformation requested:

In which regions is each vendor active? Regions are defined as follows

Germany, Austria, SwitzerlandNorthern EuropeWestern EuropeCentral EuropeSouthern EuropeNorth AmericaSouth America

Middle East, Africa (MEA)ChinaJapanIndiaRussiaAustralia / Oceania

Multiple selections were allowed.

Due to the focus of the survey Europe, especially the German speaking region havethe strongest representation. The numbers also show how global many of thesecompanies act (or probably have to act to fulfill the requirements of the markets andcustomers).

0% 20% 40% 60% 80% 100%

Germany, Austria, Switzerland

Northern Europe

Western Europe

Central Europe

Southern Europe

North America

South America

Middle East / Africa

China

Japan

India

Russia

Australia / Oceania

77,9%

67,5%

63,4%

59,3%

57,8%

63,1%

37,8%

21,3%

36,6%

29,1%

32,1%

25,0%

22,1%

Regions where vendors are present and active



5.4 INDUSTRIES- DISTRIBUTION BY VENDORS

Information requested:

In which industries are the vendors active? Multiple selections were allowed.

Aerospace Agriculture & Forestry Automotiveo BIW & Press shopo Assemblyo Logisticso Powertraino Paint shopo Tier 0.5 & Tier 1

Beverage Ceramics Consumer Goods Electronics &

Semiconductor

Entertainment Food Foundry Glass Lab Automation Life Science Logistics & Transportation Medical Engineering Medical & Health Care Metal Processing Military & Defense Mining Packaging

Paper Pharmaceuticals Photovoltaic Plastics Recycling Service Robotics Security & Surveillance Textile Wind Energy Wood

The pre-selection of industries offered in the questionnaire was focused on industries where robotvision systems and technologies are already confirmed or at least planned for near term use.

Suppliers were asked to mark the industries where they are already active in. The option was given tothe vendors to add industries relevant to robot vision.

The Automotive industry has the strongest representation within the industries.

Within the Automotive industry most of the companies are active in Assembly,Powertrain and in Tier 1.

Outside the Automotive industry Metal Processing, Aerospace, Food and Packaginghave the strongest representation.



0,0% 20,0% 40,0% 60,0% 80,0%

AerospaceAgriculture & Forestry

AutomotiveAutomotive Press shop

Automotive Body in whiteAutomotive Powertrain

Automotive LogisticsAutomotive AssemblyAutomotive Paint shopAutomotive T0,5 & T1

BeverageCeramics

Consumer GoodsElectronics & Semiconductor

EntertainmentFood

FoundryGlass

Lab AutomationLife Science

Logistic & TransportationMedical Engineering

Medical & Health CareMetal Processing

Metal & SteelMilitary & Defense

MiningPackaging

PaperPharmaceutical

PhotovoltaicPlastics

RecyclingService Robotics

Security & SurveillanceTextile

Wind EnergyWood

41,1%10,0%

80,0%34,4%

42,2%55,6%

33,3%57,8%

23,3%48,9%

20,0%21,1%

32,6%23,3%

12,2%40,0%

32,2%27,8%

12,2%7,8%

34,4%25,6%26,7%

42,2%27,8%

15,6%4,4%

48,9%13,3%

36,7%24,4%

31,1%7,8%

21,1%16,7%

10,0%14,4%15,6%

Industries where vendors are active



5.5 NO. OF ROBOT VISION SYSTEMS INSTALLED - OVERALLInformation requested:

No. of Robot Vision systems installed by the vendors, overall? Only one selection was allowed.For easier classification the following ranges have been provided to be selected from.

1-5 systems 50- 100 systems 1000+ systems

5-10 systems 100- 500 systems

10-50 systems 500+ systems

Altogether 8 companies claim to have each more than 1000 systems installed.

17 companies have installed more than 100 systems for RV and positioning.


6

1

7

4

7

2

8

0

1

2

3

4

5

6

7

8

9

1-5 5-10 10-50 50-100 100-500 500+ 1000+

Systems installed for RV and positioning / Overall



COPYRIGHT

This market report is authored by AMC Hofmann.

No parts of this report may be reproduced in any manner without prior writtenconsent of AMC Hofmann. This prohibition includes copying by a firm for itsinternal use.

While every attempt has been made to provide accurate information, AMC Hofmanncannot be held accountable for any omissions or errors.

AMC Hofmann does not ensure the use of such data against liability of any kind.

For more information on this market report please contact AMC Hofmann.

© 2010 by AMC Hofmann

All rights reserved, including the right of translation by AMC Hofmann.

AMC HOFMANNPO BOX 115664629 HEPPENHEIM, GERMANYTELEPHONE: +49 (0) 1577 530 6969EMAIL: [email protected]: WWW.AMC-HOFMANN.COM

mailto:HOFMANN:@AMC-HOFMANN.COM

http://WWW.AMC-HOFMANN.COM/