RIC-Europe Meeting Recap

ROS-I CONSORTIUM EUROPE ANNUAL MEEETING ATTENDEES - 2015

Thanks to all the participants who attended our annual meeting on April 28! A number of member institution representatives gathered in Milan, Italy, to review last year's activities and discuss current and future initiatives. The ROS-I Consortium Europe is preparing to launch joint efforts in a number of areas ranging from simulation improvements to support for video compression, so consider joining us in order to achieve faster development results through our FTP mechanism. Learn more about RIC-EU's activities during the upcoming 2015 ROS-Industrial Conference which will take place June 9, 2015, on Fraunhofer's premises in Stuttgart, Germany.

ROS Driver for the IFM Efector O3D303

Submitted by: Tom Panzarella, Love Park Robotics, LLC

Love Park Robotics has taken part in an early beta test of the new IFM Efector O3D303 3D camera system. This sensor was officially released in Germany on April 13, 2015. The O3D303 is a time-of-flight sensor, specifically designed for use in industrial environments and automation applications. The 176x132 element detector features a relative accuracy of +/-4mm. In addition to the robust design, it is able to operate in illumination conditions ranging from complete darkness to sunlight. It is also affordable, at a per-unit cost of $1250 USD. A picture of the O3D303 is shown below along with a point cloud of an imaged pallet (taken in an office environment) to highlight the quality of the sensor data.

As part of our beta test period, Love Park Robotics developed a software interface to the O3D303 that allows us to utilize the sensor within software frameworks such as PCL, OpenCV, and ROS. This code has been made available as open-source on Github in the following repositories: libo3d3xx and o3d3xx-ros. Additionally, we are working with the ROS Industrial community to make binary debian packages available as part of the core ROS and ROS-I distributions.

As mentioned above, the software is split across two separate repositories. libo3d3xx is the core C++ interface to the hardware making the 3D data available as a PCL point cloud and the depth, confidence, and amplitude images available as OpenCV images. o3d3xx-ros is a ROS wrapper around libo3d3xx making it convenient to launch the camera as a node that will participate in a larger ROS computation graph making the data available on published topics and exposing services to configure and introspect the camera settings. The reason we split the code across two distinct projects is in recognition of the fact that not all users of the O3D303 will be operating within a ROS environment yet they will likely want to take advantage of the state-of-the-art computer vision algorithms available in PCL and OpenCV. We expect that this separation of concerns will also make the code easier to maintain and port to new platforms. The Github repositories have much more technical information available that you can use to get started.

To date, Love Park Robotics has been testing the ROS interface to the O3D303 as a primary navigation sensor for a medical mobility application and as an object recognition sensor for industrial automation applications. We plan to keep the code current to hardware changes as the O3D3xx series of sensors evolve. To that end, we have been keeping in close communication with the sensor vendor, IFM Efector. The code will remain open-source and our process will remain transparent -- for now, we are using the Github issue tracker to document our roadmap for the software. We invite the ROS-I community to join us in making sure this software represents the most robust interface to this new and exciting sensor for industrial applications. We look forward to seeing how you put it to work.

Planning and Control of High-Tech Motion Stage using ROS-Industrial

Alten Mechatronics, in cooperation with Bosch Rexroth and FEI, created a motion planner application for a 5-DOF motion stage in a Transmission Electron Microscope (TEM).

The application is a little outside the scope of other application of ROS-Industrial (i.e. robots), but the problems in these high-tech system can benefit hugely from advanced in robotics. In this case motion planning libraries from MoveIt! were implemented to generate collision free paths for the 5-DOF motion stage moving in a cluttered environment (refer to our 20 October 2014 blog post).

Building on earlier simulations, the application was extend to control the physical hardware. For this a communication was set up between ROS and the motion controller of the motion stage: The Bosch Rexroth NYCe 4000 motion controller. A driver for this platform was created in accordance to the simple message protocol, so that no development on the ROS side was needed: The NYCe controller acts as any other robot controller already supported by ROS-Industrial.

The result is application with ROS tooling (MoveIt!, RVIZ, etc.) and a high-tech motion platform able to plan and execute complex motions and increasing the speed of path execution with a maximum factor of 5 compared to current implementation.

More possibilities are open, like optimizing paths or planning constrained paths using the Descartes planner.

RIC-Europe Annual Meeting - 2015

We are happy to announce the detailed agenda for the ROS-Industrial Consortium Europe Annual Meeting 2015, which will take place right before RoboBusiness Europe on April 28th in Milan, Italy. Please note that the consortium dinner (included with your registration) will follow in the evening, so that you will have the chance to further network with peer members before attending RoboBusiness the day after.

Time : 2:00 PM - 6:30 PM CET on Tuesday April 28th, 2015

Place: Atahotel Expo Fiera, Via Keplero 12 20016 Pero Milan, Italy

Agenda:

• 1400 Welcome and EU/Americas Consortia Updates, EU Projects overview
• 1445 Members' activities and introduction of new members
• 1545 Break
• 1600 Invited presentations
  • IT+Robotics Srl - cROS: how ROS meets industrial needs
  • Magazino GmbH - a microWarehouse running on ROS
• 1700 Strategic and Focused Technical Projects & other initiatives
  • TU Delft - CAD to ROS conversion
  • TU Delft - Generic IO
  • Robert Bosch GmbH - Real-time, bare-metal implementations of ROS
• 1930 Dinner

Registration: registration link

How many people can we send?

• Full Members - 3
• Associate Members - 2
• Research Members - 1
• Additional Members - 150 EUR each

Not a member? Join Now

We look forward to seeing you in Milan!

ROS-Industrial Training 2015 (April 29)

Southwest Research Institute is coordinating with Caterpillar to host a ROS-Industrial Training Class April 29 – May 2, 2015 in Peoria, Illinois. Wednesday at 1 pm, we begin with a tour of a Caterpillar manufacturing plant, and end with a dinner in downtown Peoria. The detailed agenda can be found at the ROS-I website.

The classroom portion is similar to the training held last year with the same curriculum and an added session and lab option for Descartes. Please bring a laptop to the class with the ROS-I training Virtual Machine pre-installed. This class is geared toward individuals with a programming background who seek to learn to compose their own ROS nodes. Day 1 will be split into two tracks: the ROS Basics track will focus on introductory ROS skills; the Advanced track will focus on camera/perception topics. Day 2 will examine motion planning using MoveIt! as well as the Descartes planner and Perception concepts. Day 3 offers a lab programming exercise (with a choice of):
• Simple Perception Lab
• Simple MoveIt! Application
• Simple Descartes Application.
Registration is now open: http://rosindustrial.org/training-registration.

RIC-Americas Meeting Recap

ROS-I Consortium Americas Annual Meeting Attendees - 2015

Thanks to all who attended the annual meeting on March 27th ! Our group included representatives from ABB Robotics, Boeing, CAT, Clear Path Robotics, Deere and Co., Dematic, Ford, Fraunhofer IPA, GA Tech, HDT Robotics, Innovation Machine, Intelligrated, JR Automation, Northwestern University, OmniCo AGV, Open Source Robotics Foundation, Siemens, SwRI, Textron Systems, Vetex, UNC, UT Austin NRG, Wolf Robotics, Yaskawa Motoman Robotics.

We are grateful to our presenters are who recognized in this linked agenda document. For an overview of the accomplishments from the past year, please browse the RIC-Americas 2014 Year in Review Prezi (below). Note that Consortium members have access to the full proceedings of the event via the member portal.

Click to launch the presentation

ROS-I 3-yrs. Montage Video

Thanks to those in the ROS-I community who contributed to the ROS-I 3 yrs. Montage video! We would like to ackowledge:

• Calibration of camera to robot: SwRI
• Denso VS060 path planning using ROS-Industrial Cartesian Planner: TORK
• Cartesian Planner plug-in for MoveIt!: BioRobotics Institute at Scuola Superiore Sant'Anna/MicroBio Robotics Institute at the Italian Institute of Technology/SwRI/GSoC
• Process Simulate to ROS bridge: Siemens
• Path planner optimization and planning request adapter plug-in for MoveIt!: IDEXX/RIC-Americas
• Block pick and place: Technolution
• Palletizing unknown products: Alten Mechatronics
• Plastic crate depalletizing with lightweight robot: Intermodalics
• Pick and place with obstacle avoidance: Deere and Co.
• Factory-in-a-day, EU FP7 Factory of the Future 2013 Programme (FP7-2013-NMP-ICT-FoF)
• Robotic 3D scanning: Institute Maupertuis
• ROS-I training class pick and place exercise: RIC-Americas
• ROS-Industrial Consortium Robotic Routing FTP, Testing at CNRC: RIC-Americas
• ROS-Industrial Consortium Robotic Blending FTP Milestone 2 Update: RIC-Americas
• 8-DOF microscope positioning for TEM: Alten Mechatronics
• Multiscale teleoperation: UT Austin Nuclear and Applied Robotics Group
• Mobile robotic 3D scanning: UT Austin Nuclear and Applied Robotics Group
• Rob@Work3 logistics: Fraunhofer IPA
• Euler automated warehousing: SwRI
• PRACE dual-arm robot: Fraunhofer IPA
• YouBots pick and place multiple arm cooperation: NIST
• Dual arm robot coordinated motion: Fraunhofer IPA/Yaskawa Motoman Robotics
• BMDA3 dual arm robot: Fraunhofer IPA/Yaskawa Motoman Robotics/SwRI
• Rangar TT: Blue Force Robotics

ROS-Industrial Consortium Overview 2015

RIC-Americas Annual Meeting - 2015

We are announcing the detailed agenda for the ROS-Industrial Consortium Americas annual meeting, which will take place in conjunction with Automate 2015. Note that there is also a pre-meeting dinner on March 26 that is included with your registration.

Time: 8:30 AM - 4:00 PM CST (Chicago time) on Friday, March 27

Place: Palmer House Hilton Hotel, Chicago Room on 5th floor. Only two miles from the McCormick Place convention center in Chicago, IL.

Agenda:

• 0830 Registration/check-in and light breakfast
• 0900 Welcome and Consortium updates
• 0945 ROS-I Roadmap Presentation
• 1000 Break
• 1015 Invited ROS-I Presentations
  • Cessna/CNRC - Robotic Routing
  • Caterpillar - Agile Automation
  • Wolf Robotics - Technology Transfer to Practical Applications
  • UNC - Real-Time Motion Planning & Proximity Computations for Industrial Robots
  • GA Tech - Canonical Robot Command Language
  • Siemens - ROS Simulation for Industrial Applications with Tecnomatix
• 1215 Lunch/Keynote: Fraunhofer IPA - Factory in a Day Project - Reducing System Integration Time to One Day
• 1300 Focused Technical Project Presentations - 2015
  • Robotic Blending Milestone 3: Expedite and Technology Transfer
  • Strategic Project: CAD to ROS Conversion
  • Best of Amazon Picking Challenge
  • On-Line Adaptive Cartesian Path Planner for Collaborative Robotics
• 1400 Breakout discussions about FTPs
• 1500 Lightning talks by Research Members
• 1600 Conclusion

Registration: Registration Link

How many people can we send?

• Full Members - 3
• Associate Members - 2
• Research Members - 1
• Additional Members - $300 EA

Not a member? Join Now

We look forward to seeing you there!

January Community Meeting Highlights

From the meeting host, Ulrich Reiser:

Thank you for participating in today's ROS-Industrial Community Meeting. We had more than 100 people in attendance representing both industry and academia! For those who couldn't make the live event, it is now available on YouTube (below). We are grateful to our presenters:

• Welcome and update on RIC-Americas and EU | Paul Hvass (SwRI), Ulrich Reiser (Fraunhofer IPA)
• Modelling of Realtime Communication Patterns | Arne Hamann (Bosch)
• Descartes Path Planner with Application to Robotic Routing | Shaun Edwards (SwRI)
• Amazon Picking Challenge | Dejan Pangercic (Bosch)
• Why DDS for ROS 2.0? | Dirk Thomas (OSRF)
• MoveIt! New Functionality and Status Update | Sachin Chitta (SRI)

Join us for our next Community Meeting, likely in May around ICRA!

Motoman SDA10F & ROS-Industrial

Cross-posted from the Smart Robotic Systems blog

Author: Frantisek Durovsky

At the Department of Robotics we’ve spent several weeks testing the new ROS-Industrial driver for SDA10F since it’s announcement on Dec 10th 2014. As mentioned in the original post, the driver was developed by Fraunhofer IPA in cooperation with Yaskawa Smart Robotics Center in Japan, Yaskawa Motoman Robotics and is designated to control dual arm Motoman robots. Even though only the hydro version of driver has officially been released so far, we have also scucessfully managed to test the current indigo branch in combination with Ubuntu 14.04 LTS.

Motoman SDA10F Support and Moveit Config packages follow standard ROS-Industrial naming convention so all config and xacro files are located as usual. Roslaunching “testsda10f.launch“ from “motomansda10f_support“ folder provides simple interface to check basic robot’s model behaviour and orientation of particular axes.

To read the full blog post please browse here: http://www.smartroboticsys.eu/?p=675&lang=en

Application Deadline Extended to January 16: Yaskawa Motoman Offers Robots for Amazon Picking Challenge Teams

Yaskawa Motoman has extended their application deadline for teams who are interested in using consignment Motoman robot hardware to compete in the Amazon Picking Challenge. To apply for a robot, please submit this application form by midnight PST January 16, 2015.

For more information, please refer to the Yaskawa blog post for more details about the offer.

2nd Factory-in-a-Day Newsletter!

We are happy to report that the second Factory-in-a-Day newsletter is posted. Topics covered in this newsletter include:

• Letter from the project Coordinator, Dr. Martijn Wisse, TU Delft
• Workshop on Philips use case
• Spotlight on: Philips
• Do robots kill jobs?

The Factory-in-a-Day project leverages the ROS-I repository, and is supporting improvements and maintenance of the Universal Robots package. As new general factory automation capabilities are created through this project, we anticipate that they will be added to the ROS-I repository.

Demonstration of the Fraunhofer IPA ROS-I Driver for Yaskawa Motoman Dual Arm Robots

Submitted by: Thiago de Freitas and Ulrich Reiser, Fraunhofer IPA

Under the cooperation between Fraunhofer IPA, Yaskawa Smart Robotics Center in Japan and Yaskawa Motoman Robotics, a ROS-I driver to support multi-groups control for the Motoman Robots was developed.

The first industrial dual-arm manipulator to run the driver was the Motoman SDA10F with an FS100 controller. The driver provided the capability for generating synchronous and asynchronous movements from the ROS side that could be send to the FS100 controller and then executed by the real robot groups (left arm, right arm and torso). Also, support packages were setup including driver configuration files, URDF and MoveIT! configuration files.

The driver was also demonstrated during ROSCON 2014, using a Motoman BMDA3 robot. [Remarkably, the driver worked with the hardware despite the SwRI software team never having access to the hardware prior to the event.] The demo was organized by Yaskawa Motoman and SwRI.

Erik Nieves (Yaskawa Motoman USA) Grooves with the BMDA3 at ROSCon 2014 in Chicago

Paul Hvass (SWRI/ ROS-Industrial Consortium PM) "running" with the BMDA3

Some tutorials are recommended for getting a better overview of the driver usage and system configuration:

• Installation procedure

• Creating a dual-arm system

• Dual-arm system example

Some additional information:

• The initial release is a source-only release for Hydro. The Indigo Binary Release is going to follow in Q1 2015.

• Please direct your public questions to the ROS-I developers list (swri-ros-pkg-dev@googlegroups.com)

Demonstrating the Integration of ROS with Siemens’ Process Simulate

Siemens PLM Software is a leading global provider of product lifecycle management (PLM) software. Those PLM solutions, can help make smarter decisions that lead to better products.

As always, we at Siemens PLM Software are looking for new areas that will allow us to understand the future of robotics in the industrial sector. After we came across The Robot Operating System (ROS) and ROS-Industrial, I was sent to take a closer look. In June, I participated in the “ROS Industrial Basic Developers Training Class” held at Southwest Research Institute (SwRI) to understand more about the ROS ecosystem and tools. Since then, I have been experimenting with ROS libraries and tools and thinking about a connection between ROS Industrial and our own Process Simulate software.

From what I learned, ROS Industrial has interesting potential in the area of industrial robotics by providing the following:

• Standardization for robotic languages.
• Real time path planning and collision avoidance.
• A huge library of open source components.

For the above reasons, we believe that ROS-I has the potential to play a significant role in the industry and we should be a part of it.

After a few experiments, I compiled the demo (refer to video above) which shows how Process Simulate can provide a full simulation environment for a ROS controlled robot (R2-D2 believe it or not). In the demo, you’ll see that R2-D2 has three proximity sensors which are mounted on the right, front and left (their signal values can be seen in real time on the top left of the screen). R2-D2’s objective is to leave the maze using the simple algorithm of “always try to turn to the right”. But to make its life a little more interesting, we added red barriers which are removed manually at the start of the simulation to create a more dynamic environment.

As you can see from the rqt_graph above, the sensory information received from Process Simulate is processed on the ROS side and information about the robot’s new location is sent back to Process Simulate (the objects in red are Process Simulate components).

Along the way, I overcame a couple of interesting challenges such as:

• Having to write a robotic program using third party tools.
• Connecting between Windows-based Process Simulate and Linux-based ROS.

After discovering some real added value of linking ROS Industrial with Process Simulate, I’m going to explore further capabilities, like Vision, and working with more complicated environments using advanced Process Simulate and ROS Industrial software packages (PLC and welding in Process Simulate and OpenCV and MoveIt in ROS-I).

If you are looking to share interesting view points, use cases and environment challenges which are related to ROS-I, contact me at: moshe.schwimmer (at sign) siemens.com

Additional references:

http://www.plm.automation.siemens.com/en_us/products/tecnomatix

https://www.facebook.com/Tecnomatix.NXforManufacturing

Yaskawa Motoman Offers Robots to Amazon Picking Challenge Teams

Yaskawa Motoman Robotics is pleased to announce sponsorship of the Amazon Picking Challenge to be held in conjunction with ICRA 2015. This open competition will further the development of robot skills required for e-commerce and other material handling. Yaskawa Motoman is offering consignment robots to select teams entering the challenge. Selected teams will receive their choice of robot model in January, 2015 and may keep them through June, subject to model availability. To apply for a Motoman consignment robot, please submit the following items via this form by midnight PST December 17, 2014 January 16, 2015 (extended deadline).

• Video of a simulated robot executing a picking task
• Link to your team/organization website
• Completed application describing your Motoman hardware utilization plan

Yaskawa Motoman will provide robots, software (including our MotoROS driver), and onsite technical support both at the team’s location and the event.

To support your development efforts, the ROS-Industrial Consortium will be updating its pick and place tutorial to include the Motoman MH5 II model. The tutorial will also be updated to ROS Indigo. Additional resource links:

• ROS-Industrial open source software for various Motoman robots
• ROS-I Tutorials for Motoman Robots
• All Motoman Robots

Software Engineer/PhD opening at Fraunhofer IPA

From Florian Weißhardt via ros-users@

Position: Software Engineer, possibility to obtain PhD degree

Location: Fraunhofer IPA, Stuttgart, Germany

Experience: Strong skills in software design and C/C++ development and rich experience in ROS development

Finding solutions to organizational and technological challenges, particularly within the production environment of industrial enterprises. That, in a nutshell, is the key focus of the research and development work carried out at the Fraunhofer Institute for Manufacturing Engineering and Automation IPA. With 14 individual departments engaged in the fields of Corporate Organization, Automation and Surface Engineering, our R&D projects aim to enhance production processes and make products more cost-effective and environmentally friendly by identifying and exploiting the potential for automation and streamlining at clients' companies.

The Fraunhofer IPA department for robot and assistive systems develops service robots for various application fields (e.g. domestic, inspection, logistics, production assistance, manufacturing, etc.) with the goal of reliable, robust and safe service of these robots in everyday environments. These applications require complex software systems including navigation, planning, perception and manipulation for dynamic and changing environments and intuitive human-robot interaction.

The position covers the development of concepts and tools to reduce integration effort and simplify application development for these complex robotic systems in the frame of public funded national and EU projects. The transfer of the results to industry by organizing workshops, publishing articles relevant magazines and exhibiting demonstrators at fairs and conferences is part of the job description as well.

You have completed your master or diploma degree with excellent results and are interested in interdisciplinary research with high-tech robots like Care-O-bot or KUKA iiwa. You could already gather experience in scientific working and optimally already presented your first results at an international conference. You are confident in software architectures and software engineering and have practical experiences with the development of robot applications in ROS.

We offer you a highly interesting and diverse work environment with both contact to top robotics researchers and industry. In addition to obtaining a phd degree, the position encompasses early transfer of project and staff responsibility. For the implementation and validation of your ideas, we offer exceptionally equipped laboritories and test environments.

Qualifications/Requirements:

• Rich experience in ROS development
• A Master/Diploma degree from a top university in computer science, robotics or software engineering
• Strong skills software design and C/C++ development
• Proficient oral and written English skills

Advantageous are:

• Experience with model-driven engineering approaches
• Oral and written German skills

Please include the following documents in your application:

• Cover letter that expresses your motivation and (research) goals
• CV
• Transcripts of all obtained degrees (including scholar education)
• References and certificates relevant to the position

Please send your application to martina.goetzner@ipa.fraunhofer.de referring to position IPA-2014-109.

Alten Mechatronics Applies Robotic Technology in FEI Transmission Electron Microscopes (TEM)

Submitted by Simon Jansen, Alten Mechatronics

FEI designs, produces and supports a wide variety of high performance microscope systems, which can visualize details up to the picometer scale.

In their small dual beam (SDB) systems, a moving stage platform is present. This stage platform needs to be positioned in eight degrees of freedom (DOF).  Next to the stage, the microscope contains a lot more components such as an electron and ion column, multiple detectors and a gas injection system.  These SDB systems are not only used as pure microscopes, but also as nano workshop systems. It is possible to use these systems to add or remove material of a sample, while the sample is being inspected.

Because these microscopy system have such a wide variety of applications, the microscopy chamber gets rather full with components and parts. Therefore positioning and moving the eight DOF stage becomes more and more challenging.

In current systems, an in-house developed solution is used to plan the motion trajectories for the stage. The most important requirement is that the stage moves collision free between two configurations. Using the current solution, each movement between stage configurations is programmed by hand.  In some cases, this planning problem is just too complex and the current in-house developed solution is not able to find a solution. In other cases, the found solution takes too much time because axes can only be moved sequentially. This consumes a lot of time when moving eight axes and is therefore not feasible.

FEI is investigating alternative solutions to perform motion planning for their stage. For these alternatives it is important that the stage moves collision free within a certain time. This time should be minimized to obtain the highest possible throughput. A benefit of using motion planning is the possibility to move axes in parallel instead of sequentially. 

Alten Mechatronics performed a proof of principle study in which a simulation model of the microscope was developed. By using the motion planner MoveIt! collision free paths were generated for different stage configurations. Three cases were selected and compared to the current solution. One case was meant to benchmark the new motion planner, the two other cases to show the capabilities of the motion planner in extreme situations.

Alten showed that by using MoveIt! it became possible to calculate stage trajectories up to five times faster than trajectories found by using the in-house developed solution. For the other cases, it became possible to find trajectories that were not possible when using the in-house developed solution.

By using MoveIt! it is possible to realize complex stage movements which are guaranteed to be collision free and are resulting in a much higher throughput.

For the next phase, the results of the first ROS-Industrial Focused Technical Project (http://rosindustrial.org/ftp-status/) will be used to improve the performance of planner. In the first phase, MoveIt sometimes generated sub-optimal paths, which had to be rejected. With the optimization of the FTP, we will be able to guarantee near optimal solutions and be able to predict a lower reliable cycle time.

Mark Geelen & Simon Jansen Contact: rosindustrial(at-sign)alten.nl

3rd ROS-Industrial Training at Fraunhofer IPA: ROS for Industrial Applications

ROS-Industrial Technology Seminar: "ROS for Industrial Applications"

Friday, October 24, 2014

More information: Stuttgarter Produktionsakademie

Description

The open source “Robot Operating System” ROS offers highly developed robotics software components which can be used in flexible industrial applications. In this praxis-oriented tutorial users will get in touch with the basic functionalities of the ROS framework and the ROS-Industrial initiative. Participants will get an impression about the power of the system and learn how to use it in their own application.

Especially in dynamic environments with a variety of different work pieces there is a demand for highly flexible automation solutions supported by sensors and intelligent software components. A cost efficient, reusable and powerful solution is the open source framework ROS. It offers a huge amount of intelligent algorithms, methods and integrated libraries. An advantage is that software as well as hardware components can easily be exchanged due to a network based communication layer and standardized interfaces. This allows time saving and cost effective software development, which lowers the overall development costs.

In robotics research ROS is a well-established standard. The next step is to bring this power to industrial applications. For this purpose the ROS-Industrial initiative was founded. This tutorial will get the participants in touch with the theoretical basics of ROS and teach how to practically use it for their own industrial application.

Workshop Topics:

* ROS – Introduction and Basics
* 3D-Perception using ROS
* Localization and Navigation using ROS
* Motion Planning with MoveIt!
* Application Development using ROS
* Introduction to ROS-Industrial Initiative

In small groups the attendees will have the chance to gain hands-on experience within those topics under the guidance of experts of the respective field. The seminar is suitable for attendees both with or without experience in using ROS and will be held in bilingual language (German, English).

Also please feel free to forward this E-Mail to your colleagues, project collaborators or those who may be interested in this seminar.

If you have questions please don’t hesitate to contact us!

Hope to see some of you in October (or for the 4th seminar on Thursday, March 5th, 2015)!

Contact Person:

Dipl.-Ing. Florian Weißhardt Project Manager ROS Industrial Fraunhofer-Institute For Manufacturing Engineering and Automation IPA, Stuttgart Florian.Weisshardt@ipa.fraunhofer.de

Industrial Calibration Library Update and Presentation

By Dr. Chris Lewis, SwRI:

Robotics and automation systems are increasingly reliant on both 2D and 3D imaging systems  to provide both perception and pose estimation. Calibration of these camera/robot systems is necessary, time consuming, and often a poorly executed process for registering image data to the physical world. SwRI is continuing to develop the industrial calibration library to provide tools for state-of-the-art calibration with the goal to provide reliably accurate results for non-expert users. Using the library, system designers may script a series of observations that ensure sufficient diversity of data to guarantee system accuracy. Often interfaces to motion devices such as robots may be included to fully automate the calibration procedure.

As a vision systems developer one may ask the following questions with regards to both intrinsic and extrinsic camera calibration.

1. How many images of the calibration target are needed?
2. At what ranges?
3. At what angles?
4. How many near the center of the field of view vs at the edges?
5. What accuracy is achievable?
6. What accuracy was achieved?

With our framework, a user may rapidly explore these questions.

Our framework is built using Google's Ceres Solver which is a state of the art non-linear optimization tool specifically designed to solve Bundle Adjustment problems efficiently. Our framework consists of five main parts.

1. The main script processing code which
   • Collects observations
   • Runs the optimization
   • Installs the results
2. A library of Ceres compatible cost functions.
3. The camera observer interface which ties your cameras to the system and automatically triggers the camera and locates common calibration targets within specified regions of interest.
4. The scene trigger interface which provides interfaces to motion hardware such as robots. It may also serve to communicate with users to specify how to configure each scene.
5. Transform interfaces which provide the means by which kinematic values may be fed into and out of the calibration system. Updates to these extrinsic kinematic parameters is immediate and persistent.

Using this framework, we have demonstrated three distinctly different calibrations:

• Extrinsic calibration of a camera mounted on the tool of a robot
• Extrinsic calibration of a network of cameras
• Extrinsic calibration of a static camera to a robot

In addition, the ROS-I team is currently developing an intrinsic calibration script whereby a robot moves the calibration target to create a repeatable set of calibration images. In the near future, we will be developing kinematic calibration procedures for robots using cameras to better estimate robotic joint parameters.

Additional links:

• Wiki documentation: industrial_extrinsic_cal
• Tutorials
• One-shot calibration video
• Presentation slides