From Filipe Pereira via ros-users@

Follow Inspiration is a Portuguese Technology Company.

We develop outstanding software and hardware solutions in the fields of Interaction, Image Recognition, Robotics and User Experience.


We're recruiting a Robotics Software Developer with minimum experience of two years, for our facilities in Porto.



·         Academic Background on Computer Science and Engineering or equivalent;

·         Experience in the implementation of projects of distributed systems and product development;

·         Proven experience on computational intelligence and / or artificial vision;


Technical Skills:

·         Solid knowledge of applied robotic;

·         Advanced knowledge on programing: C / C++;

·         Experience on the development of real-time applications (soft real-time);

·         Advanced skills on artificial vision (OpenCV);

·         Experience with projects in ROS;


Send us your CV to

From David Vandergucht

Position: Software Engineer, specialized on Inertial Measurement Unit on embedded systems.
At LOEMI laboratory, Institut national de l'information géographique et forestière (IGN, France)

In the context of a large-scale European nanotechnology project, IGN is seeking a software engineer specialized in Inertial Measurement Unit's data processing to contribute to the design of a real-time geolocalization system based on image processing (smart glasses).


As part of a team of researchers and engineers in IGN's MATIS and LOEMI laboratories, you will participate to the building of a 3D pose estimation tool relying mainly on video streams processing and inertial sensors.

Your work will be focused on the inertial part. This tool will then be ported on a wearable embedded system developed by our partner CEA, but may also be used on UAV for 3D scanning tasks involving photogrammetry.

The mission contains four folds:

  • You will develop a high rate 3D trajectory reconstruction software based on gyroscopes and accelerometers.

  • The sensors planned to be used being targeted for consumer applications, their precision worth their (low) price. Hence, you will develop several filtering and calibration tools to cope with the quality of these sensors and limit the drift that tends to accumulate over time.

  • You will develop a data fusion system to combine the trajectory computed from the inertial data with the pose estimation retrieved by an image processing. The knowledge of the uncertainties of both sensors and processing methods should allow the fusion, but will also have to be recomputed and propagated over time.

  • You will develop an IMU calibration system based on information provided by image analysis.

The whole software being ported on a custom-designed System-on-Chip (SoC), a hardware acceleration of some of the functions may be possible via an FPGA chip.

Required skills:

  • Classical mechanics

  • Signal processing

  • 3D pose estimation

  • Kalman filter, particle filter, complementary filter

  • Very good knowledge of C++

  • Ability to read/write/talk fluently in English.


  • Agile methods

  • Knowledge of the ROS framework (Robotic Operating System).


Work location:

The research department of the French mapping agency (IGN) hosts four laboratories. Two of them, the MATIS and the LOEMI are specialized in computer vision, photogrammetry, image analysis, 3D modeling and remote sensing for the MATIS, and optical and electronic sensors for observations and measurement for the LOEMI. Those two laboratories are composed of more than 50 researchers and engineers whose publications are recognized in their fields.

IGN is a public institute located in Saint Mandé, France (immediate border with Paris), accessible in metro or RER. The location allows quick access to the Bois de Vincennes, hosts a corporate restaurant and propose several services included a large sports offer. Professional training and courses during the contract are available through our engineering school l'École Nationale Des Sciences Géographiques.

Project Context

This position takes place within the scope of a large-scale European Project: Things2Do (call KET-ENIAC 2013-2, 120M€). The project involves 45 academic and industrial partners from the nanotechnology field. IGN's laboratories contribute to the development of a wearable lightweight pedestrian navigation aid, in collaboration with Commissariat à l'Energie Atomique (CEA).

During four years, a fifteen people team of researchers and engineers will study and design a smart camera based on FDSOI architecture and computer vision techniques embedded on this architecture. A key goal is to allow a very precise pose estimate based on image and IMU rather than GPS signals.

Required qualification

High level education in software or electrical engineering (master of science or PhD)

Type/Contract duration


CDD for 24 months.

We are hiring, ASAP.

IGN, Saint-Mandé (Paris, métro Line 1, RER Vincennes).

How to apply?

Before the 15/12/2015, please send an email to the contact address, with one pdf file containing:

  • a detailed resume with description of achieved projects.

  • A cover letter.

  • 3 contact names for recommendation.


David Vandergucht, project manager for IGN (

4th ROS Meetup in Korea!

| No Comments | No TrackBacks
From Yoonseok Pyo

The 4th ROS Korea users seminar & meetup event was held on 7-8 November 2015 at Seoul, Korea. Particularly, in the second day, we prepared the tutorial seminar for participants with hands-on experience of ROS. This tutorial seminar arranged 4 section, ROS Basic, ROS Tools like RViz and rqt, ROS Navigation using Turtlebot and ROS MoveIt using ROBOTIS Manipulator, by Jihoon Lee and Yoonseok Pyo. The goal of the seminar & meetup event is to provide the hands-on experience and programming skills when using ROS and ecosystem.


See the links below for the photos of this meetup.

See the links below for the presentation slides. It is written in Korean, but we hope to bring up new ROS Korean contributors in not far future.

OROCA (, about 30,000 members) and Korea Open Society for Robotics (, about 4,300 members) communities have been effort to expand open robotics in Korea. We want to continue this activity with ROS. We are tentatively planning to hold ROS summer school and ROS meetup event in Korea on next year. 

We'd like to give special thanks to OSRF, ROS development teams and many ROS users. Also, thanks to our sponsors Korea Institute for Robot Industry Advancement (KIRIA), Yujin Robot, ROBOTIS, MagicEco and Rubypaper.

5D Robotics now hiring for positions in Cambridge, MA

| No Comments | No TrackBacks
From David Rohr via ros-users@

5D Robotics is coming to Greater Boston!

We recently opened an algorithms group on Mass. Ave in Cambridge and are looking to hire some local talent. The work will cover a wide range of topics and fields, so candidates need to be comfortable picking up new skills. Robotics is still young, so an ideal candidate would be broadly read to understand how similar problems are being attacked in more mature fields.

All work for the immediate future will be in simulation, so candidates will need to become very comfortable operating in rviz and Gazebo. As time goes on, we will acquire test space and begin shipping robots in from our Carlsbad office.

About Us

5D Robotics is a small company that has been doing R&D projects for the past several years while developing our core capabilities. We write software for mobile robots and create solutions for companies in need of automation. 5D has developed some amazing technology and we are ready to start deploying robots in the real world. We are on a high growth trajectory, so come join us in helping to usher in the age of robots!

About You

We need roboticists with a variety of backgrounds, but with an emphasis on mobile ground robots. All experience levels are invited to apply. At 5D, every project is unique and interesting. Every employee has a chance to make meaningful contributions while working with a great team. We encourage creativity and innovation backed up by intelligent decision making.

Required skills -

  • Proficiency in programming in C++.

  • Linux proficiency

  • Theoretical and practical knowledge of robotic systems

  • Bachelor's degree in relevant field or equivalent experience

  • Familiarity with ROS

  • Practical and theoretical knowledge of some of the following areas and interest in all

    • Probabilistic Robotics

    • Behavioral Robotics

    • Control Systems

    • Computer Vision

    • Motion Planning

    • Estimation and Navigation

    • Non-Linear Optimization

    • Filtering

Nice to have -

  • Experience with large code bases

  • Experience with DVCS's such as Git and Mercurial.

  • Previous work with open source software and similar systems.

  • Comfort with Gazebo simulator and rviz

  • Experience with robotic hardware

Apply to Bossa Nova Robotics

| No Comments | No TrackBacks
From David Lu via ros-users@

Bossa Nova Robotics is a rapidly growing Pittsburgh-based startup building robots that work around people. Our mission is currently under wraps, but we are venture-funded and were recently listed as a game changing startup by CBInsights [1]. And now, we're looking to hire ROS proficient software engineers for the next exciting phase of our company. 

Check out our job listings at We need roboticists of all sorts, including people to work on navigation, computer vision, controls, point clouds, long term automation and more. 

Apply now!

ROSCon 2015 Videos Posted

| No Comments | No TrackBacks

The videos from ROSCon 2015 have been posted! If you were unable to attend or catch the live stream we recorded all of the presentations as well as the lightning talks. You can find them linked from the program or view them all here

As an example here's Brian's opening remarks:

ROSCon 2015 Hamburg: Day 1 - Opening Remarks from OSRF on Vimeo.

ROSCon 2016 Venue Survey

While you are thinking about ROSCon please take a moment to fill out our survey about potential locations for ROSCon 2016 We want your input to help us choose the next location for ROSCon!

- Your ROSCon Organizing Committee

Thanks again to our Platinum Sponsors: Canonical / Ubuntu and Fetch Robotics!

And our Gold Sponsors: 3D Robotics, Bosch, Clearpath Robotics, GaiTech, Magazino, NVIDIA, Qualcomm, Rethink Robotics, ROBOTIS, Robotnik, ROS-Industrial, Shadow Robot, SICK, and Synapticon!

Introducing A Better Inverse Kinematics Package

| No Comments | No TrackBacks
From Patrick Beeson via ros-users@

TRACLabs Inc. is glad to announce the public release of our Inverse Kinematics solver TRAC-IK.  TRAC-IK is a faster, significantly more reliable drop-in replacement for KDL's pseudoinverse Jacobian solver.

Source (including a MoveIt! plugin) can be found at:

TRAC-IK has a very similar API to KDL's IK solver calls, except that the user passes a maximum time instead of a maximum number of search iterations.  Additionally, TRAC-IK allows for error tolerances to be set independently for each Cartesian dimension (x,y,z,roll,pitch.yaw).

More details:

KDL's joint-limited pseudoinverse Jacobian implementation is the solver used by various ROS packages and MoveIt! for generic manipulation chains.  In our research with Atlas humanoids in the DARPA Robotics Challenge and with NASA's Robotnaut-2 and Valkyrie humanoids, TRACLabs researchers experienced a high amount of solve errors when using KDL's inverse kinematics functions on robotic arms.  We tracked the issues down to the fact that theoretically-sound Newton methods fail in the face of joint limits.  As such, we have created TRAC-IK that concurrently runs two different IK methods: 1) an enhancment of KDL's solver (which detects and mitigates local minima that can occur when joint limits are encountered during gradient descent) and 2) a Sequential Quadratic Programming IK formulation that uses quasi-Newton methods that are known to better handle non-smooth search spaces.  The results have been very positive.  By combing the two approaches together, TRAC-IK outperforms both standalone IK methods with no additional overhead in runtime for small chains, and significant improvements in time for large chains.

Details can be found here in our Humanoids 2015 paper here:

A few high-level results are shown in the attached (low-res) figure.


Perception Neuron Motion Capture available in ROS

| No Comments | No TrackBacks
From Alexander Rietzler and Simon Haller

We are pleased to announce a new package for making the Biovision
Hierarchy (BVH) data generated by the Perception Neuron motion capture
system [1] available under ROS in Linux.

The software perception-neuron-ros [2] contains 2 packages:
-A ROS Serial package under Windows reads the BVH data and sends it to
the ROS Server
-A ROS package reads the BVH data and broadcasts the frames to TF

[1] (currently pre-orderable)

Find this blog and more at

Please submit content to be reviewed by emailing

Find recent content on the main index or look in the archives to find all content.