October 2014 Archives

From Nick Hawes via ros-users@

I would like to announce the release of a new suite of tools to enable the persistent storage, analysis and retrieval of ROS messages in a MongoDB database.

The mongodb_store package:

http://wiki.ros.org/mongodb_store

... provides nodes to store arbitrary ROS messages in a MongoDB database, query the database and retrieve messages, with helper classes in C++ and Python. Nodes are also available to provide rosbag-like functionality using the [same db format] (http://wiki.ros.org/mongodbstore#LoggingofTopics:mongodblog) and also [parameter persistence across system runs] (http://wiki.ros.org/mongodbstore#Parameterpersistence:config_manager.py).

Packages are available on Ubuntu for Indigo and Hydro, e.g.

• ros-indigo-mongodb-log - The mongodb_log package
• ros-indigo-mongodb-store - A package to support MongoDB-based storage and analysis for data from a ROS system, eg. saved messages, configurations etc
• ros-indigo-mongodb-store-msgs - The mongodbstoremsgs package

These tools were developed by the STRANDS project to support the development, debugging and runtime introspection of long-term autonomous mobile robots, but we hope they will be useful to the ROS community more generally.

In the near future we plan to release tools for serving maps from mongodb_store and for logging streams of RGB-D data in a compressed format .

Note that there is an overlap in functionality between these tools and warehouse-ros. We developed our own solution as the existing packages appeared to be unsupported and special-purpose, but as this appears to be changing, we may want to look at combining these two packages.

For feedback, pull requests, feature requests and bug reports please go to: https://github.com/strands-project/mongodb_store/issues

From Michael Ferguson via ros-users@

FYS Systems is looking for software engineers with broad experience on real robot platforms and a deep background in one or more of the following areas: navigation, motion planning, robot perception, human-robot interfaces. We have multiple openings, for both junior and senior-level engineers, with both immediate start dates and throughout early 2015.

Required Skills:

• BS or MS in Computer Science, Robotics, or a related field.
• 1+ year of software engineering experience, or extensive software engineering experience in undergraduate or graduate program.
• Experience with C++ and/or Python in a Linux Environment.
• Experience in robot navigation, motion planning, perception, or human-robot interfaces.
• Experience with Robot Operating System (ROS).

Nice To Haves:

• Experience with MoveIt, SBPL and/or OMPL.
• Experience with OpenCV or PCL.
• Experience with web development.
• Experience with CMake.

To apply, please visit: https://fyssystems.has-jobs.com/SoftwareEngineerSunnyvale/32999/0

From Anis Koubaa via ros-users@

I am coordinating with Springer publisher to edit a handbook on Robot Operating System. There are only a few books on ROS http://wiki.ros.org/Books which mainly represent a brief introduction to ROS and a few basic applications. This does not translate the huge amount of work being done in the community and I feel the need to have a complete reference on the topic.

The prospective handbook will cover ROS from foundations and basics to advanced research works from both academia and industry. Tutorials and research papers will both be sought. The book should cover several robotics areas including but not limited to robot navigation, UAVs, arm manipulation, multi-robot communication protocols, Web and mobile interfaces using ROS, integration of new robotic platform to ROS, computer vision applications, development of service robots using ROS, development of new libraries and packages for ROS, etc. Every book chapter should be accompanied with a working code to be put later in a common repository for the readers.

To express your interest to the handbook and your intention of a chapter proposal, I would like to invite you fill in the following form. The proposed chapters are just considered as an initial expression of interest and will be included in the handbook proposal. It does not mean any kind of commitment for the author at this stage. An official call for chapters with instructions and deadline will be announced soon.

Thank you and look forward to receive your feedback,

Anis

NavVis presents new large scale mapping technology: The impressive shipping exhibition at Deutsches Museum can now be explored online and in 3D.

NavVis utilizes a trolley that's equipped with three laser scanners and six cameras. As a human operator wheels that trolley through the area to be mapped, the scanners record the horizontal and vertical dimensions of the hallway while the cameras record panoramas. The software framework is based on ROS, which allowed for a very modular design, as Suat Gedikli, Chief Software Architect at NavVis, says. One of the main advantages of their mapping device is its efficiency: In contrast to Google's mapping trolley, the operator does not have to crouch to move out of the field of view of the panoramic camera. With the patented NavVis camera head, the six cameras are assembled in a way such that the trolley operator is in their blind spot and therefore not visible. Hence 360-degree panoramas can be continuously recorded while moving the trolley.

The start-up recently teamed up with Deutsches Museum to digitize their impressive shipping exhibitions, which were mapped in less than one hour. The result is a 3D map of the exhibition, overlaid with photos of every square inch of all the surfaces.

Similar to Google Street View, their HTML5 based IndoorViewer allows people to virtually explore the museum online. Additionally, administrators can add content like text, images, video to various points on the map, which can then be accessed by users. This interactive feature also lets users do things like obtaining measurements between different points in the building (which is important for architects and in construction site monitoring applications).

NavVis, which was only founded in May of last year, is focused on public-oriented showcases and business-to-business applications. Felix Reinshagen, co-founder and managing director: "As the first step, our application is aimed at companies in the building management segment that are confronted with challenges such as documentation, inventory, path-finding and task management. We make our hardware, software and service available to them so that they can have their building and industrial plants digitized while handling the usage of their data according to their own requirements. An area the size of the Deutsches Museum is mapped and posted online within three working days for a four-figure amount."

From Paul Hvass, via ros-users@

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.

More information can be found on the ROS-I blog post.

ROS Groovy Galapagos Rosdistro Git Activity from OSRF on Vimeo.