October 2013 Archives

Release of catkinized nao_robot stack 0.2

| No Comments | No TrackBacks
From Armin Hornung via ros-users@

Dear ROS users,

it's my pleasure to announce a new release of packages for the Nao robot. Most notably, that's v0.2 of the nao_robot stack (or now, metapackage), which is now catkinized for ROS groovy and hydro thanks to Miguel Sarabia. The stacks nao_robot and nao_common also saw a complete reorganization of packages and an overhaul to enable easier usage on the robot, thanks to Séverin Lemaignan. Parts of the nao_common stack have moved into nao_robot, with the remainder in the new stack nao_extras (still rosbuild as of now).

All code should be compatible with ROS groovy and hydro, as well as NaoQI 1.12 and 1.14.

Grab the latest versions (tagged release / source package, or clone the master branch) here:

Source install instructions are available at http://wiki.ros.org/nao/Installation/remote

Thanks to Séverin, it's now really easy to compile ROS and the Nao packages to run directly on the robot:

Precompiled binary packages for the Nao and Debian packages to run your machine are in preparation.

You can find separate changelogs in the packages, the complete lists of changes are available here:

If you want to discuss future development or contribute, consider jointing the ros-sig-aldebaran mailing list:

Since this was a combined community effort, I would like to thank thank Miguel Sarabia, Séverin Lemaignan, Daniel Maier, and Sebastian Boettcher for their contributions.

Best Regards

MoveIt Feedback Survey Posted

| No Comments | No TrackBacks
From Sachin Chitta of SRI:

We are polling the community to get technical feedback about MoveIt!. It should take less than 5 minutes to fill out the survey. This form will stay open until Friday Nov 1, 2013, 11:59 PM PST
A compiled summary of the survey results will be made available to the community through the MoveIt! wiki (moveit.ros.org). We appreciate any feedback you can provide to help us improve MoveIt!

Rethinking ROS on Baxter

| No Comments | No TrackBacks
From the Open Source Robotics Foundation

Last week was an exciting one in the robotics community, as hundreds of people and dozens of robotics-related companies descended onRoboBusiness 2013 in Santa Clara, CA; just down the road from OSRF headquarters.

In all the enthusiasm and bot-gawking, we wanted to point out some exciting ROS-related news that might be have been overlooked. Rethink Robotics, the company behind Baxter, enhanced the open source capabilities of their research robot.

The Baxter Research Robot now supports MoveIt!, allowing researchers to plan and test trajectory algorithms in a virtual environment before the moves are executed on the robot; and offers integrated support for Gazebo, the widely used physics-based robotics simulator.

Those of you who have followed OSRF's contributions to the DARPA Robotics Challenge, in particular the Virtual Robotics Challenge are already familiar with the capabilities of Gazebo and MoveIt!, but it's great to see another top-notch robotics company embrace and extend ROS.

We had the opportunity to host Dave Coleman this past summer and witness the rise of Baxter in both reality and simulation. Dave worked closely with Gazebo, ROS, and Rethink to simulate Baxter and develop its controllers. His hard work helped the Rethink team deliver Baxter SDK version 2.0, and added another simulated robot for the robotics community.

Clearpath Robotics' Husky Goes Hydro

| No Comments | No TrackBacks

Clearpath Robotics announces Hydro Medusa Support

Husky software is now available for the new ROS Hydro Medusa Platform

(Kitchener, ON, Canada - October 17, 2013) Clearpath Robotics has launched Husky software for ROS' latest distribution, Hydro. The largest change in the package was the transition to the new ROS build system, catkin. Significant improvements to the Gazebo-based Husky simulator and basic autonomy are also included, completely free of charge. The package, and all other Husky ROS software, is available at www.github.com/husky.

"Along with the improvements we've made to our own libraries, upgrading to ROS Hydro makes available a significant set of other new features, including compatibility with the latest version of Gazebo and easy access to the alpha release of the MoveIt! manipulation library," said Ryan Gariepy, CTO at Clearpath Robotics. "The release of ROS Hydro validates the ability of the global robotics community to sustain a common software framework, while simultaneously adding exciting new functionality."

This Husky software release has maintained topic names and clearpath_base message types to ensure a smooth transfer from ROS versions as far back as ROS Electric. Users can continue to use rosbuild workspaces in Hydro. It is recommended to begin the migration to the catkin framework as rosbuild will be unavailable in the next ROS distribution (Indigo Igloo), currently slated for release in April 2014. Husky for Hydro can be used alongside previous versions of ROS on the same workstation. 

For those who don't yet own a Husky, a simulator package is available here (http://wiki.ros.org/husky_simulator) to enable dynamically-accurate Husky simulation on Ubuntu desktop with one command.

Clearpath will continue to publish detailed guides to the use of ROS software, starting with an introductory article which can be found here (http://www.clearpathrobotics.com/husky-for-ros-hydro/)

ROS for Labview Preview from Tufts CEEO

| No Comments | No TrackBacks
The Mechanical Engineering Department and the Center for Engineering Education and Outreach at Tufts University are releasing an alpha version of a ROS client library for Labview with the goal of creating a platform independent and intuitive ROS environment. 

In its current state, the library supports publishers, subscribers, and service clients on predefined topics and contains a partial implementation for the Baxter Research Robot from Rethink Robotics. The library was built in Labview 2010 on OS X and has been tested successfully in Windows 7.


This is a prerelease and should not be expected to remain stable. UI and structural changes should e expected along with further feature support. If you are interested please try it out and give us feedback at ros-sig-rosforlabview@googlegroups.com You can find the package on github at https://github.com/chris-smith/ROSforLabview

University of Costa Rica Explores Aerospace Research

| No Comments | No TrackBacks
Clearpath Robotics announces the use of their Husky in Costa Rica


Dr. Geovanni Martinez from the University of Costa Rica has developed a novel visual odometer algorithm for accurate and more efficient tracking of Mars rover navigation. Dr. Martinez is utilizing Clear path Robotics' Husky to test and validate the algorithm that uses one-stage maximum-likelihood estimation, rather than traditional two-stage algorithms.

"It's fantastic to witness breakthrough research of this nature, and to know that it is being validated and furthered because of our mobile robotic platform," said Matt Rendall, Chief Executive Officer at Clearpath Robotics.

Dr. Martinez' team is creating a real time image acquisition system consisting of three IEEE-1394 cameras. The system is being developed under Ubuntu 12.04.2 LTS, ROS Fuerte and the programing language C". The image acquisition system corrects, in real time, the radial and tangential distortions due to the camera lens. With regard to the hardware, Dr. Martinez commented, "We like Husky A200 because the software for image acquisition, and driving the robot, was easy to implement using ROS. It saved us a lot of development time. Additionally, it is strong enough to be driven in extreme environments."

Using the algorithm, the rover's motion will be estimated by maximizing the conditional probability of the frame to frame intensity differences at the observation points. The conditional probability is computed by expanding the intensity signal by a Taylor series and neglecting the nonlinear terms. This results in the well-known optical flow constraint, as well as using a linearized 3D observation point position transformation, which transforms the 3D position of an observation point before motion into its 3D position after motion given the rover's motion parameters. Perspective projection of the observation points into the image plane and zero-mean Gaussian stochastic intensity errors at the observation points are also assumed.

Find this blog and more at planet.ros.org.

Monthly Archives

About this Archive

This page is an archive of entries from October 2013 listed from newest to oldest.

September 2013 is the previous archive.

November 2013 is the next archive.

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