This amped up, current release of ROS features new, stable libraries for arm navigation and point cloud processing. There is also improved support for Android, Arduino, Windows, ARM, and Python 3. Please see the release page for some of the cool new tools and libraries in Electric, such as "interactive markers" for creating custom GUIs in RViz.
Batteries included, but modular
We strive to integrate ROS with the best open-source software libraries for robotics out there, while maintaining a modular system that allows you to pick and choose exactly what you need. "Smaller, lighter" was a theme of our Diamondback release, and we've continued that with ROS Electric by streamlining integration and making it more portable. We are now using standard system installs of many libraries, such as OpenCV, assimp, yaml-cpp, TinyXML, and Eigen, so that you can use them the same with or without ROS.
ROS Electric separates many of the lower-level ROS libraries into smaller components so that you only have to use what you need. KDL, nodelet, filters, xacro, and pluginlib are now available separately, and KDL in ROS is now maintained directly by the Orocos developers. We have also updated ROS to allow unary stacks.
Arm navigation 1.0
The arm navigation library has reached its 1.0 release, which has a stable ROS API. The arm navigation library enables you to execute collision-free motion plans and control for a robot arm manipulator. It is compatible with multiple motion planners, including ompl and sbpl. This release includes a new setup wizard to help you configure arm navigation for your robot manipulator. It also includes new visualization tools for working with motion planners and a powerful experimental GUI tool for generating, saving, and playing back planning scenes.
Please see the updated tutorials for a guide to these new tools.
Thanks to David Gossow, the rviz 3D visualization tool now has "interactive markers": programmable user interface components. You can use these to create custom GUIs, such as controls for moving the arms, gripper, and base of your robot.
The most recent release of Point Cloud Library is available with ROS. This release contains lots of new features, optimizations, and bug fixes, and we have also performed numerous user tests to improve PCL's documentation and tutorials. Please see the PCL 1.1 change list for more information.
OpenCV 2.3, standalone
The vision_opencv stack now integrates with the standalone version of OpenCV and has been updated to OpenCV 2.3 (change list). We are now hosting a debian binary package of OpenCV 2.3, which is used for both standalone installations of OpenCV as well as ROS integration.
Android, Pure Java
Damon Kohler at Google has created a new rosjava library that is written in pure Java and is compatible with Android. We will soon be releasing components that let you create UIs on Android devices to interact with ROS robots as well as use sensors on Android devices in your robot system.
Michael Ferguson and Adam Stambler have created a new rosserial library that lets you use Arduino hardware with ROS. This library sets up a point-to-point connection with Arduino hardware so that you can send ROS messages back and forth with another system.
Daniel Stonier (Yujin Robot) is hosting a win_ros stack that collects libraries and tools for using ROS on Windows. Daniel has also provided numerous patches to integrate the in-progress rosbuild2 system that provides more direct integration with standard Windows build toolchains.
Debian binaries of several ROS stacks are now available for armel systems. This enables binary installs on platforms like the PandaBoard and Gumstix.
Thanks to Michael Karg (TUM), Severin Lemaignan (LAAS), and Lorenz Mösenlechner (TUM) for numerous patches to many of the ROS libraries compatible with Python 3. This update enables seamless use of ROS with the MORSE simulator.