Willow Garage Blog
March 15, 2010
In recent posts, we've showcased the rviz 3-D visualizer and navigation stack, two of the many useful libraries and tools included in the ROS Box Turtle distribution. Now, we'd like to highlight what we're developingt for future distribution releases.
The first is the PR2 calibration stack. The PR2 has two pairs of stereo cameras, two forearms cameras, one high-resolution camera, and both a tilting and base laser rangefinder. That's a lot of sensor data to combine with the movement of the robot.
The PR2 calibration stack recently made our lives simpler when updating our plugging-in code. Eight months ago, without accurate calibration between the PR2 kinematics and sensors, the original plugging-in code applied a brute force spiralling approach to determining an outlet's position. Our new calibration capabilities give the PR2 the new-found ability to plug into an outlet in one go.
The video above shows how we're making calibration a simpler, more automated process for researchers. The PR2 robot can calibrate many of its sensors automatically by moving a small checkerboard through various positions in front of its sensors. You can start the process before lunch, and by the time you get back, there's a nicely calibrated robot ready to go. We're also working on tools to help researchers understand how well each individual sensor is calibrated.
The PR2 calibration stack is still under active development, but can be used by PR2 robot users with Box Turtle. In the future, we hope this stack will become a mature ROS library capable of supporting a wide variety of hardware platforms.
March 15, 2010
Logging and playback is one of the most critical features when developing software for robotics. Whether you're a hardware engineer recording diagnostic data, a researcher collecting data sets, or a developer testing algorithms, the ability to record data from a robot and play it back is crucial. ROS supports these needs with "Bag files" and tools like rosbag and rxbag.
rosbag can record data from any ROS data source into a bag file, whether it be simple text data or large sensor data, like images. The tool can also record data from programs running on multiple computers. rosbag can play back this data just as it was recorded -- it looks identical. This means that data can be recorded from a physical robot and used to create virtual robots to test software. With the appropriate bag file, you don't even need to own a physical robot.
There are a variety of tools to help you manage your stored data, such as tools for filtering data and updating data formats. There are also tools to manage research workflows for sending bag files to Amazon's Mechanical Turk for dataset labeling. For data visualization, there is the versatile rxbag tool. rxbag can scan through camera data like a movie editor, plot numerical data, and view the raw message data. You can also write plugins to define viewers for your own data types.
You can watch the video to see rosbag and rxbag in action. Both of these tools are part of the ROS Box Turtle release, which you can download from ROS.org.
March 11, 2010
Last Friday was Discovery Day at William Regnart Elementary school in Cupertino, California. Among the career day presenters were doctors, firefighters, nurses, dog trainers, and a Texai. Dallas Goecker, a Willow Garage electrical engineer working from Seymour, Indiana, and Rob Wheeler, a local Willow Garage software engineer, presented their jobs and the story of the Texai, our tele-operated tele-presence robot.
Dallas was able to weave around desks and fourth-graders, demonstrating the capabilities of his robot. Despite some teasing after losing wireless connectivity on the playground, Dallas and Rob received certificates of appreciation, one of which is still proudly taped to the Texai.
[The Texai is previously referred to as the Texas.]
March 9, 2010
The ROS community has grown quickly over the past year, greatly increasing the number and variety of robots integrated with ROS. Over on ROS.org, there's a blog series highlighting some of these robots, as well as the open source ROS Repositories behind them.
The first four installments include:
- STAIR 1: the Stanford University mobile manipulation research platform that provided the predecessor of the ROS framework.
- Aldebaran Nao: a small, commercially available humanoid robot that demonstrated the ability of the ROS community (Brown University and University of Freiburg) to come together and develop open source drivers.
- i-Sobot: an even smaller humanoid robot controlled by the ROS PS3 joystick driver. The developer has been publishing a Japanese-language blog on ROS, helping ROS reach new audiences.
- Junior: Stanford Racing's autonomous car that finished a close second in the DARPA Urban Challenge. Junior's main software framework is IPC, but ROS's modular libraries have made it easy to integrate ROS-based perception libraries into their obstacle classification system.
For more information, please see the full posts on ROS.org.
March 8, 2010
Three researchers from the JSK Lab at Tokyo University, Ryo Hanai, Kimitoshi Yamazaki and Hiroaki Yaguchi, took advantage of their two-week Spring vacation to gain some hands-on experience with our PR2 Beta robots and ROS 1.0. This was JSK's second visit to Willow Garage, and we were excited to watch this new group of researchers use the PR2 for the first time.
With only a couple of ROS tutorials under their belts, the researchers learned how to use ROS libraries for 3D perception, navigation, and controllers. During their two-week stay, the team put together four demos that accomplished various cleanup tasks. In the first, the PR2 detected items on a table and transferred them to a tray before carrying everything across the room. The second and third demos had the PR2 pick up and put away dishes and clothing, while the final demo used visual features to differentiate between two similar items (book and box).
These demos required a variety of perception approaches. The group applied a circular feature detector to identify the patternless plate, and used the shirt's wrinkles as features to identify the clothing. To discern between the box and book, the PR2 pulled each item closer to its cameras, and then used SURF for item identification. We hope our new tidy PR2 means that we can soon stop picking up after ourselves around the office.
We were very impressed with what the JSK visitors were able to accomplish, especially with community efforts to translate the ROS documentation only just underway. Check out the video to see the demos and hear more about JSK's experience at Willow Garage. You can also take a look at our visit to Japan last June, when we worked together to port the Navigation stack to JSK's Kawada HRP2-V robot.
March 7, 2010
In addition to core robotics libraries, like navigation, the ROS Box Turtle release also comes with a variety of tools for developing robotics algorithms and applications. One of the most commonly used tools is rviz, the 3-D visualization environment that is part of the ROS visualization stack.
Whether it's 3-D point clouds, camera data, maps, robot poses, or custom visualization markers, rviz can display customizable views of various types of robot data. rviz can show you the difference between the physical world and what the robot is actually seeing, and it can also help you create displays that show users what the robot is planning to do.
You can watch the video above for more details about what the ROS rviz tool has to offer, and you can read documentation and download the source code at: ros.org/wiki/rviz.
March 3, 2010
Now that the ROS Box Turtle release is out, we'd like to highlight some of its core capabilities, and share some of the features that are in the works for the next release.
First up is the ROS Navigation stack, perhaps the most broadly used ROS library. The Nagivation stack is in use throughout the ROS community, running on robots both big and small. Many institutions, including Stanford University, Bosch, Georgia Tech, and the University of Tokyo, have configured this library for their own robots.
The ROS Navigation stack is robust, having completed a marathon -- 26.2 miles -- over several days in an indoor office environment. Whether the robot is dodging scooters or driving around blind corners, the Navigation stack provides robots with the capabilities needed to function in cluttered, real-world environments.
You can watch the video above for more details about what the ROS Navigation stack has to offer, and you can read documentation and download the source code at: ros.org/wiki/navigation.
March 1, 2010
We are excited to announce that our first ROS Distribution, "Box Turtle", is ready for you to download. We hope that regular distribution releases will increase the stability and adoption of ROS.
A ROS Distribution is just like a Linux distribution: it provides a broad set of libraries that are tested and released together. Box Turtle contains the many 1.0 libraries that were recently released. The 1.0 libraries in Box Turtle have stable APIs and will only be updated to provide bug fixes. We will also make every effort to maintain backwards-compatibility when we do our next release, "C Turtle".
There are many benefits of a distribution. Whether you're a developer trying to choose a consistent set of ROS libraries to develop against, an instructor needing a stable platform to develop course materials against, or an administrator creating a shared install, distributions make the process of installing and releasing ROS software a more straightforward process. The Box Turtle release will allow you to easily get bug fixes, without worry about new features breaking backwards compatibility.
With this Box Turtle release, installing ROS software is easier than ever before with Ubuntu debian packages. You can now "apt-get install" ROS and its many libraries. We've been using this capability at Willow Garage to install ROS on all of our robots. Our plugs team was able to write their revamped code on top of the Box Turtle libraries, which saved them time and provided greater stability.
Box Turtle includes two variants: "base" and "pr2". The base variant contains our core capabilities and tools, like navigation and visualization. The pr2 variant comes with the libraries we use on our shared installs for the PR2, as well as the libraries necessary for running the PR2 in simulation. Robots vary much more than personal computers, and we expect that future releases will include variants to cover other robot hardware platforms.
We have new installation instructions for ROS Box Turtle. Please try them out and let us know what you think!
March 1, 2010
At the beginning of the year, we released our PR2 Beta Program Call for Proposals. In response, we received an overwhelming 120 letters of intent to respond. It's now March 1st, and we're excited to see the final proposals coming in. Thanks to all of those who have taken interest in this program. We appreciate the effort that has gone into assembling the proposals, and we look forward to reading each one in detail.
Check out the video above for more information on the PR2, ROS, and the PR2 Beta Program.
February 26, 2010
We've got a lot of videos in pipeline coming up for you that explain various features of ROS, PR2, and Texas platforms. In the meantime, we've put together a montage -- enjoy!
Subscribe