<<RepoHeader(roblab-whge-ros-pkg packages)>>

Examples and data sets

The first video shows a small mobile robot (Roomba), a 2D laserscanner (lms 100) and a standard digital camera (Canon). These components are combined to build 3D cognitive models. At first, the robots laserscanner takes a 3D scan with about 300.000 points. Then the point cloud is segmented into different regions, e.g walls, floors and ceiling. Depending on the laserscanners remission values, possible doorplate candidates ae calculated. Next, the robot drives to the doorplate candidate to make high resolution pictures with the Canon camera. Then OCR use those pictures to get the textual information out of the doorplate pictures. This information is then merged with the 3D point cloud, to display the doorplate information. The results of the study were that, the classical simultaneous localization and mapping (Slam), and the kidnapped robot problem is easily solved with this human like approach. Finally this is an important step to intelligent robotics, and a small revolution, since the approach is simple, universal and not expensive, because it only uses commercial hardware. The software is in the svn above.

The second video presents a novel approach to solve the localization problem for mobile robots in office environements by reading doorplates as non-artificial font based landmarks. Our implementation utilizes a high resolution Canon consumer camera, to detect and take pictures of the doorplate landmarks. The doorplates are found by a novel algorithm, which detects the layout of a doorplate in a viewfinder image (320x240) at 20 Hz in real-time. After the doorplate is found, the robot focuses the doorplate and takes a high resolution image with up to 9 mega pixels. Finally, the content of the doorplate is converted to text, by using standard OCR engines. This approach is the first important step towards reading robots since it uses only cheap consumer hardware. The laser scanner is not used.

The third video shows a novel approach, towards incorporating Peer-to-peer (P2P) networks for teleoperation of mobile robots throughout Wide Area Networks(WAN), as well as Local Area Networks (LAN). This is achieved by implementing a new operating interface, containing a Skype plugin, which is built up as a graphical user interface and comes along with audio and video windows. Utilizing Skype P2P technology, the client software is relieved of being responsible for the network layer, secure authentication and high performance audio and video streaming. Since Skype is routed through firewalls, the teleoperation system proposed is easily usable by humans. The design of the interface is led by the hypothesis, that the operator using the controls, is familiar with the Skype audio and video window, joystick control and a map visualization similar to the ones used in ego-shooter game play. The usage of the teleoperation software, is made far more intuitive and consumer friendly than in other interfaces. Two major design principles of both the robot control software, and the interface software are the reusability and scalability, so that various robot platforms are supported. In order to reach this result ROS (Robot Operating System) and generic software models are used.

Disaster areas require mobile robots with extreme capabilities. This video presents an approach based on simple disposable mobile robots in contrast to very expensive mobile systems for rescue missions. The mobile robot consists of cheap standard commercial components. It provides a view into the disaster area with its sensors e.g. cameras, and in addition and very important, it sets up a persistent communication between all mobile robots and users. This is done by using intermeshed networks to overcome the destroyed network infrastructure. Furthermore the robots are ready for cloud computing and its sensor data can be used there! (Explanations at the beginning of the video are in German e.g. this is the micro controller for generation of the PWM signals, network camera and patched router ...)

More videos can be found at [http://www.youtube.com/user/RoblabFhGe Youtube channel of our Robotic lab] and the [http://www.youtube.com/user/VolksbotCtrl Youtube channel of the Volksbot]

Kinect data sets for V-SLAM

Testing different SLAM approaches needs data sets with ground truth information. Therefore we use our Kuka KR60 and mount a Kinect on top of it. With the help of the Manipulator we drive different trajectories with a sub-millimeter precision. The following links provide the ROS bag files extend by videos of the experiments. If you use the data set reference it as:

  • Author: Thorsten Linder, Salvador Dominguez, Nils Oppermann, Hartmut Surmann, University of Applied Science Gelsenkirchen, Fraunhofer IAIS, "The Kuka KR60 and Kinect reference data set", september 2011.


We also make video lectures to introduce ROS (german / english) to students. May be the following list is helpfull for users:

A brief instruction on how to install ROS on your Ubuntu system

[http://www.youtube.com/watch?v=hJ9J6_yyotg How to install ROS(GER)]

[http://www.youtube.com/watch?v=afiSgYF3Pdo How to install ROS(ENG)]

How to create a ROS package

[http://www.youtube.com/watch?v=jZvtNSkIH_4 ROS Package(GER)]

[http://www.youtube.com/watch?v=oT3w99H_HQg ROS Package(ENG)]

How to use Eclipse with ROS

[http://www.youtube.com/watch?v=YFUifnU_nOA Eclipse with ROS(GER)]

[http://www.youtube.com/watch?v=grlBIimWXgU Eclipse with ROS(ENG)]

Creating ROS msg's and srv's

[http://www.youtube.com/watch?v=_-PaGg_EAi0 ROS msg's and srv's(GER)]

[http://www.youtube.com/watch?v=ddJm5OAt9JM ROS msg's and srv's(ENG)]

Understanding ROS-nodes

[http://www.youtube.com/watch?v=UCLLjnoFukU Understanding ROS-nodes(GER)]

[http://www.youtube.com/watch?v=OJ7yJF1dETE Understanding ROS-nodes(ENG)]

Understanding ROS Topics

[http://www.youtube.com/watch?v=NDEeK05nSZ8 Understanding ROS-Topics PART1 (GER)]

[http://www.youtube.com/watch?v=5VQgo3T-Vv0 Understanding ROS-Topics PART2 (GER)]

[http://www.youtube.com/watch?v=H9JeKbQP_1Y Understanding ROS-Topics PART1 (ENG)]

[http://www.youtube.com/watch?v=5WSWXA5kdZs Understanding ROS-Topics PART2 (ENG)]

ROS Services and Parameters

[http://www.youtube.com/watch?v=Go4rUi6bvLg ROS Services and Parameters (GER)]<br>

[http://www.youtube.com/watch?v=Ij8Kfl5D0eQ ROS Services and Parameters (ENG)]<br>

How to write a launch file

[http://www.youtube.com/watch?v=mNpm95-Mws4 How to write a launch file in ROS(GER)]

[http://www.youtube.com/watch?v=Qe12R8tDXzg How to write a launch file in ROS(ENG)]

Publisher and Subscriber in ROS<br>

[http://www.youtube.com/watch?v=pKF8aDV_x-w Publisher and Subscriber in ROS Part 1 (GER)]

[http://www.youtube.com/watch?v=2fJmzFK3L7s Publisher and Subscriber in ROS Part 2 (GER)]

[http://www.youtube.com/watch?v=_bYw2JlwHqA Publisher and Subscriber in ROS Part 1 (ENG)]

[http://www.youtube.com/watch?v=qUY7H8L_et8 Publisher and Subscriber in ROS Part 2 (ENG)]

Service and Client in ROS<br>

[http://www.youtube.com/watch?v=5qBSkwVIdE0 Service and Client in ROS Part 1 (GER)]

[http://www.youtube.com/watch?v=F27GZl61GXg Service and Client in ROS Part 2 (GER)]

Wiki: roblab-whge-ros-pkg (last edited 2012-02-08 09:35:36 by roblabfhge)