DS3 -- Environmental Constraints Exploitation

=== Description ===
This dataset contains point cloud data of the experimental setup, links tothe repositories of EC primitive detectors and grasp planner.

=== How was the data recorded? ===
The data was recorded with ROS Indigo bag file recorder tool, we recorded depth image under topic /camera/depth of the scene with an Asus Xtion PRO LIVE RGBD camera mounted on a Barrett WAM arm. The joint statetes and camera position is recorded in /joint_state and /tf topic respectively while /webcam2/image_raw (LogiLink UA00724) shows a top down view of the scene.  The planner results are included under the topic /geometry_graph (ECE Graph), the list of detected objects poses and bounidng boxes /objects and /object_poses topics respectively. an IFCO tote had a cucumber inside and the RGBD camera was pointet to look inside the bin. The data was recoreded at the Robotics and Biology Laboratory, TU-Berlin.

=== Relevant links ===
Link to Vision repository: https://github.com/SoMa-Project/vision.git
Link to Planner repository: https://github.com/SoMa-Project/ec_grasp_planner.git

=== How to open the files? ===
Open the bag files with rqt_bag within a ROS Indigo OS

=== Corresponding publication(s) ===
The data shows the grasp planner and visual perception system inputs and outputs. The paller is first published in Clemens Eppner and Oliver Brock. Planning Grasp Strategies That Exploit Environmental Constraints. Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pp. 4947 - 4952, 2015. The planner and visual preception system code are on gihub and the link is included inot the Readme.txt. More information about the mehotds can be found in Delivarable4.1 H2020 Soft Manipualtion project.
Link: http://www.robotics.tu-berlin.de/fileadmin/fg170/Publikationen_pdf/eppner_icra2015.pdf