Contents
Mission Statement
Building software for autonomous robots is a challenging, multi-disciplinary and time consuming task. Researchers often sub-optimally solve all topics and reinvent the wheel instead to provide great solutions in their central research area. In order to overcome this problem we propose a common software development based on ROS with the following goals:
- share and reuse of world-class solutions for common problems in USAR robots
- accelerate the development of search and rescue robots
- lower the entry-bar for new research groups
- continuous development of a common software framework
- train new people with existing software packages
Available Software
SLAM
hector_slam is a stack that provides a SLAM approach and associated tools that have been successfully used by numerous RoboCup Rescue League participants, including Best in Class Autonomy and overall competition champions.
Navigation
hector_exploration_planner is a package that provides a path planner for generating both goals and plans for the exploration of unknown environments.
Simulation
hector_nist_arenas_gazebo is a stack that provides tools for the fast and intuitive creation of simulated NIST standard test arenas like those in the RRL competition.
Optris Thermal Imager
optris_drivers is a stack wrapping the binary driver for Optris thermal imager devices.
QR code and victim detection
The hector_vision stack contains two packages that provide QR code detection on arbitrary camera images and victim (heat) detection based on thermal imaging. hector_qrcode_detection uses the zbar library internally. Both packages publish detections as worldmodel_msgs/ImagePercept messages that hector_object_tracker can track and visualize as markers in rviz or in the GeoTIFF map.
Manipulation
The tedusar_manipulation stack provides packages for semi-autonomouse manipulation. The packages are the outcome of a short-term visit of team TEDUSAR Graz at team Hector Darmstadt.
Tutorials for the hector_stack
- Installation of robocup_rescue software
This tutorial shows how to install software related to this initiative
- Simple Exploration of Unknown Environments in Simulation
This tutorial shows you how to start a simulation scenario with a robot exploring a maze scenario using the Gazebo simulator
Users
- Team Stabilize, Rajamangala University of Technology Phra Nakhon, Thailand
Team BARTLAB, Mahidol University, Thailand
Team Hector Darmstadt, Technische Universität Darmstadt, Germany
Team TEDUSAR, Technische Universität Graz, Graz, Austria - University Maribor, Maribor, Slovenia
- Team MRL
- Team Socrob RESCUE, Lisbon, Portugal
RRT-Team FH-Wels, University of Applied Sciences Upper Austria - Campus Wels
Team ResCube, Hungary
- others
Roadmap
Software under development
- 2D costmap determination from octomap
- Mono/Stereo Vision-based 6D-SLAM
Ideas looking for developers
- Victim detection (based on thermal images, RGB or RGBD cameras, motion, sound)
Events
In order to push this idea forward several workshops and summer schools took place. The different events serve as synchronization points, where new requirements are discussed and prototypes are developed. Future events are already planned. Stay tuned!
ROS Workshop for RoboCup Rescue teams, Koblenz, Germany, 2010
Workshop on Standard Robotic Software Architecture for RoboCupRescue based on ROS, Koblenz, Germany, 2011
ROS RoboCup Rescue Summer School 2012, Graz, Austria, 2012
SSRR Summer School 2012, Alanya, Turkey, 2012
Rescue Robotics Camp 2013 - colocated with SSRR 2013, Linkoeping, Sweden, November 2013
Links/Literature
[1] Stefan Kohlbrecher, Karen Petersen, Gerald Steinbauer, Johannes Maurer, Peter Lepej, Suzana Uran, Rodrigo Ventura, Christian Dornhege, Andreas Hertle, Raymond Sheh, Johannes Pellenz, "Community-Driven Development of Standard Software Modules for Search and Rescue Robots", Proc. IEEE International Symposium on Safety, Security and Rescue Robotics (SSRR), November 2012.
[2] S. Kohlbrecher, J. Meyer, O. von Stryk, and U. Klingauf. A flexible and scalable slam system with full 3d motion estimation. In Proc. IEEE International Symposium on Safety, Security and Rescue Robotics (SSRR), November 2011.
[3] Stephan Wirth and Johannes Pellenz. Exploration transform: A stable exploring algorithm for robots in rescue environments. Proceedings of the IEEE International Workshop on Safety, Security and Rescue Robotics (SSRR), 2007.
[4] Stefan Kohlbrecher and Johannes Meyer and Thorsten Graber and Karen Petersen and Oskar von Stryk and Uwe Klingauf. Hector open source modules for autonomous mapping and navigation with rescue robots. Proceedings RoboCup Symposium 2013