The SUTURO team is formed of members of the Institute for Artificial Intelligence (IAI) at the University of Bremen.
The team is headed by Prof. Michael Beetz, whose research interests include plan‐based control of robotic agents, knowledge processing and representation for robots, integrated robot learning, and cognitive perception. Before joining the University of Bremen, he lead the Intelligent Autonomous Systems Group at Technische Universität München that successfully investigated complex mobile manipulation tasks in a household environment.
All team members have substantial experience in a broad range of topics relevant for this challenge, including semantic task and scene representations for robots (Dr. Moritz Tenorth), object and scene perception (Ferenc Balint‐Benczedi), motion control (Georg Bartels) and plan‐based robot control (Jan Winkler).
A core part of the team will be formed by ten undergraduate students who have already implemented a mobile pick‐and‐place application on the PR2 robot that included novel perception, planning, control and reasoning modules.
The IAI investigates methods for Cognition‐enabled Robot Control, a research topic at the intersection of robotics and Artificial Intelligence.
This includes methods for intelligent perception, dexterous object manipulation, plan‐based robot control, and knowledge representation for robots. Robots performing complex tasks in open domains, such as autonomous logistics services, need to have cognitive capabilities for interpreting their sensor data, understanding scenes, selecting and parametrizing their actions, recognizing and handling failures and interacting with humans.
In our research, we are developing solutions for these kinds of issues and implement and test them on the robots in our laboratory. A particular focus of the group has always been on the integration of individual methods into complete cognition‐enabled robot control systems.
The IAI currently has a Willow Garage PR2 robot as well as an omni‐directional platform with two KuKa LWR 4 arms on a liftable torso and several kinds of grippers. The latter platform is very similar to the one to be used in the EuRoC challenges, which will facilitate integration and testing on a real platform besides the simulated environment.
Over the past years, this research has lead to demonstrations of robots making pancakes [http://youtu.be/4usoE981e7I], making popcorn and sandwiches [http://youtu.be/DTaeWITW1kI], going shopping and preparing Bavarian breakfast [http://youtu.be/gbIDPqb_2iM].
While these tasks have been demonstrated in a different domain, a human household, they included very similar (and partly more complex) actions.
Recently, the student members of the SUTURO team that will also participate in EuRoC have finished a one‐year project on mobile manipulation with the PR2 robot [http://youtu.be/WSfoMM8oAeQ].
The software that has been used for implementing these tasks has been published as open‐source code in ROS [http://www.knowrob.org, http://www.cram‐project.org] and is being used by several other research groups worldwide.
Michael Beetz, Dominik Jain, Lorenz Mösenlechner, Moritz Tenorth, Lars Kunze, Nico Blodow, Dejan Pangercic, "Cognition‐Enabled Autonomous Robot Control for the Realization of Home Chore Task Intelligence", In Proceedings of the IEEE, vol. 100, no. 8, pp. 2454‐2471, 2012.
Michael Beetz, Ulrich Klank, Ingo Kresse, Alexis Maldonado, Lorenz Mösenlechner, Dejan Pangercic, Thomas Rühr, Moritz Tenorth, "Robotic Roommates Making Pancakes", In 11th IEEERAS International Conference on Humanoid Robots, Bled, Slovenia, 2011.