DLR German Aerospace Center Institute of Robotics and Mechatronics
I am head of the "Autonomous Learning Robots Lab”. Our longterm goal in this basic research lab is to use learning as the core principle in building really autonomous robots which can operate in complex and changing environments. A prerequisite for this research program is robotic hardware which comes close to human performance with respect to sensor and motor capabilities and has a realtime connection to large and scalable computing resources. In the last years we upgraded one of our torque controlled humanoid robots to our award winning research platform Agile Justin [2] which, among others, now has realtime 3D environment modeling, high resolution spatio-temporal tactile sensing on the whole body and the hands and a wirelessly coupled GPU cluster in-the-loop.
Currently, amongst others, we use (deep) learning for tactile sensor processing and deep reinforcement learning for fast whole body motion planning in self-acquired 3D models. In the next steps we are heading for the autonomous learning of solving planning problems at the task level. Our main benchmark scenarios are to autonomously build a Mars habitat (constructing a 3D structure from tube elements) and to catch and throw balls which brings all hard- and software components to its limits.
Recent Awards
IROS2016 Best Paper on Cognitive Robotics Finalist
IROS2016 Best Student Paper Finalist
[1] S. Baishya and B. Bäuml. Robust material classification with a tactile skin using deep learning. In Proc. IEEE International Conference on Intelligent Robots and Systems, 2016.
[2] B. Bäuml, T. Hammer, R. Wagner, O. Birbach, T. Gumpert, F. Zhi, U. Hillenbrand, S. Beer, W. Friedl, and J. Butterfass. Agile Justin: An upgraded member of DLR’s family of lightweight and torque controlled humanoids. In Proc. IEEE International Conference on Robotics and Automation, 2014.
[3] O. Birbach, B. Bäuml, and U. Frese. Automatic and self-contained calibration of a multi-sensorial humanoid’s upper body. In Proc. IEEE International Conference on Robotics and Automation, 2012.
ICRA 2011 Best Video Award Finalist
[4] B. Bäuml, F. Schmidt, T. Wimböck, O. Birbach, A. Dietrich, M. Fuchs, W. Friedl, U. Frese, C. Borst, M. Grebenstein, O. Eiberger, and G. Hirzinger. Catching Flying Balls and Preparing Coffee: Humanoid Rollin’Justin Performs Dynamic and Sensitive tasks. In Proc. IEEE International Conference on Robotics and Automation, 2011.
[5] H. T ̈aubig, B. Bäuml, and U. Frese. Real-time swept volume and distance computation for self collision detection. In Proc. IEEE International Conference on Intelligent Robots and Systems, 2011.
Last updated on Wednesday, 20 April 2022 by Berthold Bäuml