Workshop IROS 2015
Physical Human-Robot Collaboration: Safety, Control, Learning and Applications
2015 IEEE/RSJ International Conference on Intelligent Robots and Systems, Hamburg, Germany
Friday, October 2nd, 2015, 08:00-17:30, Saal E
Recent advances in robotic hardware and control now enable robots to interact physically with humans. This mode of interaction opens up entirely new methods for human-robot collaboration, as physical interaction provides a direct and intuitive channel to synchronize movements and communicate intentions; in partner dance this channel is called "the connection".
Achieving "the connection" places challenging requirements on the safety, robustness and adaptivity on robots. But what exactly are the safety requirements on such robots? And how can such requirements be formalized? From a control perspective, classical high-gain position control is not feasible due to high potential contact forces. How can compliance and force control be best exploited to enable physical interaction? Should we sacrifice accurate tracking?
One of the advantages of human-robot collaboration is that humans can demonstrate solutions to novel tasks to the robot. Thus, robots will have to be able to learn new skills and adapt to novel task contexts. Which forms of imitation learning should be used? And how can we associate the correct solution with a given task context?
Finally, what are relevant and profitable applications for physical human-robot collaboration? In which cases is it beneficial over fully automated solutions? How can it facilitate customization of products in assembly lines? Can it also be used in medical applications, such as surgery and rehabilitation? Can technologies for physical human-robot collaboration be extended to wearable robots and prosthetics? These and other questions will be addressed in lectures and an expert panel discussion.
The objective of this workshop is to bring together actors from both research and industry to discuss the open research questions above, and acquire an overview of practical applications in this domain.
Topics of interest
- Physical Human-Robot Interaction
- Safety in HRI
- Compliant Control
- Imitation Learning
- Alin Albu-Schäffer, DLR Institute of Robotics and Mechatronics, Germany
- Heni Ben Amor, Arizona State University, USA
- Matthias Bjoern, ABB Corporate Research, Germany
- Sami Haddadin, Leibniz Universität Hannover, Germany
- Sandra Hirche, TUM, Germany
- Kazuhiro Kosuge, Tohoku University, Japan
- Michael Mistry, University of Birmingham, United Kingdom
- Jae-Bok Song, Korea University, Korea
The complete program is available here.
Jožef Stefan Institute
Jamova cesta 39
1000 Ljubljana, Slovenia
andrej.gams [at] ijs.si
828, Boulevard des Maréchaux
91762 Palaiseau Cedex,
freek.stulp [at] ensta-paristech.fr
Idiap Research Institute
Centre du Parc, Rue Marconi 19
sylvain.calinon [at] idiap.ch
andrej.gams [at] ijs.si
This workshop is supported by the IEEE RAS Technical Committee on Human-Robot Interaction & Coordination and by the IEEE RAS Technical Committee on Robot Learning.