We developed a system that enables a humanoid robot to imitate whole-body motions of humans in real time. To avoid falls of the robot that might occur when using direct imitation of the movements due to the different weight distribution, we developed an approach that actively balances the center of mass over the support polygon. Our system captures human motions with the Xsens MVN motion capture system. The video illustrates that the Nao is able to reliably imitate complex whole-body motions, which also include extended periods of time in single support mode.
- Real-time Imitation of Human Whole-Body Motions by Humanoids.
J. Koenemann, Felix Burget, and M. Bennewitz.
In: Proceedings of the IEEE International Conference on Robotics & Automation (ICRA), 2014.
- Whole-Body Imitation of Human Motions with a Nao Humanoid.
J. Koenemann and M. Bennewitz.
In: Video Abstract Proceedings of the ACM/IEEE International Conference on Human-Robot Interaction (HRI), 2012, second place best video award.
- Gait Optimization of Humanoids Based on Human Data.
S. Wehner and M. Bennewitz.
In: AUTOMATIKA - Journal for Control, Measurement, Electronics, Computing and Communications, 2011.
The videos below show our Nao imitating complex human motions. We capture human data with a Xsens MVN mocap suit, but in general any system providing joint angle information can be used.
Direct link to YouTube