Recent advances in robot learning, contributed by the organizers of the workshop:


Humanoid robot learns to clean a whiteboard

The humanoid robot Fujitsu HOAP-2 learns to clean a whiteboard by upper-body kinesthetic teaching. This video was presented at ICRA 2011 in Shanghai, China.


Robot learning the skill of archery

The humanoid robot iCub learns the skill of archery. This video was presented at Humanoids 2010 conference in Nashville, USA.


Robot learning to walk efficiently with passive compliance

The compliant humanoid robot COMAN learns to minimize the energy consumption during walking using reinforcement learning. This video was presented at IROS 2011 in San Francisco, USA.


Robot learns to place new objects

The ability to place objects in an environment is an important skill for a personal robot. In this video, a supervised learning algorithm is used for finding good placements given the point-clouds of the object and the placing area. This video was presented at RSS 2011 workshop on mobile manipulation.


Scene Understanding for Personal Robots

Using Kinect sensors, robots can easily obtain colored 3D pointclouds of their environment. The following video focuses on the high-level scene understanding based on the 3D data.