Company Logo
  • fibremap_940x230.jpg
  • intro1b_940x230.jpg
  • intro2_940x230.jpg
  • intro3b_940x230.jpg
  • intro4_940x230.jpg

Human action recognition oriented to humanoid robots action reproduction

Our research aims at providing a humanoid robot with the ability of observing, learning, and reproducing actions performed by humans in order to acquire new skills. In other words, we want to apply arti cial intelligence techniques to automatically recognize a human activity in order to make a humanoid robot able to reproduce it.This system has not only to distinguish between di erent actions, but also to represent them in a proper manner to allow a robot to reproduce the motion trajectories the demonstrator showed and learn new skills. Since the nal system is going to be integrated in an autonomous humanoid robot (speci cally model Aldebran Nao), we are working with an RGB-D sensor (Microsoft Kinect) that can be easily applied to it. This objective introduces also strict real-time constrains to the action recognition algorithm: we have opted for a probabilistic approach that o ers good modeling and fast recognition performances.
Research areas:
Type of Publication:
In Proceedings
Action recognition; Humanoid Robots; Programming by Demonstration; Imitation Learning
Book title:
In Matteo Baldoni, Federico Chesani, Bernardo Magnini, Paola Mello, Marco Montai (eds.), Popularize Artificial Intelligence, proceedings of the AI*IA Workshop and Prize for Celebrating 100th Anniversary of Alan Turing's Birth (PAI. 2012), Rome, Italy, Jun
June 15
Hits: 275

Powered by Joomla!®. Valid XHTML and CSS