Loughborough University
Towards a learning framework for dancing robots.pdf (168.59 kB)

Towards a learning framework for dancing robots

Download (168.59 kB)
conference contribution
posted on 2012-07-25, 12:57 authored by Ibrahim S. Tholley, Qinggang MengQinggang Meng, Paul Chung
How can we make robots learn how to dance? How do humans learn to dance? An emerging culture of dancing robots is becoming more prominent in the research community with more emphasis on how we can show of our own creativity rather than allowing the robots to develop their own cognitive and psychological behaviours to the music being played. There are many different types of music and indeed, many different robots and many ways, in which they can dance to music however, much of the work carried out in this field concern limiting robots to dance in particular ways to a specific music and no adaptive behaviour implemented in them to be able to respond intuitively to music in general. We propose in this paper, a way in which such a problem can begin to be looked into, by introducing fundamental things that should be learnt that are necessary for dancing. We programmed a virtual robot to learn to dance to the beat as well as recognise the downbeat of any time-signature and tailor its movements to the loudness of music, using the Sarsa and the Sarsa(lambda) algorithms from reinforcement learning as the learning framework. Experimental results show that it is possible to make robots learn to dance to these fundamental rhythmic features of music.



  • Science


  • Computer Science


THOLLEY, I.S., MENG, Q. and CHUNG, P.W.H., 2009. Towards a learning framework for dancing robots. IN: Proceedings of the IEEE International Conference on Control and Automation, Christchurch, New Zealand, Vols 1-3, pp. 1581 - 1586




  • AM (Accepted Manuscript)

Publication date



This conference paper was presented at the 2009 IEEE International Conference on Control and Automation Christchurch, New Zealand, December 9-11, 2009 [© IEEE]. It is also available at: http://ieeexplore.ieee.org/. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.




  • en