Learning to bounce: First lessons from a bouncing robot

Lungarella, Max and Berthouze, Luc (2003) Learning to bounce: First lessons from a bouncing robot. In: 2nd International Symposium on Adaptive Motion of Animals and Machines, Kyoto, Japan.

Full text not available from this repository.


The study of how infants learn to bounce, while being supported by a harness attached to a spring, sheds light on how infants learn to exploit the dynamics of their exploratory motion. The emerging rhythmical activity result of an entrainment among neural system, musculo-skeletal system, and surrounding environment is a salient characteristic of a developing body control during the rst year of life. In this paper, we describe and discuss the results of four preliminary experiments realized with a small-sized humanoid robot harnessed in a jolly jumper, and whose leg joints are controlled by neural oscillators. While the two rst experiments see the robot oscillate freely in space, the last experiments have the robot touch the ground during oscillations so as to characterise the effects of ground interaction. An appropriate choice of the parameters of the neural oscillators lead to sustained and stable bouncing.

Item Type: Conference or Workshop Item (Paper)
Schools and Departments: School of Engineering and Informatics > Informatics
Depositing User: Luc Berthouze
Date Deposited: 06 Feb 2012 19:26
Last Modified: 12 Apr 2012 14:40
URI: http://sro.sussex.ac.uk/id/eprint/20551
📧 Request an update