Jump to content

Motor babbling

From Wikipedia, the free encyclopedia

Motor babbling is a process of repeatedly performing a random motor command for a short duration. It is similar to the vocal babbling of infants, where the brain learns the relation between vocal muscle activities and the resulting sounds. However, it was found that the general motor-control system is already exploring itself in the womb, in animals, in a similar way. Originally, the random spasms and convulsions of the embryo were seen as the non-functional consequences of growth. Later it was realized that the motor system is already calibrating its sensorimotor system before birth.[1] After birth, motor babbling in primates continues in the random grasping movements towards visual targets, training the hand–eye coordination system. These insights are used since the early nineteen nineties in models of biological movement control and in robotics.[2][3] In robotics, it is a system of robot learning whereby a robotic system can autonomously develop an internal model of its self-body and its environment. Early work is by Kuperstein (1991)[4] using a robot randomly positioning a stick in its workspace, while being observed by two cameras, using a neural network (multilayer perceptron) to associate poses of the stick with joint angles of the arm. This type of research has led to the research field of developmental robotics.[5][6][7]

The random exploration of the motor-control state and its effects can lead to the identification of parameters for a inverse kinematics and inverse dynamics model, which the robot can use to find appropriate control signals, given a particular task that requires positioning and/or force control.[8][9]

See also

[edit]

References

[edit]
  1. ^ Bernstein, N.A. (1967). The Co-Ordination and Regulation of Movements. Oxford: Pergamon Press.
  2. ^ Bullock, Daniel; Grossberg, Stephen; Guenther, Frank (1993). "A Self-Organizing Neural Model of Motor Equivalent Reaching and Tool Use by a Multijoint Arm". Journal of Cognitive Neuroscience. 5 (4): 408–435. doi:10.1162/jocn.1993.5.4.408. PMID 23964916. S2CID 207610285.
  3. ^ Villa, Alessandro E. P.; Masulli, Paolo; Rivero, Antonio Javier Pons (2016). Artificial Neural Networks and Machine Learning – ICANN 2016: 25th International Conference on Artificial Neural Networks, Barcelona, Spain, September 6-9, 2016, Proceedings. Cham: Springer. p. 167. ISBN 978-3-319-44777-3.
  4. ^ Kuperstein, Michael (1991). "INFANT neural controller for adaptive sensory-motor coordination". Neural Networks. 4 (2): 131–145. doi:10.1016/0893-6080(91)90001-L. ISSN 0893-6080.
  5. ^ Saegusa, Ryo; Metta, Giorgio; Sandini, Giulio; Sakka, Sophie. "Active Motor Babbling for Sensory-Motor Learning" (PDF). {{cite journal}}: Cite journal requires |journal= (help)
  6. ^ Demiris, Yiannis, and Anthony Dearden. "From motor babbling to hierarchical learning by imitation: a robot developmental pathway." (2005): 31-37.
  7. ^ Saegusa, Ryo, et al. "Active motor babbling for sensorimotor learning." Robotics and Biomimetics, 2008. ROBIO 2008. IEEE International Conference on. IEEE, 2009.
  8. ^ Arik, Sabri; Huang, Tingwen; Lai, Weng Kin; Liu, Qingshan (2015). Neural Information Processing: 22nd International Conference, ICONIP 2015, Istanbul, Turkey, November 9–12, 2015, Proceedings. Cham: Springer. p. 27. ISBN 978-3-319-26554-4.
  9. ^ McFarland, David; Stenning, Keith; McGonigle, Maggie (2012). The Complex Mind: An Interdisciplinary Approach. New York: Palgrave Macmillan. p. 119. ISBN 978-0-230-35445-6.