Embodiment and Manipulation Learning Process for a Humanoid Hand

Title: en_US Embodiment and Manipulation Learning Process for a Humanoid Hand
Creator: en_US Matsuoka, Yoky
Date: 2004-10-20T20:27:54Z
Date Available: 2004-10-20T20:27:54Z
Date Issued: en_US 1995-05-01
Identifier: en_US AITR-1546
uri: http://hdl.handle.net/1721.1/7064
Abstract: en_US Babies are born with simple manipulation capabilities such as reflexes to perceived stimuli. Initial discoveries by babies are accidental until they become coordinated and curious enough to actively investigate their surroundings. This thesis explores the development of such primitive learning systems using an embodied light-weight hand with three fingers and a thumb. It is self-contained having four motors and 36 exteroceptor and proprioceptor sensors controlled by an on-palm microcontroller. Primitive manipulation is learned from sensory inputs using competitive learning, back-propagation algorithm and reinforcement learning strategies. This hand will be used for a humanoid being developed at the MIT Artificial Intelligence Laboratory.
Extent: 9161027 bytes; 7404933 bytes
Format: application/postscript; application/pdf
Language: en_US
Relation: en_US AITR-1546