Accession Number : ADA299421

Title :   Embodiment and Manipulation Learning Process for a Humanoid Hand.

Descriptive Note : Master's thesis,

Corporate Author : MASSACHUSETTS INST OF TECH CAMBRIDGE ARTIFICIAL INTELLIGENCE LAB

Personal Author(s) : Matsuoka, Yoky

PDF Url : ADA299421

Report Date : MAY 1995

Pagination or Media Count : 90

Abstract : 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.

Descriptors :   *NEURAL NETS, *ROBOTS, *LEARNING MACHINES, ALGORITHMS, SIGNAL PROCESSING, COMPUTERIZED SIMULATION, OPTIMIZATION, ADAPTIVE CONTROL SYSTEMS, COMPUTATIONS, COGNITION, THESES, VOLTAGE, FEEDBACK, PATTERN RECOGNITION, ARTIFICIAL INTELLIGENCE, SYSTEMS ANALYSIS, CONTROL THEORY, MANIPULATORS, CENTRAL NERVOUS SYSTEM, HANDS, REFLEXES, POTENTIOMETERS, CONDITIONING(LEARNING), FINGERS.

Subject Categories : Cybernetics
      Bionics

Distribution Statement : APPROVED FOR PUBLIC RELEASE