Design of a high precision Data glove based on inertial sensors

Abstract

This work presents a design proposal for the development of a system for human hand motion caption system, aiming to create a data glove. This prototype will be composed of a set of inertial sensors placed in the user's hand. The sensors are used for transmitting the movements of the proximal, medial phalanges of each finger in the hand. For this prototype, the flexion of the distal phalanx is not relevant, because this phalanx cannot be normally controlled by the humans. The flexion of the distal element is calculated with relation to the medial phalanx flexion. The motion captured by the sensors are transmitted through an Arduino™ board to a computer, where it is possible to render a virtual hand applying the transformations captured from the sensors. The virtual hand can mimic the configuration of the real hand, including movements of abduction, adduction, flexion and extension of the fingers. Some experimental design is carried out in order to perform a motion capture validation of the system, obtaining the precision and the resolution of the system.