A low-cost instrumented glove for monitoring forces during object manipulation.

A rehabilitation program toward restoring upper limb movements based on neuromuscular electrical stimulation (NMES) depends on closed-loop control performance, which has been limited by the development of sensors for practical daily use. This work proposes a system to obtain force feedback. The system is comprised of a Lycra commercial glove with force sensing resistors (FSR's) attached to the distal phalanxes of the thumb, index and long fingers. After amplification and filtering, the signal is digitized through an analog-to-digital (A/D) converter. The polynomial fitting coefficients for the characteristic curves, obtained during the sensor calibration process, were inserted in the software thus enabling the reading of forces exerted during object manipulation. The system was applied to 30 normal subjects in order to verify its feasibility and to acquire knowledge of the normal hand function. Different ways of grasping have been detected according to the Force versus Time curve pattern and to the fingers predominantly used in grasping. Results have also shown the influence of parameters such as gender, age, hand size, and object weight in the normal function. The system did show efficacy. It was able to determine grasp forces during object manipulation for up to 73% of the studied sample. This is significant since a single glove was used in a wide range of subjects. For best results in medical applications, the glove should be tailored to the particular characteristics of an individual user.