This sensory feedback system would help amputees execute fine tasks or handle delicate objects easily.

A new type of sensory feedback system for a myoelectrically controlled biomimetic prosthetic hand was developed. One of the characteristic features of the human neuromuscular control system is the increase in the compliance around a joint with decreasing activity of a muscle. Interferential current that uses two waves and interferences was used to synthesize a low frequency in the body. One frequency was fixed at 4,000 Hz and the other frequency was changed from 3,700 Hz to 4,000 Hz. As a result, stimulus frequency in the body was changed from 0 Hz to 300 Hz. It was measured that the subjects could distinguish the change of stimulus frequency, and confirmed that the interference current is a useful method for sensibility feedback of a prosthetic hand.

In developing a prosthetic hand, three points are important. The first is that its shape should be similar to a human hand. The second is that an amputee must be able to voluntarily control the opening and closing of the hand. The third is the ability to transmit the status of the hand to the amputee.

If a prosthetic hand has the same mechanisms and mechanical properties as the neuromuscular control system of the human hand, an amputee may be able to utilize almost the same subconscious control as used before amputation. Consequently, training periods required for the amputee to operate such a prosthetic hand would be much shorter than that for a conventional myoelectric hand. If the prosthetic hand has the ability to transmit the status of the hand, such as finger angle or grasping force, to the amputee, the amputee will be able to execute fine tasks or handle breakable objects more easily with this hand than with conventional myoelectric prosthetic hands.

Four stainless steel electrodes were fixed on the forearm with electrode paste. In the experiment, the channel-1 frequency was fixed for 4,000 Hz and the channnel-2 frequency was changed as follows: 1) How an interference wave was felt when both frequencies were at 4,000 Hz was measured; 2) The frequency at which the subject felt a change each time the channel-2 frequency gradually decreased from 3,999 Hz to 3,700 Hz was measured; and 3) The frequency at which the subject felt a change each time when the channel-2 frequency gradually increased from 3,700 Hz to 3,999 Hz was measured. Each experiment repeated ten times for each of eight healthy young volunteers.

The result shows that the subjects could feel the frequency change precisely in cases of frequency decrease. This means that more information will be needed when the prosthetic hand grasps an object as compared with releasing an object. Therefore, the change of channnel-2 frequency should decrease when the prosthetic hand grasps the object and increase when the prosthetic hand releases the object.

This work was done by M. Yoshida of the Department of Biomedical Engineering and and Y. Sasaki of the Graduate School of Engineering at Osaka Electro-Communication University, Japan, for the Army Research Laboratory. ARL-0068