Force myography controlled multifunctional hand prosthesis for upper-limb amputees
[Display omitted] •FMG sensor was designed to extract the muscle contraction information from amputee’s forearm.•Captured FMG signals were used to drive multifunctional custom-made transradial prosthesis.•Hand prototype 1 coupled with proportional control provides intuitive operation for grasping.•H...
Gespeichert in:
Veröffentlicht in: | Biomedical signal processing and control 2020-09, Vol.62, p.102122, Article 102122 |
---|---|
Hauptverfasser: | , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | [Display omitted]
•FMG sensor was designed to extract the muscle contraction information from amputee’s forearm.•Captured FMG signals were used to drive multifunctional custom-made transradial prosthesis.•Hand prototype 1 coupled with proportional control provides intuitive operation for grasping.•Hand prototype 2 with fuzzy classification scheme yields six grip patterns for dexterous grasping.•Hand prototype 1 presented an affordable, practical, and faster version of the prosthetic hand.
Existing myoelectric prostheses can provide solutions to amputees for performing activities of daily livings. However, there are several issues with these prostheses (1) performance are affected mainly by sweat, electrode shift, motion artifact, interference, etc. (2) are expensive (3) have complex control schemes.
This paper presents an affordable hand prosthesis solution for transradial amputees, which uses force myography (FMG) as a control signal in place of electromyography (EMG) signal. A novel FMG sensor was designed to detect mechanical muscle contractions from the residual forearm of amputees. Two distinct control strategies were formulated for translating the FMG signals from the sensor to the control commands for driving custom-made prosthetic hands.
Prosthetic hand prototype 1 with a proportional control scheme was developed and verified on five amputees. Subjects wearing the hand were able to grasp various objects using their intent of muscular contractions.
Fuzzy logic based classification scheme with an offline success rate above 97 % was established for identifying six different hand gestures from only single-channel FMG data. Further, the system, along with the threshold control, was implemented for the developed hand prototype 2 to obtain six distinct grip patterns. The real-time testing of hand on thirteen subjects (five amputees and eight intact) showed a success percentage above 95 % in performing dexterous grasping operations.
As compared to the presently available hands, the hand prototype 2 displayed comparable features at such a low-cost price. However, with limited functionality, hand prototype 1 is an affordable, practical, intuitive, and faster version of the prosthetic hand. |
---|---|
ISSN: | 1746-8094 1746-8108 |
DOI: | 10.1016/j.bspc.2020.102122 |