Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Learning and Online Replication of Grasp Forces from Electromyography Signals for Prosthetic Finger Control

Robin Arbaud, Elisa Motta, Marco Domenico Avaro, Stefano Picinich, Marta Lorenzini, Arash Ajoudani

TL;DR

This work demonstrates a force-controlled prosthetic finger driven by EMG signals, implemented on a wrist-mounted device with a tendon-driven, underactuated finger and load-cell feedback. A subject-specific Convolutional LSTM model robustly maps EMG RMS features to fingertip force, enabling online modulation of grip via an admittance controller that translates force commands into tendon tension. The system achieves accurate force tracking and pattern following in online tests with four users, and exhibits strong generalization across ten subjects in offline force estimation. Overall, EMG-based force estimation coupled with a compliant control scheme offers a viable, intuitive pathway to enhance dexterity for partial-hand prostheses, with practical implications for daily tasks and quality of life.

Abstract

Partial hand amputations significantly affect the physical and psychosocial well-being of individuals, yet intuitive control of externally powered prostheses remains an open challenge. To address this gap, we developed a force-controlled prosthetic finger activated by electromyography (EMG) signals. The prototype, constructed around a wrist brace, functions as a supernumerary finger placed near the index, allowing for early-stage evaluation on unimpaired subjects. A neural network-based model was then implemented to estimate fingertip forces from EMG inputs, allowing for online adjustment of the prosthetic finger grip strength. The force estimation model was validated through experiments with ten participants, demonstrating its effectiveness in predicting forces. Additionally, online trials with four users wearing the prosthesis exhibited precise control over the device. Our findings highlight the potential of using EMG-based force estimation to enhance the functionality of prosthetic fingers.

Learning and Online Replication of Grasp Forces from Electromyography Signals for Prosthetic Finger Control

TL;DR

This work demonstrates a force-controlled prosthetic finger driven by EMG signals, implemented on a wrist-mounted device with a tendon-driven, underactuated finger and load-cell feedback. A subject-specific Convolutional LSTM model robustly maps EMG RMS features to fingertip force, enabling online modulation of grip via an admittance controller that translates force commands into tendon tension. The system achieves accurate force tracking and pattern following in online tests with four users, and exhibits strong generalization across ten subjects in offline force estimation. Overall, EMG-based force estimation coupled with a compliant control scheme offers a viable, intuitive pathway to enhance dexterity for partial-hand prostheses, with practical implications for daily tasks and quality of life.

Abstract

Partial hand amputations significantly affect the physical and psychosocial well-being of individuals, yet intuitive control of externally powered prostheses remains an open challenge. To address this gap, we developed a force-controlled prosthetic finger activated by electromyography (EMG) signals. The prototype, constructed around a wrist brace, functions as a supernumerary finger placed near the index, allowing for early-stage evaluation on unimpaired subjects. A neural network-based model was then implemented to estimate fingertip forces from EMG inputs, allowing for online adjustment of the prosthetic finger grip strength. The force estimation model was validated through experiments with ten participants, demonstrating its effectiveness in predicting forces. Additionally, online trials with four users wearing the prosthesis exhibited precise control over the device. Our findings highlight the potential of using EMG-based force estimation to enhance the functionality of prosthetic fingers.
Paper Structure (18 sections, 2 equations, 6 figures, 1 table)

This paper contains 18 sections, 2 equations, 6 figures, 1 table.

Table of Contents

  1. INTRODUCTION
  2. FINGERTIP FORCE ESTIMATION
  3. PROSTHETIC SYSTEM
  4. Hardware
  5. Embedded Software
  6. ROS 2 Interface
  7. EXPERIMENTS
  8. Signal Processing
  9. Experimental setup and protocol
  10. Force Estimation Experiment
  11. Prosthesis Control Experiment
  12. Experimental Analysis
  13. Force Estimation Experiment
  14. Prosthesis Control Experiment
  15. RESULTS AND DISCUSSION
  16. ...and 3 more sections

Figures (6)

  • Figure 1: Overview of the main components of the control and actuation scheme of the proposed finger prosthesis
  • Figure 2: Global schematics of the signal processing and control of the proposed prostetic finger
  • Figure 3: Setup and sensors employed in the experiments
  • Figure 4: Statistical performance of each force estimation method, on training and test datasets, across all ten subjects
  • Figure 5: RMS error statistics across all four participants during the prosthesis control experiments, between the command and the output (force following), between the pattern and the command (force targeting), and between the pattern and the output (force reaching)
  • ...and 1 more figures