Journal of Bionic Engineering ›› 2024, Vol. 21 ›› Issue (1): 270-287.doi: 10.1007/s42235-023-00453-8

• • 上一篇    下一篇

Robust Machine Learning Mapping of sEMG Signals to Future Actuator Commands in Biomechatronic Devices

Ali Nasr1; Sydney Bell1; Rachel L. Whittaker2; Clark R. Dickerson2; John McPhee1   

  1. 1 Systems Design Engineering, University of Waterloo, Waterloo N2L 3G1, Canada  2 Kinesiology and Health Sciences, University of Waterloo, Waterloo N2L 3G3, Canada
  • 出版日期:2024-01-16 发布日期:2024-02-26
  • 通讯作者: Ali Nasr E-mail:a.nasr@uwaterloo.ca
  • 作者简介:Ali Nasr1; Sydney Bell1; Rachel L. Whittaker2; Clark R. Dickerson2; John McPhee1

Robust Machine Learning Mapping of sEMG Signals to Future Actuator Commands in Biomechatronic Devices

Ali Nasr1; Sydney Bell1; Rachel L. Whittaker2; Clark R. Dickerson2; John McPhee1   

  1. 1 Systems Design Engineering, University of Waterloo, Waterloo N2L 3G1, Canada  2 Kinesiology and Health Sciences, University of Waterloo, Waterloo N2L 3G3, Canada
  • Online:2024-01-16 Published:2024-02-26
  • Contact: Ali Nasr E-mail:a.nasr@uwaterloo.ca
  • About author:Ali Nasr1; Sydney Bell1; Rachel L. Whittaker2; Clark R. Dickerson2; John McPhee1

摘要: A machine learning model for regression of interrupted Surface Electromyography (sEMG) signals to future control-oriented signals (e.g., robot’s joint angle and assistive torque) of an active biomechatronic device for high-level myoelectric-based hierarchical control is proposed. A Recurrent Neural Network (RNN) was trained using output data, initially obtained from ofine optimization of the biomechatronic (human–robot) device and shifted by the prediction horizon. The input of the RNN consisted of interrupted sEMG signals (to mimic signal disconnections) and previous kinematic signals of the assistive system. The RNN with a 0.1-s prediction horizon could predict the control-oriented joint angle and assistive torque with 92% and 86.5% regression accuracy, respectively, for the test dataset. This proposed approach permits a fast, predictive, and direct estimation of control-oriented signals instead of an iterative process that optimizes assistive torque in the inverse dynamic simulation of a multibody human–robot system. Training with these interrupted input signals signifcantly improves the regression accuracy in the case of sEMG signal disconnection. This Robust Predictive Control-oriented Machine Learning (Robust-MuscleNET) model can support volitional high-level myoelectric-based control of biomechatronic devices, such as exoskeletons, prostheses, and assistive/resistive robots. Future work should study the application to prosthesis control as well as the repeatability of the high-level controller with electrode shift. The low-level hierarchical controller that manages the human–robot interaction, the assistance/resistance strategy, and the actuator coordination should also be studied.

关键词: Myoelectric-based control , · Surface electromyography , · Machine learning , · Multibody system dynamics , · Exoskeleton , · Bionic

Abstract: A machine learning model for regression of interrupted Surface Electromyography (sEMG) signals to future control-oriented signals (e.g., robot’s joint angle and assistive torque) of an active biomechatronic device for high-level myoelectric-based hierarchical control is proposed. A Recurrent Neural Network (RNN) was trained using output data, initially obtained from ofine optimization of the biomechatronic (human–robot) device and shifted by the prediction horizon. The input of the RNN consisted of interrupted sEMG signals (to mimic signal disconnections) and previous kinematic signals of the assistive system. The RNN with a 0.1-s prediction horizon could predict the control-oriented joint angle and assistive torque with 92% and 86.5% regression accuracy, respectively, for the test dataset. This proposed approach permits a fast, predictive, and direct estimation of control-oriented signals instead of an iterative process that optimizes assistive torque in the inverse dynamic simulation of a multibody human–robot system. Training with these interrupted input signals signifcantly improves the regression accuracy in the case of sEMG signal disconnection. This Robust Predictive Control-oriented Machine Learning (Robust-MuscleNET) model can support volitional high-level myoelectric-based control of biomechatronic devices, such as exoskeletons, prostheses, and assistive/resistive robots. Future work should study the application to prosthesis control as well as the repeatability of the high-level controller with electrode shift. The low-level hierarchical controller that manages the human–robot interaction, the assistance/resistance strategy, and the actuator coordination should also be studied.

Key words: Myoelectric-based control , · Surface electromyography , · Machine learning , · Multibody system dynamics , · Exoskeleton , · Bionic