Journal of Bionic Engineering ›› 2024, Vol. 21 ›› Issue (1): 256-269.doi: 10.1007/s42235-023-00448-5

• • 上一篇    下一篇

STGNN-LMR: A Spatial–Temporal Graph Neural Network Approach Based on sEMG Lower Limb Motion Recognition

Weifan Mao1,2; Bin Ma1,2; Zhao Li3; Jianxing Zhang1; Yizhou Lu1,2; Zhuting Yu4; Feng Zhang5   

  1. 1 Shanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, China  2 University of Chinese Academy of Sciences, Beijing 101408, China  3 University of Science and Technology of China, Hefei 230026, China  4 Shanghai Zhongyan Jiuyi Technology Co., Ltd., Shanghai 201821, China 5 Academy of Military Science, Beijing 100091, China
  • 出版日期:2024-01-16 发布日期:2024-02-26
  • 通讯作者: Bin Ma E-mail:mab@sari.ac.cn
  • 作者简介:Weifan Mao1,2; Bin Ma1,2; Zhao Li3; Jianxing Zhang1; Yizhou Lu1,2; Zhuting Yu4; Feng Zhang5

STGNN-LMR: A Spatial–Temporal Graph Neural Network Approach Based on sEMG Lower Limb Motion Recognition

Weifan Mao1,2; Bin Ma1,2; Zhao Li3; Jianxing Zhang1; Yizhou Lu1,2; Zhuting Yu4; Feng Zhang5   

  1. 1 Shanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, China  2 University of Chinese Academy of Sciences, Beijing 101408, China  3 University of Science and Technology of China, Hefei 230026, China  4 Shanghai Zhongyan Jiuyi Technology Co., Ltd., Shanghai 201821, China 5 Academy of Military Science, Beijing 100091, China
  • Online:2024-01-16 Published:2024-02-26
  • Contact: Bin Ma E-mail:mab@sari.ac.cn
  • About author:Weifan Mao1,2; Bin Ma1,2; Zhao Li3; Jianxing Zhang1; Yizhou Lu1,2; Zhuting Yu4; Feng Zhang5

摘要: Lower limb motion recognition techniques commonly employ Surface Electromyographic Signal (sEMG) as input and apply a machine learning classifer or Back Propagation Neural Network (BPNN) for classifcation. However, this artifcial feature engineering technique is not generalizable to similar tasks and is heavily reliant on the researcher’s subject expertise. In contrast, neural networks such as Convolutional Neural Network (CNN) and Long Short-term Memory Neural Network (LSTM) can automatically extract features, providing a more generalized and adaptable approach to lower limb motion recognition. Although this approach overcomes the limitations of human feature engineering, it may ignore the potential correlation among the sEMG channels. This paper proposes a spatial–temporal graph neural network model, STGNN-LMR, designed to address the problem of recognizing lower limb motion from multi-channel sEMG. STGNN-LMR transforms multi-channel sEMG into a graph structure and uses graph learning to model spatial–temporal features. An 8-channel sEMG dataset is constructed for the experimental stage, and the results show that the STGNN-LMR model achieves a recognition accuracy of 99.71%. Moreover, this paper simulates two unexpected scenarios, including sEMG sensors afected by sweat noise and sudden failure, and evaluates the testing results using hypothesis testing. According to the experimental results, the STGNN-LMR model exhibits a signifcant advantage over the control models in noise scenarios and failure scenarios. These experimental results confrm the efectiveness of the STGNN-LMR model for addressing the challenges associated with sEMG-based lower limb motion recognition in practical scenarios.

关键词: Lower limb motion recognition , · Exoskeleton , · sEMG , · Graph neural network , · Noise , · Sensor failure

Abstract: Lower limb motion recognition techniques commonly employ Surface Electromyographic Signal (sEMG) as input and apply a machine learning classifer or Back Propagation Neural Network (BPNN) for classifcation. However, this artifcial feature engineering technique is not generalizable to similar tasks and is heavily reliant on the researcher’s subject expertise. In contrast, neural networks such as Convolutional Neural Network (CNN) and Long Short-term Memory Neural Network (LSTM) can automatically extract features, providing a more generalized and adaptable approach to lower limb motion recognition. Although this approach overcomes the limitations of human feature engineering, it may ignore the potential correlation among the sEMG channels. This paper proposes a spatial–temporal graph neural network model, STGNN-LMR, designed to address the problem of recognizing lower limb motion from multi-channel sEMG. STGNN-LMR transforms multi-channel sEMG into a graph structure and uses graph learning to model spatial–temporal features. An 8-channel sEMG dataset is constructed for the experimental stage, and the results show that the STGNN-LMR model achieves a recognition accuracy of 99.71%. Moreover, this paper simulates two unexpected scenarios, including sEMG sensors afected by sweat noise and sudden failure, and evaluates the testing results using hypothesis testing. According to the experimental results, the STGNN-LMR model exhibits a signifcant advantage over the control models in noise scenarios and failure scenarios. These experimental results confrm the efectiveness of the STGNN-LMR model for addressing the challenges associated with sEMG-based lower limb motion recognition in practical scenarios.

Key words: Lower limb motion recognition , · Exoskeleton , · sEMG , · Graph neural network , · Noise , · Sensor failure