Abstract:Introduction: Human gait motion intention recognition is very important for the lower extremity exoskeleton robot to accurately synchronize and respond to the user's natural motion. And motion intention recognition is generally performed through sEMG. Deep learning neural networks perform well in dealing with high-dimensional data and nonlinear relationships such as sEMG, but different deep learning neural networks have their own advantages in dealing with different types of data. Therefore, a multi-branch deep learning neural network, which enables different neural networks to process different feature items, could achieve more accurate and efficient motion intention recognition. The purpose of this study is to 1) Establish a multi-branch deep learning neural network model to achieve accurate gait recognition and effective estimation of joint angles. 2) Quantify the performance of the multi-branch deep learning neural network model in gait recognition and joint angle prediction using sEMG. Methodology: This study involved the collection of sEMG and plantar pressure data during walking in human subjects. Firstly, the collected signals are filtered and denoised to ensure the quality and reliability of the data. Calculate the time domain features and the frequency domain features to capture the key information of gait. Then, using the sensitivity difference of different structural neural networks to different feature data, a multi-branch deep learning neural network model is developed, in which the extracted features are used as the input of the model. The output of the model includes gait cycle and joint angle, so as to realize the accurate recognition of human gait and the effective estimation of joint angle. Results: The results show that the proposed method has high accuracy in identifying human gait and estimating joint angles. The multi-branch neural network model successfully integrates time-domain and frequency-domain features and provides reliable prediction of gait cycle and joint angle. The highest accuracy of gait recognition is 95.42%, the lowest is 90.11%, and the average is 92.16%. The average error of joint angle estimation is 3.19. Discussion: This study designed a human walking gait recognition and joint angle prediction model to achieve accurate human lower limb motion intention recognition.The model can be integrated into the sEMG sensor to design a angular biosensors, which can predict the human joint angle in real time.

A neural network-based model for lower limb continuous estimation against the disturbance of uncertainty

A Multi-Joint Continuous Motion Estimation Method of Lower Limb Using Least Squares Support Vector Machine and Zeroing Neural Network based on sEMG signals

Recurrent Neural Network Enabled Continuous Motion Estimation of Lower Limb Joints From Incomplete sEMG Signals

Continuous Prediction of Lower-Limb Kinematics From Multi-Modal Biomedical Signals

Prediction of Lower Limb Action Intention Based on Surface EMG Signal

A noise-suppressing neural network approach for upper limb human-machine interactive control based on sEMG signals

Continuous Prediction of Wrist Joint Kinematics Using Surface Electromyography from the Perspective of Muscle Anatomy and Muscle Synergy Feature Extraction

Motion estimation of elbow joint from sEMG using continuous wavelet transform and back propagation neural networks

Continuous Estimation of Human Joint Angles From sEMG Using a Multi-Feature Temporal Convolutional Attention-Based Network

Multi-branch deep learning neural network prediction model for the development of angular biosensors based on sEMG

Multi-layer CNN-LSTM network with self-attention mechanism for robust estimation of nonlinear uncertain systems

Gait Intention Prediction Using a Lower-Limb Musculoskeletal Model and Long Short-Term Memory Neural Networks

Intelligent Prediction of Human Lower Extremity Joint Moment: An Artificial Neural Network Approach

Prediction of Contralateral Lower-Limb Joint Angles Using Vibroarthrography and Surface Electromyography Signals in Time-Series Network

Recognizing Continuous Multiple Degrees of Freedom Foot Movements with Inertial Sensors.

A Bi-Directional LSTM Network for Estimating Continuous Upper Limb Movement From Surface Electromyography

Neural Dynamics for Computing Perturbed Nonlinear Equations Applied to ACP-Based Lower Limb Motion Intention Recognition

Gait Cycle-Inspired Learning Strategy for Continuous Prediction of Knee Joint Trajectory from sEMG

An artificial neural network framework for lower limb motion signal estimation with foot-mounted inertial sensors

Knee Joint Angle Prediction Based on Muscle Synergy Theory and Generalized Regression Neural Network

Motion intention prediction of upper limb in stroke survivors using sEMG signal and attention mechanism