A Pilot Study on Motion Intention Mapping and Direct Myoelectric Control Method for Prosthetic Knee Based on LSTM...

Area of Science:

• Biomedical Engineering
• Robotics
• Neuroscience

Background:

• Prosthetic knees require advanced control for adaptability and natural movement.
• Current myoelectric control methods face challenges in real-time coordination and responsiveness.

Purpose of the Study:

• To develop a direct myoelectric control method for intelligent prosthetic knees.
• To enhance human-machine coordination and adaptability using an LSTM network and coupling model.

Main Methods:

• Collected multichannel surface electromyography (sEMG) and knee joint angle data during walking.
• Extracted time-domain features to build an LSTM prediction model for muscle activity-joint kinematics mapping.
• Integrated a human-machine coupling dynamics model with a hydraulic actuation system for a prosthetic knee control framework.

Main Results:

• The LSTM model demonstrated superior prediction accuracy and temporal consistency compared to traditional neural networks.
• Achieved continuous damping adjustment and smooth gait transitions in a variable-damping prosthetic knee.
• Validated the effectiveness of the direct myoelectric control for human-machine coordinated prosthetic limb function.

Conclusions:

• The proposed LSTM-based direct myoelectric control method significantly enhances prosthetic knee adaptability and human-machine coordination.
• This approach offers a feasible and effective solution for intelligent prosthetic limb control, improving user mobility.