Data-Mined Continuous Hip-Knee Coordination Mapping With Motion Lag for Lower-Limb Prosthesis Control.

Area of Science:

• Biomechanics and Robotics
• Prosthetics and Orthotics

Background:

• Lower limb prosthesis control relies heavily on knee joint trajectory planning.
• Current methods for mapping healthy limb motion to prosthetic joints are complex and lack intuition.
• Developing simpler, robust coordination mappings is crucial for advanced prosthetic control.

Purpose of the Study:

• To explore and establish a simple, intuitive coordination mapping between healthy hip and prosthetic knee motion.
• To introduce a novel Motion-Lagged Coordination Mapping (MLCM) for prosthetic knee trajectory generation.
• To investigate the relationship between hip-knee coordination indexes and motion lag.

Main Methods:

• Experimental data mining was used to analyze coordination between hip and knee joints.
• Key coordination indexes, mean absolute relative phase (MARP) and deviation phase (DP), were calculated.
• A polynomial model was constructed for the hip-knee Motion-Lagged Coordination Mapping (MLCM), incorporating a time lag.

Main Results:

• A stable phase difference between hip and knee motion was identified across subjects.
• The MLCM demonstrated higher efficiency compared to Gaussian process regression and neural network learning.
• A significant linear correlation between hip-knee MARP and motion lag was discovered.

Conclusions:

• The MLCM enables prosthetic knee trajectory generation using only healthy hip motion, reducing sensing requirements and enhancing robustness.
• This approach facilitates more natural prosthetic gait at various walking speeds.
• The MLCM offers a promising, efficient, and intuitive method for lower limb prosthesis control.