Model predictive controller-based spatiotemporal path tracking method for transhumeral prostheses

Area of Science:

• Biomedical Engineering
• Robotics
• Human-Machine Interaction

Background:

• Transhumeral amputees utilize prostheses to replace the upper limb segment between the shoulder and elbow.
• Effective prostheses should replicate natural limb function for optimal user benefit.
• Natural upper limb motions, like reaching and pointing, follow a straight-line path with a specific velocity profile.

Purpose of the Study:

• To develop a dynamic path-tracking method for transhumeral prostheses.
• To enable prosthetic limbs to adhere to a straight-line path during movement.

Main Methods:

• Utilized a model predictive controller (MPC) based on the prosthesis's kinematic model.
• Implemented a shoulder matcher to align actual and predicted shoulder poses for optimal joint angle selection.
• Enabled dynamic path adjustments in response to unplanned human shoulder motions.

Main Results:

• Experimental validation confirmed the proposed method's effectiveness.
• The transhumeral prosthesis demonstrated the ability to follow a straight-line path, mimicking natural human movement.

Conclusions:

• A dynamic path-tracking method using MPC was successfully developed for transhumeral prostheses.
• The prosthetic hand can achieve a straight-line trajectory comparable to a natural human hand.