Hybrid learning control for improving suppression of hand tremor

Area of Science:

• Biomedical Engineering
• Control Systems Engineering
• Robotics

Background:

• Hand tremors significantly impair daily activities like writing and object manipulation.
• Existing control strategies for tremor attenuation have limitations.

Purpose of the Study:

• To present an online hybrid proportional-integral control with active force control (PI+AFC) strategy for tremor attenuation.
• To incorporate iterative learning control (ILC) for enhanced active force control performance.

Main Methods:

• Developed an online hybrid PI+AFC system with ILC for tremor suppression.
• Utilized a linear voice coil actuator as the active tremor suppression element.
• Tested the controller on a dummy hand model in a tremor test rig simulating postural tremors.

Main Results:

• The proposed intelligent active force control with ILC demonstrated superior tremor error reduction compared to traditional controllers.
• Sensitivity analysis confirmed the controller's robustness in real-time applications.
• Significant reduction in tremor severity was achieved.

Conclusions:

• The intelligent active force control and iterative learning controller offers an effective solution for hand tremor attenuation.
• This approach shows potential for restoring functional capabilities in patients with hand tremors.
• The study highlights the efficacy of advanced control strategies in mitigating neurological movement disorders.