An Adaptive Iterative Learning Based Impedance Control for Robot-Aided Upper-Limb Passive Rehabilitation

Area of Science:

• Robotics
• Biomedical Engineering
• Control Systems

Background:

• Upper-limb rehabilitation is crucial for patients with various diseases.
• Existing rehabilitation devices may lack adaptivity and precision.
• Pneumatic artificial muscles offer a promising approach for soft robotic actuation.

Purpose of the Study:

• To introduce an anthropomorphic arm for passive upper-limb rehabilitation therapy.
• To develop a combined control strategy for enhanced system stability and adaptivity.
• To enable accurate position tracking and appropriate contact force control during therapy.

Main Methods:

• Construction of an anthropomorphic arm using pneumatic artificial muscles.
• Implementation of a combined control strategy integrating iterative learning control (ILC) and adaptive ILC-based impedance control.
• Analysis of control convergence and validation through numerical simulations and real-world experiments.

Main Results:

• The proposed combined control effectively stabilizes the system and enhances adaptivity.
• Accurate position tracking is achieved using iterative learning control.
• Appropriate contact forces are executed via adaptive iterative learning-based impedance control.
• The control strategy demonstrates robustness against model uncertainties and system nonlinearities.

Conclusions:

• The developed anthropomorphic arm and combined control strategy are effective for upper-limb passive rehabilitation.
• The approach offers a model-independent solution for nonlinear and redundant robotic systems.
• The findings support the use of this technology for improving therapeutic exercises.