Performance-Based Hybrid Control of a Cable-Driven Upper-Limb Rehabilitation Robot
IEEE Transactions on Bio-Medical Engineering
|September 30, 2020
Summary
This study introduces a hybrid controller for upper-limb rehabilitation robots, adapting to patient needs. The controller adjusts resistance, assistance, and restriction based on motor performance for effective stroke rehabilitation.
Area of Science:
- Robotics
- Rehabilitation Engineering
- Biomechanics
Background:
- Stroke survivors often require tailored rehabilitation due to varying motor impairments.
- Existing upper-limb rehabilitation robots may lack adaptive control to meet diverse patient needs.
Purpose of the Study:
- To propose and validate a performance-based hybrid controller for a cable-driven upper-limb rehabilitation robot (CDULRR).
- To enable adaptive control modes (resistance, assistance, restriction) based on patient motor performance.
Main Methods:
- Developed a hybrid controller with three modes: resistance, assistance (fuzzy logic-tuned stiffness), and restriction.
- Utilized tracking error as an index for switching between control modes.
- Conducted task-oriented experiments with ten healthy subjects under varying disturbance levels.
Main Results:
- The controller successfully switched between working modes based on subjects' motor performance.
- Increased disturbance led to decreased motor performance and consequently, increased robot assistance.
- Adaptive damping force and stiffness adjustments promoted greater active patient effort.
Conclusions:
- The proposed performance-based hybrid controller effectively adapts to user motor performance during upper-limb robot-assisted rehabilitation.
- This adaptive approach enhances training by providing appropriate levels of support and challenge.
- The CDULRR controller shows promise for personalized stroke rehabilitation.
