In recent years, the robotic devices have been used in hand rehabilitation training practice. The majority of existing robotic devices for rehabilitation belong to the rigid exoskeleton. However, rigid exoskeletons may have some limitations such as heavy weight, un-safety and inconvenience. This paper presents a device designed to help post-stroke patients to stretch their spastic hands. This hand rehabilitation device actuator is fabricated by soft material, powered with fluid pressure, and embedded in one glove surface. The distinguished features of this device are: safety, low cost, light weight, convenience and pneumatic actuation. In clinical practice, rehabilitation therapists should help the post-stroke patients to stretch fingers to a desired joint position. Therefore, the control objective of the proposed hand rehabilitation device is to drive the patient’s finger bending angle to a predesigned position. To this end, curvature sensors embedded in the glove are used to measure the finger’s bending angle. A commercial data glove is used to collect the actual finger’s bending angle for calibrating the curvature sensors based on a three-layer back-propagation (BP) neural network. Then the error between the designed joint position and the actual joint position can be calculated. An error proportional control strategy is adopted for the positioning control objective (the controller’s input is the pump speed). Finally, experiments are conducted to validate the effectiveness of control method and the capacity of the proposed hand rehabilitation device.
Source: Preliminary study on the design and control of a pneumatically-actuated hand rehabilitation device – IEEE Xplore Document
Recent neural science research suggests that a robotic device can be an effective tool to deliver the repetitive movement training that is needed to trigger neuroplasticity in the brain following neurologic injuries such as stroke and spinal cord injury (SCI).
In such scenario, adaptive control of the robotic device to provide assistance as needed along the intended motion trajectory with exact amount of force intensity, though complex, is a more effective approach. A critic-actor based reinforcement learning neural network (RLNN) control method is explored to provide adaptive control during post-stroke fine hand motion rehabilitation training.
The effectiveness of the method is verified through computer simulation and implementation on a hand rehabilitation robotic device.
Results suggest that the control system can fulfil the assist-as-needed (AAN) control with high performance and reliability. The method demonstrates potential to encourage active participation of the patient in the rehabilitation process and to improve the efficiency of the process.
Source: IEEE Xplore Abstract (Abstract) – Reinforcement learning neural network (RLNN) based adaptive control of fine hand motion rehabilitati…