CONTROL SYSTEM DESIGN OF SOFT ROBOTIC-PNEUMATIC NETWORK FOR HAND REHABILITATION WITH REPETITIVE MOVEMENT

• Main Author: Selvia
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/66721
• Institution: Institut Teknologi Bandung

Studies show that 65% of post-stroke patients often experience hand disability. Disability after stroke could be cured by rehabilitation during the golden time, the first month after stroke. One of hand rehabilitation methods is repetitive flexion-extension exercise. Normally, this rehabilitation would be guided by a therapist. However, there is shortage of therapists so that rehabilitation devices are excessively needed to help patients carrying out independent and repetitive rehabilitation. In addition, the difference of the patient's residual motion ability requires personalization of assistive device. In the past few years, robots for people with disabilities using rigid bodies have been developed, but these aids still have many weaknesses, like stiffness, being expensive, unsafety, and inconvenience for patients to use. These weaknesses encourage the emergence of soft robotic tools such as Pneu-Net which offers greater flexibility, cheaper, safe, adaptive, and comfortable to use by patients. However, to meet the needs of independent, repetitive, and personal rehabilitation, designing a soft robotic electro-pneumatic control system is necessary to produce repetitive flexion-extension movements in a chageable mode according to residual motion ability of the patient. In this study, the designed control system method is to send a PWM duty cycle to the control valve to control the flow of air that enters the soft actuator to produce flexion movement and the on/off control system on the release valve to release air from the soft actuator so that the extension movement is obtained. Meanwhile, to meet the patient's personalization needs according to their movement abilities, a dynamic response test of the system was carried out by varying the duty cycle from 10% - 100%. From this test, the response time is divided into three modes of movement, which are slow, medium, and fast with a duration of 6 seconds, 4 seconds, and 2 seconds. Based on the division of motion modes, designing a time scheduling control algorithm with the three modes of movement and adding release mode with constant duration between modes were done. Based on implementation, the highest RMSE value is 19.09o in slow mode and the lowest is 11.27o in medium mode. The distribution of the RMSE values between modes and their repetitive cycles was not much different, indicating that the flexion-extension repetitive control performance was quite good and stable.