Modelling and angle control of fiber braided bending actuator for finger rehabilitation
Stroke is a prominent cause of disability on a global scale, often resulting in hand impairment that significantly hinders a person's ability to carry out daily activities. Soft actuators present a promising technology for addressing hand impairment in stroke patients, offering a more versatile...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English English |
| Published: |
2023
|
| Subjects: | |
| Online Access: | http://eprints.uthm.edu.my/10979/ |
| Abstract | Abstract here |
| Summary: | Stroke is a prominent cause of disability on a global scale, often resulting in hand impairment that significantly hinders a person's ability to carry out daily activities. Soft actuators present a promising technology for addressing hand impairment in stroke patients, offering a more versatile and adaptable approach to actuation. Despite the benefits of soft actuators, their nonlinearity presents a challenge when it comes to modeling, controlling, and achieving swift response times. Due to the nonlinearity of the system, open-loop systems are not suitable for soft actuator applications. Open-loop controlled pneumatic actuator muscles often struggle with high precision control. The drawbacks can be addressed by implementing a closed-loop control system. The objective of a closed-loop control approach is to perform a dynamic task while enhancing precision, robustness, and actuator conformance to the environment. In this study, one approach to implementing closed-loop control is through system identification (SI), using a transfer function that simulates the actual actuator. The auto-regressive model structure was selected for this study. Pseudo-random binary sequences were employed as the input signal for the SI process. The implementation of a proportional-integral-derivative (PID) controller enabled the control of the angle of the Fiber Braided Bending Actuator (FBBA). Additionally, two tuning techniques were proposed for the PID controller, namely the auto-tuning method and the genetic algorithm method. Both controllers' real-time experiments and simulations are analyzed. The results indicate that, compared to PID tuned using the auto-tuning method, PID tuned using GA demonstrates a significant improvement in both simulation and real-time experiments |
|---|