Model reference adaptive control for one dof hand rehabilitation robot

• Main Author: Mazlan, Nur Syahirah
• Format: Thesis
• Language: English; English; English
• Published: 2020
• Subjects: TJ212-225 Control engineering systems. Automatic machinery (General)
• Online Access: http://eprints.uthm.edu.my/407/1/24p%20NUR%20SYAHIRAH%20MAZLAN.pdf; http://eprints.uthm.edu.my/407/2/NUR%20SYAHIRAH%20MAZLAN%20COPYRIGHT%20DECLARATION.pdf; http://eprints.uthm.edu.my/407/3/NUR%20SYAHIRAH%20MAZLAN%20WATERMARK.pdf; http://eprints.uthm.edu.my/407/

This project presents one type of adaptive control technique to control the position of Direct Current (DC) motor for one-DOF rehabilitation robot. A DC motor is generally act as actuator in one-DOF rehabilitation robot and usually the robots are developed without gearing or reducer to make them drivable and can give easiness to stroke patients. Besides that, DC motor has variations in its parameters during operation and it also sensitive to disturbances, presence load and environment. Therefore, it may require an advance technique to make the DC motor position stable. Many different types of controllers were used to provide the accurate positioning of DC motor of one degree of freedom (DoF) rehabilitation robot. One of them is Proportional Integral Derivative (PID) controller. Even though PID controller is simple, stable and easy adjustment, but it not be able to adapt the variation load of stroke patient handgrip stiffness and it can only tune for each standard stiffness only. Therefore, this project will designed and develop the Model Reference Adaptive Control (MRAC) using one of adaptive mechanism that is Lyapunov Method. The MRAC with Lyapunov method is used in order to obtain good controller for position control of DC motor and able to cope with variations handgrip stiffness of stroke patients using one-DoF rehabilitation robot. Besides that, while using the MRAC there will be the uncertainties such as parameter drift and in order to cope with these uncertainties and have robust MRAC, this system required a robust modification such as sigma modification and -modification. In this project, the comparison between MRAC using Lyapunov method, MRAC with sigma modification and MRAC with - modification being examined. As conclusion, simulated results show that 90% of the sigma modification will reduce steady state error, system stable and can reduce settling time rather than using MRAC using Lyapunov method and MRAC with -modification. So, MRAC with sigma modification is the good type of MRAC for controlling the position of DC motor in one-DoF rehabilitation robot where the stroke patients can train their hand in good way.