Control of lower limb exoskeleton with simulated EMG signal
No Thumbnail Available
Date
2016-04
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis Ababa University
Abstract
An exoskeleton robot is a kind of a man-machine system which mostly uses combination of human intelligence and machine power. The structure of an exoskeleton robot consists of joints and links which correspond to the human body.
This thesis presents a control system for exoskeletons that utilizes the simulated electrical signals from the muscles, EMG signals, as the main means of information transportation between the human and the exoskeleton. A support action is computed in accordance to the patient’s intention and is executed by the exoskeleton. The mathematical model of the exoskeleton system was based on the mathematical model of a permanent magnet DC servo motor whose parameters can be selected by either Using system identification techniques on a prototype built or by selecting an actuator based on the requirement of the load torque. A linear quadratic Gaussian (LQG) controller for the lower-limb exoskeleton is designed and implemented. Furthermore the robustness of the system to sensor noise and unmodelled system dynamics were analyzed.
The effectiveness of the proposed system is tested through simulation studies using Simulink® software. The 3-D model of the system designed using solidworks® is interfaced with the Simulink® model via Simulink® simmechanics product. The proposed control strategy has shown satisfactory performances in terms robustness and gentleness. The knee joint is to track the ideal range of motion with an error of less than 5 % with the use of LQG controller. The control law is also found to be robust with respect to external disturbances.
Key words: Lower limb exoskeleton, Electromyography (EMG), Modeling, linear quadratic Gaussian (LQG) controller.
Description
Keywords
Lower limb exoskeleton; Electromyography (EMG);Modeling; linear quadratic Gaussian (LQG) controller