Development of a control system for robotic rehabilitation devices using co-activation principle

• Main Author: Sy, Armyn Chang
• Format: text
• Language: English
• Published: Animo Repository 2010
• Subjects: Robots—Control systems; Robotic exoskeletons; Manufacturing
• Online Access: https://animorepository.dlsu.edu.ph/etd_masteral/6589; https://animorepository.dlsu.edu.ph/context/etd_masteral/article/12839/viewcontent/CDTG004836_P.pdf
• Institution: De La Salle University

Over the past decade, more and more robotic devices have been tested in the physical rehabilitation setting. Initial results from some devices indicate signs of improvements in subject’s motor functions, some studies reveal the contrary. An important point from these studies is that there are some strategies that actually result to reduced recovery as compared to conventional rehabilitation practices due to the “slacking hypothesis” in which robotic devices could potentially decrease recovery if it encourages slacking, i.e. decrease in motor output, effort, energy consumption and/or attention during motor training. The contradicting results from these studies highlight the need of optimizing the design strategies of robotic therapy devices. This research focuses on the development of a control system for such devices based on proven principles of conventional physical rehabilitation practices, namely (1) practice, (2) specificity and (3) effort. The system gives emphasis on the latter principle of effort through the implementation of the coactivation principle, which ensures that robotic assistance would only be provided when the patient exerts effort, and no corresponding movement from the effort was detected. This aims to reduce over-dependence on the part of the patients and encourages them to be more active in their recovery. The control system is applied through the use of a microcontroller, while an EMG data acquisition system and an angle monitoring potentiometer equips the system with the necessary sensing capabilities. Implementation of the control system on the developed powered exoskeleton revealed great potential. Experimental results show that the system is capable of detecting effort and movement from the feedforward and feedback parameters gathered from the subject. The system also provided only when needed, which is upon detection of effort from the subject and it is not translated into the desired movement. Lastly, extensive tests done on the system indicate that the system is capable of performing all these tasks in real time.