学位論文

生体信号による操縦装置を用いた電動車いすの操縦方法の改善

IMPROVEMENT OF A CONTROL DEVICE SYSTEM FOR ELECTRIC WHEELCHAIR ON THE BASIS OF BIOSIGNAL

jpn

1

57

2016-03-24

Not Applicable (or Unknown)

2016-03-24

修士(工学)

理工学研究科機械工学専攻; 指導教授 石井千春

In this study, a control system of an electric wheelchair (EWC) to assist disabled people such as persons with spinal cord injury, which was developed in our previous study, was improved to enhance maneuverability of the EWC. The control device is attached to a joy-stick of the ready-made EWC and controls the joy-stick by motor drive. The surface electromyograms (SEMGs) of four muscles, which are the neck, both sides of shoulder and the cheek, are measured, and the EWC is navigated based on the SEMG of the user. By identifying the SEMGs, seven patterns of the navigation of the EWC are possible, which are stop, advance, back, right turn, left turn, right back turn, and left back turn. Thus, users can navigate the EWC by their own intention without using their hands. There are two problems in the previous study. The first one is the power source of the system, and the second one is the maneuverability of the EWC. As for the first problem, the electric power to the system was supplied from a home power source. Therefore, motion space of the EWC was restricted due to the cable from the outlet. This problem was solved by replacing the power source of the system with the external battery. As for the second problem, this study focused on improvement of the four motions, which are right turn, left turn, right back turn, and left back turn. In the previous study, these motions were performed monotonically, because rotation angle of the joy-stick to achieve these motions was constant. To enhance the maneuverability of the EWC, two new methods to determine the target angle of the joy-stick based on both the degree and duration of the muscular activity during straining the corresponding muscle are proposed for these four motions. In the first method, duration of the muscular activity is reset by each straining of the corresponding muscle. In the second method, duration of the muscular activity is accumulated for each straining of the corresponding muscle. The proposed two new methods enable users to adjust the rotation angle of the joy-stick. Therefore, users can tweak the traveling direction of the EWC by their own intention using the SEMG. Thus, the maneuverability of the EWC is improved. In order to verify the maneuverability of the EWC by the proposed two new methods, run experiments were carried out for an S-Slalom course and an N-crank course, respectively, and results were compared with the previous method.

Thesis