Abstract
Recently, along with the declining birthrate and an aging population, labor shortage of nursing care support is recognized as a social problem. Among such social backgrounds, Brain-Computer Interface (BCI), which non-contact controls an external device with electroencephalograms, is being actively researched. However, it is still hard to say that BCI has returned to society. The reason for this is as follows: (1) high cost, and (2) instability of the system caused by requiring complicated control action. From this fact, in this research, BCI specialized for serial operation was tried by simplifying the system, and its operation reliability (task success rate / malfunction rate) has been evaluated. From the experiments, as a technical problem, the motor control method was inappropriate, and the evaluation accuracy of the simple electroencephalograph was low. In this paper, the control method of the motor was changed, and the effect of (1) improvement of the success rate of the problem and (2) reduction effect of the malfunction rate between the conventional method and the proposed method was evaluated. In the experiment, motor control experiments by both methods were performed 5 times on each subject (n = 10). As a result, in the evaluation item (1), the task success rate improved in all subjects. Moreover, similarly in the evaluation item (2), reduction in malfunction rate was confirmed in all subjects. After that, the average was calculated from 10 data for both items, and paired t-test with a significance level of 1 % was carried out. Consequently, the task success rate increased significantly by 13.3 %, and the malfunction rate was significantly decreased by 8.8 % (p < .01). The present result suggested that the proposed method is more voluntary and more secure than the conventional method.
