Study on BMI operation GUI for more flexible control of autonomous wheelchairs

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K11207
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 59010:Rehabilitation science-related
• Research Institution: Kagawa National College of Technology
• Principal Investigator: Onishi Akinari 香川高等専門学校, 電子システム工学科, 講師 (20747969)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: 脳波 / 車いす / 自律走行 / P300 / Brain-computer interface / Brain-machine interface

Outline of Final Research Achievements

Brain-computer interface (BCI) or brain-machine interface (BMI) can decode our brain signals into control commands for, e.g., electric wheelchair. Previous BMI application studies focused only on a specific device. To be more practical, it will be preferable if the BMI could control autonomous wheelchair in addition to the varieties of devices that are located near the user. This study developed a BMI system that can control autonomous wheelchair by designating goal of a 2D map. In addition, household appliances near the user is automatically installed to the BMI by pairing with ESP32 via Bluetooth. I successfully controlled the BMI so that the autonomous wheelchair moved to a next room, and successfully turned a desk light on via the BMI simultaneously. Furthermore, appending and removing household appliances to the BMI stimulator does not influence the accuracy of the BMI, keeping almost 100% classification accuracy.

Free Research Field

福祉工学

Academic Significance and Societal Importance of the Research Achievements

本研究を通して脳波で目的地を指定する新たな方法を提案した．この方法と自律走行技術を融合させることで，脳波で目的地を入力してそこまで自律走行することを可能とした．
また，P300 BMIの刺激提示器を訓練時から変化させた場合の影響について調べた．その結果，文字を入れ替えるなどの変更の場合，刺激提示画面を訓練時から変更しても大きな影響を受けずに制御できることが分かった．また，従来の灰/白刺激よりもスマイリーを命令の上に描画する形式のほうが高い精度が得られたため，電動車いすや家電制御にはスマイリーを用いた刺激提示器のほうが適していることが分かった．