電波センシングと機械学習を用いた運転者の状態識別システムの開発

カテゴリ: 部門大会

論文No: GS10-3

グループ名: 【C】2023年電気学会電子・情報・システム部門大会

発行日: 2023/08/23

タイトル(英語): Development of Driver's State Identification System Using Radio Wave Sensing and Machine Learning

著者名: 許 子良（東京工科大学）,天野 直紀（東京工科大学）

著者名(英語): ZiLiang Xu (Tokyo University of Technology),Naoki Amano (Tokyo University of Technology)

キーワード: チャネル状態情報|機械学習|ドライバーの状態ドライバーの状態|Channel state information|Machine learning|Driver state

要約(日本語): Driver's dangerous driving is one of the main causes of road traffic accidents. This study uses CSI and machine learning to detect the driver's breathing state, to judge the driver's driving state and reduce the incidence of traffic accidents. This study designed a CSI non-contact human breathing detection system based on the Fresnel zone model. The system mainly includes CSI data collection and processing of data outliers and noise. And select the sub-carrier that most represents the breathing of the human body for mechanical learning and prediction of the breathing pattern. In the simulation experiment, the four breathing states of "normal breathing", "normal deep breathing", "rapid deep breathing" and "breathing suddenly stopped" were predicted, with an average accuracy rate of 86%.