研究課題/領域番号 |
18K11408
|
研究種目 |
基盤研究(C)
|
配分区分 | 基金 |
応募区分 | 一般 |
審査区分 |
小区分61020:ヒューマンインタフェースおよびインタラクション関連
|
研究機関 | 法政大学 |
研究代表者 |
Jianhua Ma 法政大学, 情報科学部, 教授 (70295426)
|
研究分担者 |
Huang Runhe 法政大学, 情報科学部, 教授 (00254102)
|
研究期間 (年度) |
2018-04-01 – 2024-03-31
|
研究課題ステータス |
交付 (2022年度)
|
配分額 *注記 |
4,030千円 (直接経費: 3,100千円、間接経費: 930千円)
2020年度: 1,170千円 (直接経費: 900千円、間接経費: 270千円)
2019年度: 1,560千円 (直接経費: 1,200千円、間接経費: 360千円)
2018年度: 1,300千円 (直接経費: 1,000千円、間接経費: 300千円)
|
キーワード | Wearable / Recognition / Activity / Emotion / Human / Dog / Modeling / Attribute / Personality / Platform / organic / smart / service |
研究実績の概要 |
Our research in 2022 fiscal year was focused mainly on the following aspects. First, we used multiple IMU wearables for human activity recognition including sleep posture and body turn classifications for sleeping healthcare. Second, we studied estimation of mountain trails using inertial data from sensors worn by mountain climber. Third, the IMUS sensors were also used for monitoring a dog’s postures and breath rate. Fourth, wearable biosensors including EEG, ECG, breath chest pressure, and galvanic skin response (GSR) were used for detecting various emotion state, such as stress levels, concentration degrees, arousal extents, and valence values. Our research achievements have been published in international conferences and journals.
|
現在までの達成度 (区分) |
現在までの達成度 (区分)
3: やや遅れている
理由
The research in 2022 fiscal year was carried out basically as we planned though the experiments were taken longer time. In the activity recognition, we have done research on a human’s sleep monitoring and a dog’s state monitoring, especially extracting breath data from an inertial sensor worn by the dog. In the mountain trail condition detection, we have achieved at 96% accuracy for subject-depended recognition and 87% accuracy for subject-independent recognition. In the stress recognition of five levels, we have achieved at over 90% accuracy using the combined data of GSR, RRI and breath. In 2022, we also developed an integrated platform for synchronous and auto-tagged data collection from diverse sensors, which is very useful and efficient in experiments using multi wearables.
|
今後の研究の推進方策 |
The research in 2023 will be focused mainly on the following aspects. First, we are going to further improve the integrated platform for taking data synchronously and reliably from more types of wearables. The platform will play an essential role to automatically manage various devices and collect data from them. Second, we plan to extract not only breath data but also heartbeat data of both humans and dogs using wearable IMU sensors for novel applications. Third, we are going to study the detection of drowsiness, fatigue and motion sickness using physiological data, e.g., EEG, ECG, GSR and respiration, from corresponding wearable devices. To be more practical, we shall research more on subject-independent emotion state recognition.
|