Heartbeat Detection Technology for Monitoring Driver's Physical Condition

Kento Tsuchiya, Kenta Mochizuki, Tomoaki Ohtsuki, Kohei Yamamoto

研究成果: Conference article査読


In recent years, the number of reported traffic accidents due to sudden deterioration in driver's physical condition has been increasing, it is expected to develop a system that prevents accidents even if physical condition suddenly changes while driving, or reduces damage through vehicle body control. For this purpose, it is necessary to detect sudden changes of the driver's physical condition, and research is being conducted widely. Among them, it is reported that some of such changes may appear in the heartbeat interval. In other words, by acquiring the driver's heartbeat interval in real time, it may be possible to detect the sudden changes, and reduce traffic accident. Even if a traffic accident occurs, the damage can be reduced by emergency evacuation immediately after detecting sudden changes. Therefore, we focused on the technology to detect the heartbeat interval with 24GHz microwave Doppler radar, which can detect heartbeat non-contactly while maintaining the interior design and passenger's privacy. Doppler radar with microwave is sensitive enough to detect heartbeat, however vibration noise is also superimposed on the sensor signal easily. Thus, in general algorithm studied widely, the problem is that the accuracy of detecting the heartbeat interval deteriorates particularly while driving, because the car vibration is superimposed on the frequency band used for heartbeat analysis, and the SNR (Signal to Noise power Ratio) decreases. To solve this problem, we developed an algorithm that extracts power spectrum over wide band including harmonic components of heartbeat and detects the peaks. This time, our new peak-searching method, in which multiple peak intervals are selected simultaneously based on the previous estimation results, suppresses false detection of peaks caused by vibration noise. As a result, the error rate of the heartbeat interval detection was reduced by half compared to the conventional algorithm in which a peak is detected based on previous heartbeat intervals.

ジャーナルSAE Technical Papers
出版ステータスPublished - 2020 4月 14
イベントSAE 2020 World Congress Experience, WCX 2020 - Detroit, United States
継続期間: 2020 4月 212020 4月 23

ASJC Scopus subject areas

  • 自動車工学
  • 安全性、リスク、信頼性、品質管理
  • 汚染
  • 産業および生産工学


「Heartbeat Detection Technology for Monitoring Driver's Physical Condition」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。