Fatigue is a problem with severe consequences for drivers and is one of the most contributing cause for the road accidents in India. Existing attempts to implement fatigue detection for automobile drivers have employed several techniques such as blink frequency detection, posture/movement analysis, lane changes etc. However, the reliability of these methods is not very high due to varying lighting conditions and inherent person to person variation. Further, different road traffic and driving conditions makes these methods less accurate in the Indian scenario. Compared to these techniques, physiological measures derived from captured ECG signals give a more accurate assessment of the driver fatigue.
CSIR-CEERI has developed an algorithm using Heart Rate Variability (HRV) as a potential indicator of fatigue. Changes in HRV, especially Low Frequency to High Frequency Ratio (LF/HF ratio) are quantified and a successful correlation has been established with the fatigue level variation. Customized hardware (integrating ECG, PPG sensors and smart cameras) and advanced machine learning algorithms are being developed which can be readily deployed in the vehicles for non-intrusive and accurate fatigue and drowsiness detection and early warning systems for improved road safety.
Salient Features
- Fatigue derived from heart rate variability.
- The computer program performs analysis of the ECG signals in real time and computes pulse rate & heart rate variability and a parameter gives a LF/HF ratio which is correlated to fatigue.
The know-how was transferred to the sponsorer M/s.Tereso Ventures Pvt. Ltd., Pune.