Development of a High-Resolution Human-Specific Breath Gas Sensor for Survivor Detection in Disaster Zones

Abstract

Abstract This paper presents a new type of ultrasonic gas molecule concentration sensor for rescue robotics. This device can measure the change of gas concentration with a sampling rate of over 400 kHz. The performance is evaluated by measuring the CO2 concentration in human respiration gas. The experiments show that the proposed sensor could detect a difference between 5% CO2-containing air, humidified air and dry air with over 50 dB signal-to-noise, which are the main components of our respiration gas. Another important result was that our sensor could give information about the 'dead space', which is distributed from the lungs to the mouth. The 'dead space' could not be detected by previously proposed commercially distributed gas sensors because of the time needed to analyze the gas. We verified the distance dependency of the respiration detection in a open space that was considered for use for finding survivors. These results make the proposed sensor especially applicable for finding survivors in disaster zones. Keywords: HUMAN RESPIRATION SENSORULTRASOUNDHIGH SAMPLING RATERESCUE ROBOTICS