Non-contact Vital Signals Measurement on a Mobile Rescue Robot

Improving the mobile robot’s ability to perceive the physiological parameter of the wounded during search and rescue missions has attracted the attention from re-searchers. This paper proposed a vision-based vital signals measurement, we detect the human body from video images captured by the camera that mounted on a mobile rescue robot, then track the face. Extract heart rate (HR) and respiration rate (RR) by detecting colour variations of the face skin. We select human posture and face position as experimental variables, the evaluation results show that this method has good consistency with medical monitor, and the overall root mean square error is HR: 1.58, RR: 1.92. This non-contact vital signals detection method will effectively enhance the performance of the mobile robot in search and rescue tasks.