Abstract
We developed drowsiness detection technology using heart rate variability to reduce traffic accidents. Evaluation of this system was carried out during real road driving of a heavy-duty truck, and the validity of the technology was verified. Concerning the capability to distinguish drowsiness and alertness, which denotes off-line performance, it was assured by all the test data. For online use, we found that it could detect all of the extremely drowsy periods for all data within five minutes. As for the result, we found that there is a possibility that the drowsiness technology can be useful for on-line use as well as off-line use.
