Activity Summarisation and Fall Detection in a Supportive Home Environment
ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 4 - Volume 04
A wearable airbag to prevent fall injuries
IEEE Transactions on Information Technology in Biomedicine - Special section on body sensor networks
Hi-index | 0.00 |
We have developed a wearable airbag that incorporates a fall detection system that uses both acceleration and angular velocity signals to trigger inflation of the airbag. The fall detection algorithm was devised using a thresholding technique with the signals of an accelerometer and gyro sensor. The thresholds of acceleration less than ±3 m/s2 and the integral of angular velocity exceed 0.52 rad/s were used. Five young healthy subjects mimicked falls, and their signals of acceleration and angular velocity were monitored. Then, we developed a fall detection algorithm that could detect signals 300 ms before the fall. This signal was used as a trigger to inflate the airbag to a capacity of 2.4 L. The system has been manufactured but the accuracy was not 100% of operation. In this study we have improved fall detection algorithm to operate correctly in daily life.