In this study, pre-impact fall detection algorithms were developed using an IMU sensor at the waist. Forty male volunteers participated in the experiments (four types of falls and six types of ADLs). An IMU was used to measure acceleration, angular velocity and vertical angle during all activities. Thresholds of acceleration, angular velocity, and vertical angle were set to 0.9 g, 47.3°/s, and 24.7° respectively for algorithm using vertical angle. Thresholds of acceleration, angular velocity, and triangle feature were set to 0.9 g, 47.3°/s, and 0.19 respectively for pre-impact fall detection algorithm using triagle feature. Pre-impact fall detection algorithms with the vertical angle and the triangle feature resulted in the lead time of 402 ms and 427 ms respectively. Both algorithms showed 100% accuracy to detect falls.
"PRE-IMPACT FALL DETECTION USING VERTICAL ANGLE AND ACCELERATION TRIANGLE FEATURE,"
ISBS Proceedings Archive: Vol. 36:
1, Article 271.
Available at: https://commons.nmu.edu/isbs/vol36/iss1/271