Fall Detection Through Inferencing at the Edge

In this work, we design and implement a prototype wireless, wearable, low-power fall detection sensor (FDS) to predict an imminent fall and detect when a fall occurs, for elderly at-risk adults. The neural network model is trained and validated using the Caffe deep learning framework with fall data collected from human subject participants. We also acquire and publish a data set of 3D accelerometer and gyroscope measurements sampled from predefined ADLs (Activities of Daily Life) and falls, using volunteer human subjects.