Desenvolvimento de um sistema embarcado para detecção de queda para idosos

Abstract

This work has as objective to develop a fall detection algorithm to be used by elderly. This algorithm uses the data from the body acceleration to distinguish activities daily living (ADL) from falls. The sensor used in the project is an accelerometer with 3 axis, model ADXL345, that is integrated to a microcontroller board Arduino UNO, wherein the developed algorithm must be embedded, so it can detect a fall in real time. Techniques based on resultant acceleration thresholds, such as the Upper Fall Threshold (UFT) and the Lower Fall Threshold (LFT) that are widely used in the literature, were studied to develop the fall detection algorithm. However, activities that requires large body effort would lead to a false alarm. For that reason, this work also purpose a new technique, which takes into account the energy associated to the signal measured by the accelerometer to discern the activities.