Fall prevention is a major research interest nowadays as issues related to falls have become increasingly concerning with the world’s aging population. A fall intervention exosuit project has previously been created to assist in trip-falls. However, successful implementation implies that falls are no longer observed using conventional kinematics and dynamics sensors. Hence, a fall detection method based on Galvanic Skin Response (GSR) sensor has been proposed. The GSR sensor is capable of sensing changes in stress levels which can be induced by falls. This allows for detection of instability felt by the user even when the fall intervention device successfully prevents a fall. The method applies receiver operating characteristics (ROC) method to analyze the raw data collected from the GSR sensor. In an experiment that involves fifteen subjects, the method has been proven to be able to distinguish a fall correctly with a 0.9938 area under curve (AUC). In this paper, the method has been described and some findings on the performance of the GSR sensor in fall detection have been presented.

Master of Science (Systems and Project Management)