Dynamical Analysis for Structure Fall Forecasting
Abstract
Many approaches of evaluating fall risk are being investigated as the incidence of senior fall victims in the UK keeps rising. Conventional evaluations carried out in lab and hospital settings are sometimes costly and cumbersome for individuals and medical providers. By taking use of the increasing number of sensors available, passive in-home surveillance systems provide an alternative, affordable option. The massive volumes of data produced by these sensors may be processed by machine learning algorithms, which help with risk assessment, the detection of falls, forecasts, and activity identification. The intricacies of sensor data, the necessary analysis, and the automated learning techniques used for fall evaluations are all explored in this work. The viability of active tracking using wearables and vision-based sensors is investigated, and the state of the art of passive in-home monitoring research is also explored. Lastly, a comparison is made between several methods for risk assessment, fall prediction, and detection.