A multi-user assistive system for the user safety monitoring in care facilities

This paper deals with the development of a multiuser assistive system for the monitoring of elders and people with neurological pathologies (e.g. Alzheimer) during their staying in social care or health facilities. The system adopts wearable devices to monitor the dynamics of the users and a dedicated wireless network for communication. The wearable device is able to recognize critical events like falls or prolonged inactivity, to monitor the user posture and to send an alarm to a centralized monitoring system which will broadcast the alarm to caregivers alert devices. The paper focuses on the smart algorithms developed for the classification of the ADL (Activities of Daily Living) which use the information from the inertial sensors installed on the user device. In particular a novel, with respect to the State Of The Art, multi-sensor data fusion approach, combining data from the onboard accelerometer and the gyroscope is presented. Apart from alerts management, the information provided by this system is useful to track the evolution of the user pathology, also during rehabilitation tasks.