Lead University: Carnegie Mellon University
PI: Hae Young Noh, Civil and Environmental Engineering
Co-PI(s): Pei Zhang, Electrical and Computer Engineering

A major goal of elder care is to maintain independence of the elderly for as long as possible. This will improve the quality of life for the elderly and reduce costs and capacity needs for care-professionals. The major obstacle in this is the reduction of daily physical activities, which expedites loss of independence, hospitalization, and premature deaths. However, in elder care facilities, however, independent walking is often discouraged, or prohibited. This is due to the issues of safety of elderly and liability and availability of caregivers. Prior works have been focused on fall detection, but most of them are diagnostic rather than preventive. Furthermore, many of them often require special instrumentation on persons, which is intrusive and inconvenient.

In order to address these challenges, we propose to develop a system to sense, identify, and characterize persons’ gait pattern (to understand their ability to walk, tiredness, activity level, dizziness, etc.) on a fine-grained level with non-dedicated building vibration monitoring sensors. The system consists of three modules: sensing, identification, and characterization. The sensing module collects building vibration signals from structural vibration monitoring systems and detects footstep-induced vibrations at a sensor level. This footstep event detection then triggers identification module to identify each person’s footstep based on various features extracted from the signal. The final module conduct a network level analytics to understand the detected person’s status, such as location, walking path, ability to walk, tiredness, and emotional status. This system will inform doctors and caregivers health status of elderlies around the clock, allowing them more chances to walk independently by providing appropriate and necessary services.