Gait monitoring for the elderly using a robotic walking aid

The Personal Aid for Mobility and health Monitoring (PAMM) is a robot prototype that was built as part of a doctoral thesis in MIT, back in 2002. The device´s purpose is to assist elderly people in their everyday walking. The robot automatically adjusts itself to the needs of its user by monitoring his/hers gait signals such as direction, velocity and walker grasp force. The objective of this research was to analyze the signals acquired by PAMM in an experiment conducted at an eldercare facility. We used signal processing methods to extract features from the robot´s data and correlated them to the subjects´ gait condition. Time- and frequency-domain features were extracted and then clustered using a K-Mean classifier. A subgroup of the subjects was medically diagnosed as suffering from gait problems. A set of 5 features was indicated as best clustering this subgroup. The results indicate that PAMM, or a similar prototype, can be used as a diagnostic tool that will enable doctors to monitor from a distance the medical conditions of their elderly patients, thus dramatically reducing clinic visits.