Date of Publication :31st December 2023

Abstract: As the population continues to age, the frequency of falls at home among older adults is increasing. Falls stand as the primary cause of injuries in this demographic. We proposed a novel design of a quadruped robot which combines autonomous cruising and AI-based accidental falls detection on a commercial available quadruped robot. The proposed system utilizes both human skeleton recognition and object recognition to identify accidental falls in home environments. The point clouds of a millimeter wave radar are processed in order to detection obstacles and navigation. If the predetermined conditions for an accident are met, an emergency notification is sent to the designated contacts.

Reference :