Robotic Healthcare Companion for the Elderly and the Differently Abled with Indoor Human Following and Fall Detection Capability

摘要

The number of falls among elderly people increases each year. The lack of appropriate care and support often leads to serious injuries and loss of lives. Therefore, the elderly and the differently abled people require constant monitoring. It is also quite expensive to hire a caretaker or move the person to a nursing/elder care home. Commercial robots are available only for either healthcare assistance or human following and fall detection. Therefore, in this paper, we propose and develop a robot combining both the functions stated above, with a novel algorithm that is used to lower the risks of a person falling, and to alert the guardian, if necessary, when a fall is detected. Additionally, it can serve as a companion who reminds them to take their medication, to contact their guardian, and maintain company. The robot can identify falls of the three basic types: slip, trip, and faint, with an overall accuracy beyond 84.21% and follow the human with an overall accuracy of over 85.91%, which outperforms the state of the art, where the maximum accuracy obtained was 80%. The primary focus of this study is on the fall detection aspect of the robot.