Preliminary investigations into human fall verification in static images using the NAO humanoid robot

Advances in medicine have led to an ever aging population who generally wish to retain a significant amount of independence. To this end there has been significant research into technologies that allow for this level of independence while maintaining an appropriate level of safety. One of the most significant risks to the elderly is the danger of falling and in light of this fall detection and alarm systems have been the focus of much of this research. This technology has often been resisted by those it is trying to help. Failing to strike the balance between several factors including reliability, complexity and invasion of privacy has been prohibitive in the adoption of this technology. Whereas some systems rely on cameras being mounted so as to allow complete coverage of a user's home, others rely on being worn 24 hours a day; this paper explores a system using the mobile humanoid NAO robot to perform fall detection.