Generation of Consistent Slip, Trip and Fall Kinematic Data via Instability Detection and Recovery Performance Analysis for Use in Machine Learning Algorithms for (Near) Fall Detection.

Slip, trip and fall (STF) accidents are a major problem for companies in many industries and cause massive work absences. According to the German employers' liability insurance associations, this type of accident has been leading the list of reportable occupational accidents in the traffic, transport and logistics sector for years. A total of 172,045 STF accidents occurred during the reporting period of the German Social Accident Insurance (DGUV Report 2021) seven of which were fatal, and 2,694 new accident pensions. This paper focuses on the standardized acquisition of kinematic data of STF events and its use to assess dynamic stability control during locomotion for machine learning. Accurate detection of near falls via machine learning could help to identify subjects at high risk of falling and ensure that fall prevention interventions are provided to targeted individuals.