PATRIA II
PATTERN RECOGNITION IN AUTOMATIC RISK-ASSESSMENT
In our increasingly complex society, more and more situations arise where errors are made by workers than may have serious consequences for other people: a nurse who is administering medication, an operator of a chemical factory, the driver of a truck with hazardous materials. Each time the fate of individuals depends on the correct task performance by the people who are responsible.
Human Error Detection and Prevention
In the
PatRiA II project VicarVision is working with
Noldus, the
University of Twente,
Fachhochschule Muenster,
Rittmeyer and
Koldijk Consultancy to build a system that helps in the detection and prevention of such errors. The system uses information from body posture and movements, facial expressions, and gaze direction to detect deviations from a known protocol at an early stage, as well as to detect mental states that increase the risk of future errors. By issuing warnings it is possible to prevent serious errors from occurring.
In many professions such as the medical sector, with large systems such as power plants, chemical plants and transportation, people are busy with work that can be risky. Nurses are engaged in certain operations, pilots perform a flight or engineers are controlling large installations. A safe task performance is crucial, but sometimes mistakes that can have catastrophic consequences are made. These errors are usually of the type avoidable adverse events (AAE).