Transferable belief models for human welfare assessment with wearable sensors

This paper describes the use of the transferable belief model in its discrete and continuous forms for fusing and classifying data and information from various wearable sensors that allow for an assessment to be made on the welfare of the user. Data and information is received from sensors and systems that report on individual factors that can affect the user of the wearable sensor system. Some of these are time related and so can be fused at each time step, while others will only be classified, both of which are then fused to allow welfare analysis to take place. Decisions regarding the ability of the user to perform tasks can be made dependant on the results of this analysis, as well as the need for the user to receive medical assistance or support from fellow workers.