Deskwork support system based on the estimation of human intentions

We have proposed a system called an attentive workbench (AWB) that is designed to support desktop work. AWB is the system which estimates a worker´s condition using some sensors and assists a worker using movable trays and projectors from both physical and informational viewpoints. A method for estimating an AWB user´s intention is proposed and presented. An AWB user expresses his/her intention by pointing a finger at a necessary tray. For estimating which tray a user needs, the system uses dynamic Bayesian network to integrate the sensor data obtained from user´s pointing gestures and user´s past action sequences. A description of the experimental results are demonstrated the usefulness of the method proposed here.