Eye/Voice Mission Planning Interface (EVMPI)

摘要

The goal of the Eye/Voice Mission Planning Interface (EVMPI) research is tointegrate voice recognition and eye-tracking technology with aviation displays in order to reduce pilot cognitive and manual workload. The EVMPI technology allows an operator to gaze on user interface items of interest and issue verbal commands/queries that can be interpreted by the system, thus permitting hands-free operation of cockpit displays. This report describes the concept for the EVMPI, general principles for integrating eye and voice input, an EVMPI architecture, and the user interface implementation and evaluation of several aviation mission planning tasks. A primary benefit that arises from the integration of multiple input modalities to infer user intent is that robust performance can be obtained, even when the component technologies (eye and voice) are imperfect. This reduces the accuracy requirements on the individual technologies. GOMS models for eye/voice and conventional throttle/stick interaction protocols were developed for a select set of operator tasks. A preliminary comparison reveals that eye/voice interaction can significantly reduce the total number of operations that need to be performed in particular tasks.