Speech improves human-automation cooperation in automated driving

Abstract

During highly automated driving, upcoming automated manoeuvres (e.g., lane changes) should be communicated to the driver in order to ensure system transparency. As driving time can be used for non-driving-related tasks (NDRT), such as office work or in-vehicle entertainment, drivers might prefer to be informed in a non-distracting way as interruptions of ongoing NDRTs may be perceived as a nuisance (e.g., when drivers are required to retrieve information from attention demanding displays). In this paper, the potential for using speech output to improve human-automation communication is explored. A sample of 17 drivers participated in a simulated automated driving experience in a motion-based driving simulator, which replicated different situations for the participants that required communication between the automation and the driver (e.g., lane changes, avoiding of obstacles). Compared to generic auditory output (i.e., standard information tones), communicating upcoming automated manoeuvres by speech led to a decrease in self-reported visual workload and reduced interference with NDRTs. Participants clearly favoured the speech-based output.