Skip to main content

Research Repository

Advanced Search

Stimulating Conversation: Engaging Drivers in Natural Language Interactions with an Autonomous Digital Driving Assistant to Counteract Passive Task-Related Fatigue

Large, David; Burnett, Gary; Antrobus, Vicki; Skrypchuk, Lee

Stimulating Conversation: Engaging Drivers in Natural Language Interactions with an Autonomous Digital Driving Assistant to Counteract Passive Task-Related Fatigue Thumbnail


Authors

DAVID LARGE David.R.Large@nottingham.ac.uk
Senior Research Fellow

Gary Burnett

Vicki Antrobus

Lee Skrypchuk



Abstract

Engaging in conversation has been shown to be an effective countermeasure to passive task-related fatigue. To investigate the effectiveness of a digital assistant to fulfil the role of conversational partner in counteracting driver fatigue, twenty participants undertook two 30-minute drives in a medium fidelity driving simulator, within a low-feature, monotonous driving environment – following a lead car at 68-mph in lane one of a UK-style motorway. All testing occurred between 13:00 and 16:30, whencircadian and homeostatic influences naturally reduced participants’ alertness. During one of thecounterbalanced drives, participants engaged in natural language interactions with a state-of-the-art digitaldriving assistant, delivered using a Wizard-of-Oz approach. Results suggest that the digital assistant had apositive effect on driver alertness compared with the control condition (no assistant): there was a trendtowards lower perceived sleepiness and significantly higher arousal after driving with the digital assistant.Objectively, interacting with the digital assistant improved lane keeping (lower SDLP) and attractedearlier responses to a hazard situation. There were also significant differences in eye activity whenconversing with the digital assistant, revealed by fewer fixations directed towards the road centre andlarger pupil diameter (suggesting increased arousal). The findings have implications for the design offuture in-vehicle natural language interfaces.

Conference Name 5th International Conference on Driver Distraction and Inattention (DDI2017)
Start Date Mar 20, 2017
End Date Mar 22, 2017
Acceptance Date Mar 20, 2017
Publication Date Mar 20, 2017
Deposit Date May 9, 2019
Publicly Available Date May 23, 2019
Public URL https://nottingham-repository.worktribe.com/output/2034939
Related Public URLs https://www.ifsttar.fr/collections/ActesInteractifs/AII2/index.html

Files





You might also like



Downloadable Citations