Indicating the limits of partially automated vehicles with drivers’ peripheral vision: an online study

Noé Monsaingeon (1, 2) , Yanna Carli (2) , Loïc Caroux (1) , Sabine Langlois (2) , Céline Lemercier (1)

1 CLLE - Cognition, langues, langage, ergonomie
2 Technocentre Renault [Guyancourt]

Automated systems of partially automated vehicles are able to perform the driving task, but can give back the driver all controls in specific conditions. This study aimed to evaluate the effectiveness of an Indicator of Proximity to the Limits of Assistance (IPLA) to anticipate transitions of control. The study was performed online, presenting videos representing situations in which assistance deactivated. A classical interface was compared to an IPLA dedicated to peripheral vision. Participants decided which action to perform. The results revealed that the participants who had the IPLA performed more actions before the system deactivated and expressed greater psychological comfort than the participants with the classic interface. The participants with the IPLA performed less appropriated actions and the IPLA was rated as more cognitively demanding. These results highlight the pertinence of integrating an IPLA dedicated to peripheral vision, but should not encourage the driver to deactivate assistance when not necessary.

Mots clés: Reliability of automation; Human-machine interface; Online study

Domaines: Sciences cognitives; Interface homme-machine [cs.HC]

Noé Monsaingeon, Yanna Carli, Loïc Caroux, Sabine Langlois, Céline Lemercier. Indicating the limits of partially automated vehicles with drivers’ peripheral vision: an online study. 12th International Conference on Applied Human Factors and Ergonomics - AHFE 2021, Jul 2021, New York City, NY, United States. pp.78-85. ⟨hal-03114194⟩