Estimating the out-of-the-loop phenomenon from visual strategies during highly automated driving

Schnebelen, Damien; Charron, Camilo; Mars, Franck

doi:10.1016/j.aap.2020.105776

Cited by 11 publications

(11 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Indeed, even the potential of a task demand has been shown to affect where we look . These findings are coherent with research suggesting that being out of the loop is associated with more visual attention below the windscreen (Schnebelen et al, 2020) and less cumulative time spent looking at the road ahead (Feldhütter et al, 2017;Jamson et al, 2013;. Gaze off the road in manual driving has been associated with higher crash likelihood (Seppelt et al, 2017) and slower reaction times (Lamble et al, 1999).…”

Section: The Perception-action In Loop In Conditionally Automated Dri...supporting

confidence: 83%

Keeping the Driver in the Loop While Driving With Conditional Automation

Dillmann¹

View full text Add to dashboard Cite

The automation of vehicles can be divided into several steps (SAE International, 2018). Level 0 contains no automation at all. Level 1 contains driving assistance that is either longitudinal or lateral. For instance, cruise control or adaptive cruise control would assist longitudinal vehicle control by holding or adapting velocity. At partial automation (Level 2) both lateral and longitudinal driving assistance systems are combined to support the driver. An example is the combination of adaptive cruise control and lane centering functions. In combination these two assistance functions support the driver in staying in the lane and maintaining a certain speed. However, at this level of automation, the driver remains the labels entail (Flach, 1995;Woods, 1993) and shifting scientific endeavors away from observable human performance (Dekker & Hollnagel, 2004). The Loop Defined From Perception-Action Theory PerspectiveGenerally speaking, perception-action theory originates from James Gibson's seminal work on understanding drivers' and pilots' capacity to control movement adequately (Gibson et al., 1955;Gibson & Crooks, 1938). Building on Gibson's later work (Gibson, 1986) the perception-action theory approach investigates how drivers locomote by interacting with the environment through the process of perception and action loops (Bootsma, 1998;Harrison et al., 2016;Mathieu et al., 2017a). For example, the optical expansion of a stop sign in the retina of drivers guides their braking actions towards it (DeLucia et al., 2016;Lee, 1976;Morando et al., 2016). In turn, the action of braking changes the optical expansion of the stop sign relative to the driver, thereby closing the perception-action loop. Iterations of these perception-action loops allow driving behavior to emerge (Harrison et al., 2016;Mathieu et al., 2017a).In order to understand how drivers can be kept in the loop, it is necessary to understand two key processes underlying the perception-action loop: perceptual attunement and perceptual-motor calibration. Perceptual attunement entails that a driver is exposed to -and able to detect and track -the specifying (i.e. task-relevant) visual information (Bootsma, 1998). In the vehicle braking example above, drivers are perceptually attuned to visual variables such as looming, specifying the relative distance to other vehicles (Fajen & Devaney, 2006). In lane keeping tasks, drivers have been shown to perceptually attune to visual variables such as optical flow, bearing-and slay angles (Li & Chen, 2010), or the inner point of approaching curves (Land & Lee, 1994).To close the perception-action loop, drivers must be proficient in scaling their motor responses adequately to this information, a process termed perceptualmotor calibration (Brand & de Oliveira, 2017;Van Andel et al., 2017). Perceptualmotor calibration is not static but dynamically dependent on the situation. Drivers have been shown to need a period of perceptual-motor calibration when tailwind is introduced in a vehicle braking task (Fajen, 2005b) or wh...

show abstract

Section: The Perception-action In Loop In Conditionally Automated Dri...supporting

confidence: 83%

Keeping the Driver in the Loop While Driving With Conditional Automation

Dillmann¹

View full text Add to dashboard Cite

show abstract

“…During autonomous driving, the driver may also gradually get out of the loop, which may lead to a decrease in the quality of takeover when it is requested (Merat et al, 2019). Recently, we have shown that PLS modelling can discriminate between drivers who supervise the driving scene correctly and drivers who are out of the loop with a high level of mind wandering (Schnebelen et al, 2020). The approach could also be transposed to other areas such as the evaluation of the out-of-the-loop phenomenon in aeronautics, for example (Gouraud et al, 2017).…”

Section: Discussionmentioning

confidence: 99%

Model-based estimation of the state of vehicle automation as derived from the driver’s spontaneous visual strategies

Schnebelen¹,

Charron²,

Mars³

2021

JEMR

Self Cite

View full text Add to dashboard Cite

When manually steering a car, the driver’s visual perception of the driving scene and his or her motor actions to control the vehicle are closely linked. Since motor behaviour is no longer required in an automated vehicle, the sampling of the visual scene is affected. Autonomous driving typically results in less gaze being directed towards the road centre and a broader exploration of the driving scene, compared to manual driving. To examine the corollary of this situation, this study estimated the state of automation (manual or automated) on the basis of gaze behaviour. To do so, models based on partial least square regressions were computed by considering the gaze behaviour in multiple ways, using static indicators (percentage of time spent gazing at 13 areas of interests), dynamic indicators (transition matrices between areas) or both together. Analysis of the quality of predictions for the different models showed that the best result was obtained by considering both static and dynamic indicators. However, gaze dynamics played the most important role in distinguishing between manual and automated driving. This study may be relevant to the issue of driver monitoring in autonomous vehicles.

show abstract

“…Schooler et al (2014) found MW to be an antecedent to poor performance. Similar to Schnebelen, Charron, & Mars, (2020), this study assimilates the OOTL phenomenon to MW, and studies it's effect on spontaneous gaze behavior as a way of identifying the OOTL phenomenon, as well as examining it's effect on take over behavior.…”

Section: Introductionmentioning

confidence: 99%

Identifying the Out Of The Loop phenomenon during driving automation using spontaneous gaze behavior

Chouchane

Nakamura

Sakai

et al. 2022

Proceedings of the Human Factors and Ergonomics Society Annual

View full text Add to dashboard Cite

Human factors research that addresses driver-automation interaction is required to mitigate crashes related to the deployment of automated vehicles. While Automated Driving Systems still require driver supervision, there will remain room for human error. This paper presents results from the first in a series of experiments that aim to estimate driver situation awareness using spontaneous gaze behavior and to ensure the driver is ready for takeover. Spontaneous gaze behavior was studied and compared among 13 participants between partial driving automaton and driver assistance automations condition to extract indicators of the out of the loop phenomenon.

show abstract

Estimating the out-of-the-loop phenomenon from visual strategies during highly automated driving

Cited by 11 publications

References 38 publications

Keeping the Driver in the Loop While Driving With Conditional Automation

Keeping the Driver in the Loop While Driving With Conditional Automation

Model-based estimation of the state of vehicle automation as derived from the driver’s spontaneous visual strategies

Identifying the Out Of The Loop phenomenon during driving automation using spontaneous gaze behavior

Contact Info

Product

Resources

About