Drivers’ Performance in Non-critical Take-Overs From an Automated Driving System—An On-Road Study

Rydström, Annie; Mullaart, Mattias Söderholm; Novakazi, Fjollë; Johansson, Mikael; Eriksson, Alexander

doi:10.1177/00187208211053460

Cited by 9 publications

(4 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Overall, the non-critical average take-over time of 5.62 seconds we found, is between an on the road driving study reporting 2.70 and 5.50 seconds (Naujoks et al, 2019) and a similar study reporting an average of 8.70 seconds (Rydström et al, 2022). These results align well with the typical 6 seconds drivers needed to take back control from automation when engaged with a non-driving related task in a simulator study (Eriksson & Stanton, 2017b).…”

Section: Discussionsupporting

confidence: 82%

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: Discussionsupporting

confidence: 82%

Keeping the Driver in the Loop While Driving With Conditional Automation

Dillmann¹

View full text Add to dashboard Cite

show abstract

“…In both studies, drivers managed to transition back to manual driving in response to TORs. However, Rydström et al [17] reported take-over times up to 25 s, whereas Naujoks et al [16] only report take-over times up to 10 s. The likely explanation behind the longer take-over times in [17] is that the participants had no prior experience of the TOR and deactivation strategy prior to the test, while the participants in [16] had practiced beforehand.…”

Section: Need To Validate Previous Findings In Real Trafficmentioning

confidence: 93%

“…Therefore, there is a need to extend current findings (from the virtual environment involving critical scenarios) with studies conducted in real traffic in which drivers are given more than 7 s to deactivate AD. To the knowledge of the authors, only two studies have investigated drivers' responses to TORs during AD (hands-free driving allowed) in real traffic and under noncritical conditions (see, [16], [17]). Both these studies allowed participants to engage in NDRTs (as a manipulated variable in [16] and as voluntary engagement in [17]).…”

Section: Need To Validate Previous Findings In Real Trafficmentioning

confidence: 99%

“…To the knowledge of the authors, only two studies have investigated drivers' responses to TORs during AD (hands-free driving allowed) in real traffic and under noncritical conditions (see, [16], [17]). Both these studies allowed participants to engage in NDRTs (as a manipulated variable in [16] and as voluntary engagement in [17]). In both studies, drivers managed to transition back to manual driving in response to TORs.…”

Section: Need To Validate Previous Findings In Real Trafficmentioning

confidence: 99%

See 1 more Smart Citation

Driver Response to Take-Over Requests in Real Traffic

Pipkorn,

Tivesten,

Flannagan

et al. 2023

IEEE Trans. Human-Mach. Syst.

View full text Add to dashboard Cite

Existing research on control-transitions from automated driving (AD) to manual driving mainly stems from studies in virtual settings. There is a need for studies conducted in real settings to better understand the impacts of increasing vehicle automation on traffic safety. This study aims specifically to understand how drivers respond to take-over requests (TORs) in real traffic by investigating the associations between 1) where drivers look when receiving the TOR, 2) repeated exposure to TORs, and 3) the drivers' response process. In total, thirty participants were exposed to four TORs after about 5-6 min of driving with AD on public roads. While in AD, participants could choose to engage in non-driving-related tasks (NDRTs). When they received the TOR, for 38% of TORs, participants were already looking on path. For those TORs where drivers looked off path at the time of the TOR, the off-path glance was most commonly towards an NDRT item. Then, for 72% of TORs (independent on gaze direction), drivers started their response process to the TOR by looking towards the instrument cluster before placing their hands on the steering wheel and their foot on the accelerator pedal, and deactivating automation. Both timing and order of these actions varied among participants, but all participants deactivated AD within 10 s from the TOR. The drivers' gaze direction at the TOR had a stronger association with the response process than the repeated exposure to TORs did. Drivers can respond to TORs in real traffic. However, the response should be considered as a sequence of actions that requires a certain amount of time.

show abstract

Design friction in autonomous drive—exploring transitions between autonomous and manual drive in non-urgent situations

Bley,

Eriksson,

Johansson

et al. 2023

Pers Ubiquit Comput

View full text Add to dashboard Cite

In the ongoing turn to automation, the growing trend towards the design of conditionally and highly automated vehicles (C/HAV) is evident. In a CAV, the driver no longer needs to partake in the driving. However, the vehicle might send a takeover request (TOR) when the CAV’s system reaches its operational boundaries, i.e. a call for a transition from autonomous to manual drive. Previous research on TORs has focused on the context of urgent situations, e.g. hazards and unpredictable events. Furthermore, it has been noted that drivers’ situation awareness (SA) deteriorates after being in autonomous drive. However, less is known about TORs in non-urgent situations. Motivated by this need, the study explores how design friction can serve as a guiding concept for transferring control between autonomous and manual drive in non-urgent situations to increase situation awareness. Design friction is defined as elements of interactions that steer attention and guides the driver to take informed decisions. The work resulted in prototypes that leveraged design friction as part of a takeover sequence. The proposed design was empirically evaluated in a fixed-base medium-fidelity driving simulator. The results indicated that the level of friction might have been too extensive, as some annoyance was expressed. However, participants claimed to feel calm and aware of their surroundings at the moment of regaining control of the vehicle. This suggests that design friction is a promising tool for guiding concept design to enhance transitions from autonomous to manual drive.

show abstract

Drivers’ Performance in Non-critical Take-Overs From an Automated Driving System—An On-Road Study

Cited by 9 publications

References 36 publications

Keeping the Driver in the Loop While Driving With Conditional Automation

Keeping the Driver in the Loop While Driving With Conditional Automation

Driver Response to Take-Over Requests in Real Traffic

Design friction in autonomous drive—exploring transitions between autonomous and manual drive in non-urgent situations

Contact Info

Product

Resources

About