Do drivers change their manual car-following behaviour after automated car-following?

Abstract

There is evidence that drivers’ behaviour adapts after using different advanced driving assistance systems. For instance, drivers’ headway during car-following reduces after using adaptive cruise control. However, little is known about whether, and how, drivers’ behaviour will change if they experience automated car-following, and how this is affected by engagement in non-driving-related tasks (NDRT). The aim of this driving simulator study, conducted as part of the H2020 L3Pilot project, was to address this topic. We also investigated the effect of the presence of a lead vehicle during the resumption of control, on subsequent manual driving behaviour. Thirty-two participants were divided into two experimental groups. During automated car-following, one group was engaged in an NDRT (SAE Level 3), while the other group was free to look around the road environment (SAE Level 2). Both groups were exposed to Long (1.5 s) and Short (.5 s) Time Headway (THW) conditions during automated car-following, and resumed control both with and without a lead vehicle. All post-automation manual drives were compared to a Baseline Manual Drive, which was recorded at the start of the experiment. Drivers in both groups significantly reduced their time headway in all post-automation drives, compared to a Baseline Manual Drive. There was a greater reduction in THW after drivers resumed control in the presence of a lead vehicle, and also after they had experienced a shorter THW during automated car-following. However, whether drivers were in L2 or L3 did not appear to influence the change in mean THW. Subjective feedback suggests that drivers appeared not to be aware of the changes to their driving behaviour, but preferred longer THWs in automation. Our results suggest that automated driving systems should adopt longer THWs in car-following situations, since drivers’ behavioural adaptation may lead to adoption of unsafe headways after resumption of control.

Physiological Indicators of Driver Workload During Car-Following Scenarios and Takeovers in Highly Automated Driving

“…They were also asked not to overtake the lead vehicle, but otherwise, follow the normal rules of the road, ensuring the safe operation of the vehicle, and maintaining their desired distance from the lead vehicle. After each experimental drive, the participants were given a 10-min break, during which they were asked to complete a set of questionnaires, including Arnett's Sensation Seeking Questionnaire (Arnett, 1994), traffic locus of control ( Özkan & Lajunen, 2005) and the Driver Style Questionnaire (French et al, 1993), see Louw et al (2020). However, results from these questionnaires are not reported here, since they did not include questions about driver workload.…”

“…However, participants' subjective ratings of riskiness due to the THW maintained by the vehicle was also dependent on environmental factors such as visibility (fog), and traffic conditions such as following a truck, with driver discomfort increasing with lower visibility, increased traffic, and when following a truck, as opposed to a car. In Louw et al (2020), we observed, via subjective ratings, that shorter headways maintained by HAD are perceived as riskier or unsafe by drivers, especially when they are not in control of the driving task. Resuming control from automation in the presence of a closer lead vehicle is also likely to be more demanding, especially following engagement in an NDRT (Mehler et al, 2009), further exacerbating the OOTL effect, with studies on HAD showing that engagement in NDRTs increases driver workload, and negatively affects their driving performance after takeovers (Du et al, 2020;Wandtner et al, 2018;Zeeb et al, 2016).…”

Study of the Technogenic Impact of Motor Transport on the Environment in Cities

“…Estimates can vary greatly depending on the country and region, so many experimental studies report high concentrations of pollutants on roads with a large number of diesel vehicles. It was found that a small part of dirty vehicles, for example, with outdated emission standards, high mileage, or failures in engine emission control systems, can be characterised by more significant emissions than the rest [17]. Driving conditions can affect the quality of roadside air in two ways, namely the level of emissions and turbulence caused by the vehicle.…”

Do you trust me? Driver responses to automated evasive maneuvers

“…Monitoring the environment but not being in physical control of the vehicle is defined as the “on-the-loop” state. Due to the expectation for the driver to resume manual control while in CAD and shift to an “in-the-loop” state from an “on-the-loop” or “out-of-the-loop” state, a significant portion of the human factors research focused on the higher levels of driving automation systems has focused on the resumption of manual control ( Seppelt and Victor, 2016 ; Louw et al, 2020 ). In contrast, there has been limited research on how drivers respond to evasive maneuvers initiated by a CAD vehicle.…”