SAE Level Research Articles

Effects of In-vehicle Auditory Interactions on Takeover Performance in SAE L2 Semi-automated Vehicles

Commercially available automated vehicles require drivers to maintain focus on their driving environment and be prepared to fully control their vehicles (i.e., perform a takeover) when critical incidents occur (e.g., sudden automation failures). Therefore, drivers are discouraged from engaging in non-driving tasks that cause visual or manual distractions. Auditory interactions, despite being considered a safe alternative, can consume the attentional resources of drivers, causing them to respond poorly to critical situations. This study investigates (1) how varying levels of auditory interactions affect takeover performance and (2) what physiological contexts are related to the takeover performance in SAE Level 2 automated driving. For the investigation, 50 drivers wore wearable devices that collected various physiological signals and performed six different auditory tasks during L2 automated driving in a simulator-based experiment. The results showed that auditory interactions could degrade the takeover performance and that the task demand for auditory interactions nonlinearly affected the takeover performance, possibly owing to behavior changes intended to prevent the task difficulty from becoming excessively high. Additionally, physiological contexts such as pupil diameter, dispersion of eye movements, and inter-beat interval, were found to be related to the takeover performance. Subsequently, we discussed drivers’ behavior changes, practical deployment of in-situ physiological measures, and design implications for mitigating the degradation of takeover performance due to auditory tasks.

Supporting Driver Attention Toward Potential Hazards During Takeover: A Preliminary Result

This study investigates the impact of a supportive system on takeover transitions in conditionally automated driving (SAE Level 3). The supportive system is designed to direct drivers’ attention toward potential hazards in the environment when a takeover request occurs. The study comprises two components: (a) identifying various types of potential hazards using naturalistic driving data, and (b) conducting a driving simulator study to develop and assess a gaze guidance system based on the N-SEEV model of visual attention. Results indicate that drivers using a highly salient attention guidance system were less likely to collide with a secondary hazard during takeover transitions. This suggests that gaze guidance support is an effective approach for assisting drivers during takeover transitions.

Comparing Static and Dynamic Vibration Cues in Wristband Haptic Feedback for Enhanced Driver Response in Automated Vehicles

Higher levels of driving automation make effective takeover requests critical. The wrist’s sensitivity to vibration makes wristband devices a potential carrier for sending these requests. However, the impacts of conveying takeover requests through directional vibrotactile patterns such as dynamic patterns (sequential stimuli occurring at different locations on the wrist) and static patterns (fixed stimuli at the same locations on the wrist) are unclear. Therefore, this study examined the effects of directional vibrotactile patterns on takeover performance among younger and older adults. Participants responded to four patterns (two dynamic, one static, and one baseline) in a simulated SAE Level 3 automated vehicle. Takeover performance was evaluated using reaction time and takeover time. The results show that the static and baseline patterns had shorter reaction and takeover times compared to the dynamic patterns. In addition, younger adults react faster to takeover requests compared to older adults. Findings provide important insights for the future design of human-machine interfaces via wristband devices for automated vehicles.

Assessing Meaningful Visual and Tactile Feedback for Effective Automated Vehicle Takeover by Hearing and Non-hearing Drivers

With SAE Level 3 automated vehicles handling most driving tasks, there are still situations when the driver needs to take over. Multimodal displays have been introduced to inform drivers of the need to take over for critical scenarios (e.g., in construction zones) in instructional or informative formats. However, the effects of multimodal displays on takeover performance for drivers with hearing impairments are still unclear. Therefore, the goal of this study was to investigate how signal type (single tactile (T), single visual (V), and visual and tactile combined (VT)), information type (instructional, informative, and baseline), and hearing impairment (hearing-impaired and non-hearing-impaired drivers) affect drivers’ takeover performance. Findings show that signal type significantly influenced reaction and takeover times, with multimodal signals (VT) resulting in faster reactions compared to single modal signals. Additionally, the baseline condition yielded the faster reaction times compared to both instructional and informative formats. Hearing impairment, however, did not significantly affect reaction and takeover times. Findings may inform the development of future vehicle interfaces to assist drivers with hearing impairments.

User comfort and naturalness of automated driving: The effect of vehicle kinematic and proxemic factors on subjective response

User comfort in higher-level Automated Vehicles (AVs, SAE Level 4+) is crucial for public acceptance. AV driving styles, characterised by vehicle kinematic and proxemic factors, affect user comfort, with “human-like” driving styles expected to provide natural feelings. We investigated a) how the kinematic and proxemic factors of an AV's driving style affect users' evaluation of comfort and naturalness, and b) how the similarities between automated and users' manual driving styles affect user evaluation.Using a motion-based driving simulator, participants experienced three Level 4 automated driving styles: two human-like (defensive, aggressive) and one machine-like. They also manually drove the same route. Participants rated their comfort and naturalness of each automated controller, across twenty-four varied UK road sections. We calculated maximum absolute values of the kinematic and proxemic factors affecting the AV's driving styles in longitudinal, lateral, and vertical directions, for each road section, to characterise the automated driving styles. The Euclidean distance between AV and manual driving styles, in terms of kinematic and proxemic factors, was calculated to characterise the human-like driving style of the AV.We used mixed-effects models to examine a) the effect of AV's kinematic and proxemic factors on the evaluation of comfort and naturalness, and b) how similarities between manual and automated driving styles affected the evaluation. Results showed significant effects of lateral and rotational kinematic factors on comfort and naturalness, with longitudinal kinematic factors having a less prominent effect. Similarities in vehicle metrics, such as speed, longitudinal jerk, lateral offset, and yaw, between manual and automated driving styles, enhanced user comfort and naturalness.This research facilitates an understanding of how control features of AVs affect user experience, contributing to the design of user-centred controllers and better acceptance of higher-level AVs.

User expectations for an automated motorway access assistant system

IntroductionTo avoid the uncertainty present in a cooperation between manual vehicles (MVs) and automated vehicles (AVs), cooperative connected automated mobility (CCAM) is frequently proposed. In this contribution, we consider the situation of a motorway slip road where an AV (SAE level 3 or higher) on the motorway cooperates by opening a suitable gap for an entering AV. To ensure a meeting of the cooperation partners, speed adjustment by the system might often be necessary.MethodsTo investigate the acceptance of an automated cooperative motorway access assistant a driving simulator study (N = 41) was conducted. Participants drove six laps on a virtual test track in a level 3 AV. Three levels of speed adjustment and two levels of an HMI (with or without HMI) were tested. They experienced the automated cooperative motorway access assistant in two situations. Firstly, when they entered the motorway themselves and benefited from the system, and secondly, when the system reduced their vehicle's speed to allow for cooperation with another vehicle on a motorway slip road ahead.ResultsThe results show that the acceptance of the assistant depends significantly on the magnitude of speed reduction: the greater the speed adjustment, the worse the user rating, and the more often the cooperation was terminated. Obviously, the acceptance depends on the perspective: If the users themselves benefit from the speed adjustment of others, they rate the system significantly better than if they themselves must accept the restriction. Only in the case of a small speed reduction, the assistant is rated as well as if the user benefits from the system. The availability of a human-machine interface (HMI) providing information about the system status influenced acceptance only marginally but had a significant influence on the trust in the automation.DiscussionThe results suggest that people are willing to accept a personal disadvantage in order to give others an advantage in the context of CCAM. However, acceptance depends on the magnitude of the disadvantage. The overall agreement scores of the system with HMI did not differ from those without HMI, although the HMI was found to slightly reduce the termination of cooperation. Three possible reasons for this result, which differs from previous research, are discussed.

Drivers’ situational awareness of surrounding vehicles during takeovers: Evidence from a driving simulator study

This study aimed to understand the influence of surrounding vehicle configuration, driving lane, and traffic density on drivers’ situational awareness (SA), takeover performance, and eye-tracking behaviors in conditionally automated driving. An experiment was conducted with the participation of 40 university students using a fixed-base driving simulator configured to simulate SAE Level 3 automation. During the experiment, participants were engaged in playing Tetris on a tablet as a non-driving related task in automated driving mode. Upon hearing an auditory takeover request, participants were instructed to take control of the vehicle, and then complete a scene reconstruction task to report their SA after transferring control back to the automated driving system. Our findings showed that drivers often neglected vehicles at their sides and rear during the takeover, which was associated with higher collision risks. Higher oncoming traffic density led to drivers’ worse SA of surrounding vehicles but more cautious driving behavior. Driving in the right lane generally resulted in smoother takeovers with lower collision risks. Interestingly, while SA did moderate the impacts of driving conditions on safety margins, a higher level of SA did not consistently relate to improved performance, especially in complex scenarios. This suggests the need for support systems that guide drivers to focus on safety–critical objects rather than simply amplifying SA in general. These insights have significant implications for the design of driver monitoring and support systems in automated vehicles.

Hello, is it me you're Stopping for? The Effect of external Human Machine Interface Familiarity on Pedestrians' Crossing Behaviour in an Ambiguous Situation.

We investigated how different deceleration intentions (i.e. an automated vehicle either decelerated for leading traffic or yielded for pedestrians) and a novel (Slow Pulsing Light Band - SPLB) or familiar (Flashing Headlights - FH) external Human Machine Interface (eHMI) informed pedestrians' crossing behaviour. The introduction of SAE Level 4 Automated Vehicles (AVs) has recently fuelled interest in new forms of explicit communication via eHMIs, to improve the interaction between AVs and surrounding road users. Before implementing these eHMIs, it is necessary to understand how pedestrians use them to inform their crossing decisions. Thirty participants took part in the study using a Head-Mounted Display. The independent variables were deceleration intentions and eHMI design. The percentage of crossings, collision frequency and crossing initiation time across trials were measured. Pedestrians were able to identify the intentions of a decelerating vehicle, using implicit cues, with more crossings made when the approaching vehicles were yielding to them. They were also more likely to cross when a familiar eHMI was presented, compared to a novel one or no eHMI, regardless of the vehicle's intention. Finally, participants learned to take a more cautious approach as trials progressed, and not to base their decisions solely on the eHMI. A familiar eHMI led to early crossings regardless of the vehicle's intention but also led to a higher collision frequency than a novel eHMI. To achieve safe and acceptable interactions with AVs, it is important to provide eHMIs that are congruent with road users' expectations.

Examining Psychological Conflict-Handling Strategies for Highly Automated Vehicles to Resolve Legal User-Vehicle Conflicts

As vehicle automation progresses, with SAE levels 3 to 5 introducing varying degrees of control transfer from driver to vehicle, a key challenge emerges: aligning driver interests with the automated vehicle's (AV) operational goals. Despite technical feasibility, drivers' tendencies to intervene due to distrust in or dissatisfaction with AVs necessitate consideration of control mechanisms in future AV designs. This study investigates the potential of persuasion strategies inspired by Human-Robot Interaction research to harmonize driver actions with AV goals in legal conflict situations. We conducted a Virtual Reality driving simulator study with 36 participants, comparing 11 persuasive conflict-handling strategies and a baseline of no persuasion. Our research helps understand the effects of persuasion on users' compliance with law-abiding AV goals, trust, and acceptance towards the AV and evaluates the effectiveness of these strategies based on their working mechanism (cognitive, emotional, social, politeness) and valence (negative, neutral, positive). Results show that none of the strategies could substantially increase compliance with the AV, but neutral and positive persuasion strategies did not decrease acceptance and trust towards the AV compared to no persuasion. These findings contribute to the discourse on cooperation design and control allocation in AVs, particularly in scenarios involving legal conflicts.

Comparison of Waymo rider-only crash data to human benchmarks at 7.1 million miles

Objectives This article examines the safety performance of the Waymo Driver, an SAE level 4 automated driving system (ADS) used in a rider-only (RO) ride-hailing application without a human driver, either in the vehicle or remotely. Methods ADS crash data were derived from NHTSA’s Standing General Order (SGO) reporting over 7.14 million RO miles through the end of October 2023 in Phoenix, Arizona, San Francisco, California, and Los Angeles, California, and compared to human benchmarks from the literature. Results When considering all locations together, the any injury reported crashed vehicle rate was 0.6 incidents per million miles (IPMM) for the ADS vs. 2.80 IPMM for the human benchmark, an 80% reduction or a human crash rate that is 5 times higher than the ADS rate. Police-reported crashed vehicle rates for all locations together were 2.1 IPMM for the ADS vs. 4.68 IPMM for the human benchmark, a 55% reduction or a human crash rate that was 2.2 times higher than the ADS rate. Police-reported crashed vehicle rate reductions for the ADS were statistically significant when compared in San Francisco and Phoenix, as well as combined across all locations and the any-injury-reported reductions were statistically significant in San Francisco and in all locations. The any property damage or injury comparison had statistically significant decreases in 3 comparisons but also nonsignificant results in 3 other benchmarks. When excluding ADS crashes with a delta-V less than 1 mph (a measure of sensitivity to lower reporting threshold), about half of the ADS collisions were excluded, resulting in comparisons that showed a large statistically significant reduction in all comparisons except for one comparison from San Francisco. Conclusions The statistically significant reductions in police-reported and any injury reported crash rates indicate a promising positive safety impact of ADS. The direction and significance of comparisons in the any property damage or injury outcome group are inconclusive due to difficulties in estimating a matching human benchmark. More research is needed on defining any property damage or injury benchmarks with clear lower reporting thresholds to reduce the systematic uncertainty in the benchmark rates. Together, these crash rate results contribute to the continuous growth in confidence, together with other methodologies, in a safety case approach.

A note on Tesla's revised safety report crash rates

Between June 2018 and December 2023, Tesla released quarterly safety reports citing average miles between crashes for Tesla vehicles. Prior to March 2021, crash rates were categorized as (1) with their SAE Level 2 automated driving system Autopilot engaged, (2) without Autopilot but with active safety features such as automatic emergency braking, and (3) without Autopilot and without active safety features. In January 2023, Tesla revised past reports to reflect their new categories of with and without Autopilot engaged, in addition to making small adjustments based on recently discovered double counting of reports and excluding previously recorded crashes that did not meet their thresholds of airbag or active safety restraint activation. The revisions are heavily biased towards no-active-safety-features—a surprising result given prior research showing that drivers predominantly keep most active safety features enabled. As Tesla's safety reports represent the only national source of Level 2 advanced driver assistance system crash rates, clarification of their methods is essential for researchers and regulators. This note describes the changes and considers possible explanations for the discrepancies.

Using driver monitoring to estimate readiness in automation: a conceptual model based on simulator experimental data

This paper provides a theoretical overview of how the concept of driver readiness can be objectively measured, using controlled experimental data. First, a literature review regarding the concept of driver readiness is provided. Then, it highlights challenges for a standardized readiness estimation model. A conceptual readiness estimation model is presented, and a methodology is proposed for defining readiness thresholds for use by Driver State Monitoring (DSM) systems. The paper then explores how this model can be used to estimate readiness thresholds. A proof of concept for the model application is presented, using previously collected experimental involving SAE Level 2 automation. This paper contributes to the state of the art in DSM-development, by providing a methodology for estimating driver readiness, while considering variabilities across individual drivers. The model also allows readiness thresholds to be defined with data from driving simulator experiments, without relying on subjective assessment of readiness as its ground truth.

Benchmarks for retrospective automated driving system crash rate analysis using police-reported crash data

Objectives With fully automated driving systems (ADS; SAE level 4) ride-hailing services expanding in the U.S., we are now approaching an inflection point in the history of vehicle safety assessment. The process of retrospectively evaluating ADS safety impact (as seen with seatbelts, airbags, electronic stability control, etc.) can start to yield statistically credible conclusions. An ADS safety impact measurement requires a comparison to a “benchmark” crash rate. Most benchmarks generated to-date have focused on the current human-driven fleet, which enable researchers to understand the impact of the introduced ADS technology on the current crash record status quo. This study aims to address, update, and extend the existing literature by leveraging police-reported crashes to generate human crash rates for multiple geographic areas with current ADS deployments. Methods: All of the data leveraged is publicly accessible, and the benchmark determination methodology is intended to be repeatable and transparent. Generating a benchmark that is comparable to ADS crash data is associated with certain challenges, including data selection, handling underreporting and reporting thresholds, identifying the population of drivers and vehicles to compare against, choosing an appropriate severity level to assess, and matching crash and mileage exposure data. Results Consequently, we identify essential steps when generating benchmarks, and present our analyses amongst a backdrop of existing ADS benchmark literature. One analysis presented is the usage of established underreporting correction methodology to publicly available human driver police-reported data to improve comparability to publicly available ADS crash data. We also identified several important crash rate dependencies (geographic region, road type, and vehicle type), and show how failing to account for these features in ADS comparisons can bias results. Conclusions Working with police-reported crash data to create crash rate benchmarks is fraught with challenges. Researchers should be cautious in their selection of crash rate benchmarks. We present these challenges, discuss their consequences, and provide analytical guidance for addressing them. This body of work aims to contribute to the ability of the community - researchers, regulators, industry, and experts - to reach consensus on how to estimate accurate benchmarks.

Exploratory analysis of injury severity under different levels of driving automation (SAE Levels 2 and 4) using multi-source data

Vehicles equipped with automated driving capabilities have shown potential to improve safety and operations. Advanced driver assistance systems (ADAS) and automated driving systems (ADS) have been widely developed to support vehicular automation. Although the studies on the injury severity outcomes that involve automated vehicles are ongoing, there is limited research investigating the difference between injury severity outcomes for the ADAS and ADS equipped vehicles. To ensure a comprehensive analysis, a multi-source dataset that includes 1,001 ADAS crashes (SAE Level 2 vehicles) and 548 ADS crashes (SAE Level 4 vehicles) is used. Two random parameters multinomial logit models with heterogeneity in the means of random parameters are considered to gain a better understanding of the variables impacting the crash injury severity outcomes for the ADAS (SAE Level 2) and ADS (SAE Level 4) vehicles. It was found that while 67 percent of crashes involving the ADAS equipped vehicles in the dataset took place on a highway, 94 percent of crashes involving ADS took place in more urban settings. The model estimation results also reveal that the weather indicator, driver type indicator, differences in the system sophistication that are captured by both manufacture year and high/low mileage as well as rear and front contact indicators all play a role in the crash injury severity outcomes. The results offer an exploratory assessment of safety performance of the ADAS and ADS equipped vehicles using the real-world data and can be used by the manufacturers and other stakeholders to dictate the direction of their deployment and usage.

AUTOMATED DRIVING SYSTEMS: COMPARISON OF TRAINING METHODS’ EFFECTIVENESS

The second decade of the 21st century saw the introduction of vehicles equipped with various driver assistance systems. However, the producers' plans go much further. In January 2023, Mercedes-Benz's DRIVE PILOT system was approved for use on public roads in Nevada (United States) as the first and so far only SAE Level 3 series-production solution. Despite increasing automation, the human factor remains important for the safety and efficiency of road transport. In the coming years, and possibly decades, this will mainly be due to the co-existence of vehicles with varying degrees of automation and necessary driver-vehicle interactions in cases where an automated driving system crosses the boundaries of the operational design domain. The latest research reveals the limitations of driving assistance systems, pointing not only to the need for changes in the technological domain but also to the need to pay more attention to the psychological aspects of using assistance systems in cognitive and behavioral areas. Therefore, an additional element of automation implementation should be considered, which is driver training focused on modern technological solutions. Properly arranged and conducted training can be treated as an additional factor facilitating the implementation of automated mobility. The article describes the methods of assessing the effectiveness of training in terms of familiarizing drivers with the functions that automate driving. The authors present the results of tests carried out on a driving simulator with an implemented Level 4 system (according to SAE classification). Three forms of knowledge transfer were analyzed: practical training, e-learning, and training with a short user manual. The results were compared in terms of driver-vehicle interaction effectiveness (e.g., system activation), ensuring that the driver is sufficiently reactive in dangerous situations and understands the limitations of the system. The research was conducted on a group of 81 drivers aged 18–65. Driver training was one of the main pillars of the Trustonomy project, which received funding from the EU research and innovation program Horizon 2020 under grant agreement No. 815003.

Investigating levels of remote operation in high-level on-road autonomous vehicles using operator sequence diagrams

The continuing development of autonomous vehicle technology is making the presence of fully autonomous vehicles (SAE Level 5 of Driving Automation) on the road an ever more likely possibility. Similarly, regulation changes show countries are preparing for autonomous vehicles to increase their presence on public roads for both testing and use after sale. With this in mind, solutions to the problem of disengagement from the autonomous driving system by Level 5 vehicles, due to damage, operation outside of expected parameters or software failure among other reasons are being investigated including remote operation. This research aims to give evidence for the inclusion of remote operation into the autonomous driving and define the types of remote operation that may occur from existing literature. The four types of remote operation are Remote Monitoring, Remote Assistance, Remote Management and Remote Driving. Operator sequence diagrams are used to evaluate these types of remote operation in likely scenarios they may occur and draw conclusions about the role and the tasks the operator will be required to complete.

Cooperative truck platooning trial on Canadian public highway under commercial operation in winter driving conditions

Cooperative truck platooning, a convoy of trucks driving together while communicating and coordinating with each other, represents a technology-driven approach to improve energy conversion efficiency, lower greenhouse gas emissions, and enhance road safety. Despite numerous studies have explored these potentials, there is a scarcity of empirical investigations into on-road cooperative truck platooning during commercial operations, particularly in winter driving conditions. This paper presents the findings of an experimental study on the first commercially focused truck platooning implementation on a Canadian public highway in the winter season, using two SAE level 2 class 8 trucks. The on-road trials took place on the Queen Elizabeth II Highway, between Calgary and Edmonton, with ambient temperatures ranging from −27°C to 12°C, and truck weights spanning 16–39 tons. Nine well-trained and experienced drivers conducted 41 incident-free (platooning and baseline) test trips, covering a distance of 22,855 km. The experimental results confirmed the feasibility of operating commercial truck platooning with 3–5 s time gaps on public roads during the Canadian winter season including various road surface conditions. The results also show that the platooning engagement ratio reached up to 88.9%, with an average of 61.6% across 25 platooning trips. Furthermore, the follower truck achieved a 1.6% fuel savings on flat road sections during platooning, but its freight transportation specific fuel consumption was higher than that of the lead truck on hilly terrain. Test results indicate the lighter truck exhibited higher specific nitrogen oxides (NOx) emissions. Moreover, the frequent engagement and disengagement of the cooperative truck platooning system had adverse effects on the powertrain system of the truck, leading to increased fuel consumption and engine-out NOx emissions. This study provides real-world data to identify limitations and needed areas for improvement in adapting cooperative truck platooning technology to commercial operations on public roads.

Aiding Automated Shuttles with Their Driving Tasks as an On-Board Operator: A Case Study on Different Automated Driving Systems in Three Living Labs

Highly automated shuttle vehicles (SAE Level 4) have the potential to enhance public transport services by decreasing the demand for drivers, enabling more frequent and flexible ride options. However, at least in a transitionary phase, safety operators that supervise and support the shuttles with their driving tasks may be required on board the vehicle from a technical or legal point of view. A crucial component for executing supervisory and intervening tasks is the human–machine interface between an automated vehicle and its on-board operator. This research presents in-depth case studies from three heterogenous living laboratories in Germany that deployed highly automated shuttle vehicles with on-board operators on public roads. The living labs differed significantly regarding the on-board operators’ tasks and the design of the human–machine interfaces. Originally considered a provisional solution until the vehicle automation is fully capable of running without human support, these interfaces were, in general, not designed in a user-centered way. However, since technological progress has been slower than expected, on-board operator interfaces are likely to persist in the mid-term at least. Hence, this research aims to assess the aptitude of interfaces that are in practical use for the on-board operators’ tasks, in order to determine the user-centered design of future interfaces. Completing questionnaires and undergoing comprehensive, semi-structured interviews, nine on-board operators evaluated their human–machine interfaces in light of the respective tasks they complete regarding user variables such as work context, acceptance, system transparency, and trust. The results were highly diverse across laboratories and underlined that the concrete system setup, encompassing task and interface design, has a considerable impact on these variables. Ergonomics, physical demand, and system transparency were identified as the most significant deficits. These findings and derived recommendations may inform the design of on-board operator workspaces, and bear implications for remote operation workstations as well.

Development of an Integrated Longitudinal Control Algorithm for Autonomous Mobility with EEG-Based Driver Status Classification and Safety Index

During unexpected driving situations in autonomous vehicles, such as a system failure, the driver should take over control from the vehicles in SAE Level 3 to cope with unexpected situations. Therefore, it is necessary to develop reasonable takeover technologies to ensure safe driving. In this study, an electroencephalogram (EEG)-based driver status classification model and a safety index-based integrated longitudinal control algorithm considering the takeover time and driving characteristics are proposed. The driver status is classified into two states: road monitoring and non-driving-related tasks. EEG data are acquired while the driver performs certain tasks. The driver status classification model is presented using the EEG data based on a machine learning method. It is designed such that the desired takeover time is determined based on the classified driver state. To design the integrated longitudinal control algorithm, a safety index is designed and calculated based on the vehicle state and driver’s driving characteristics. The desired clearances based on the desired takeover time and driver characteristics are calculated and selected based on the safety index. A sliding-mode control algorithm is adopted to allow the vehicle to track the desired clearance reasonably. The performance of the proposed control algorithm is evaluated using the MATLAB/Simulink R2019a (Mathworks, Natick, Massachusetts, U.S.A) and CarMaker software 8.1.1 (IPG Automotive, Karlsruhe, Germany).

A multimodal driver monitoring benchmark dataset for driver modeling in assisted driving automation

In driver monitoring various data types are collected from drivers and used for interpreting, modeling, and predicting driver behavior, and designing interactions. Aim of this contribution is to introduce manD 1.0, a multimodal dataset that can be used as a benchmark for driver monitoring in the context of automated driving. manD is the short form of human dimension in automated driving. manD 1.0 refers to a dataset that contains data from multiple driver monitoring sensors collected from 50 participants, gender-balanced, aged between 21 to 65 years. They drove through five different driving scenarios in a static driving simulator under controlled laboratory conditions. The automation level (SAE International, Standard J3016) ranged from SAE L0 (no automation, manual) to SAE L3 (conditional automation, temporal). To capture data reflecting various mental and physical states of the subjects, the scenarios encompassed a range of distinct driving events and conditions. manD 1.0 includes environmental data such as traffic and weather conditions, vehicle data like the SAE level and driving parameters, and driver state that covers physiology, body movements, activities, gaze, and facial information, all synchronized. This dataset supports applications like data-driven modeling, prediction of driver reactions, crafting of interaction strategies, and research into motion sickness.