How to keep drivers engaged while supervising driving automation? A literature survey and categorisation of six solution areas

Abstract

This work aimed to organise recommendations for keeping people engaged during human supervision of driving automation, encouraging a safe and acceptable introduction of automated driving systems. First, heuristic knowledge of human factors, ergonomics, and psychological theory was used to propose solution areas to human supervisory control problems of sustained attention. Driving and non-driving research examples were drawn to substantiate the solution areas. Automotive manufacturers might (1) avoid this supervisory role altogether, (2) reduce it in objective ways or (3) alter its subjective experiences, (4) utilize conditioning learning principles such as with gamification and/or selection/training techniques, (5) support internal driver cognitive processes and mental models and/or (6) leverage externally situated information regarding relations between the driver, the driving task, and the driving environment. Second, a cross-domain literature survey of influential human-automation interaction research was conducted for how to keep engagement/attention in supervisory control. The solution areas (via numeric theme codes) were found to be reliably applied from independent rater categorisations of research recommendations. Areas (5) and (6) were addressed by around 70% or more of the studies, areas (2) and (4) in around 50% of the studies, and areas (3) and (1) in less than around 20% and 5%, respectively. The present contribution offers a guiding organisational framework towards improving human attention while supervising driving automation.     

Effects of higher-order driving skill training on young, inexperienced drivers' on-road driving performance

The aim of the current study was to compare the effects of training in higher-order driving skills (e.g., perceptual, motivational, insight) and vehicle handling skill training in relation to on-road driving performance, hazard perception, attitudes to risky driving and driver confidence levels in young, inexperienced drivers. Thirty-six young drivers (23 males and 13 females, average age 16.3 years), mostly on a restricted NZ driver licence, participated in a Driver Training Research camp. Participants were randomly allocated to one of three equally sized groups according to the type of driving skill training (5 days) they received: higher-order, vehicle handling or control (no training). Professional driver assessors conducted a comprehensive driving assessment before (Baseline) and after the training (Post Training). At both time points, participants also carried out a computerised hazard perception task, and completed self-report questionnaires to assess attitudes to risky driving and driver confidence. In terms of on road driving, the participants who received higher-order driving skill training showed a statistically significant improvement in relation to visual search and the composite driving measure. This was accompanied by an improvement in hazard perception, safer attitudes to close following and to dangerous overtaking and a decrease in driving related confidence. The participants who received vehicle handling skill training showed significant improvements in relation to their on-road direction control, speed choice and the composite driving score. However, this group showed no improvement in hazard perception, attitudes to risky driving or driver confidence. The findings will be discussed in the context of driver training as a viable crash prevention intervention in regard to young, inexperienced drivers.     

Statistical transfer learning: A review and some extensions to statistical process control

ABSTRACT

The rapid development of information technology, together with advances in sensory and data acquisition techniques, has led to the increasing necessity of handling datasets from multiple domains. In recent years, transfer learning has emerged as an effective framework for tackling related tasks in target domains by transferring previously-acquired knowledge from source domains. Statistical models and methodologies are widely involved in transfer learning and play a critical role, which, however, has not been emphasized in most surveys of transfer learning. In this article, we conduct a comprehensive literature review on statistical transfer learning, i.e., transfer learning techniques with a focus on statistical models and statistical methodologies, demonstrating how statistics can be used in transfer learning. In addition, we highlight opportunities for the use of statistical transfer learning to improve statistical process control and quality control. Several potential future issues in statistical transfer learning are discussed.     

Predictors of injury among younger and older adults in fatal motor vehicle crashes

With 2003 Fatality Analysis Reporting System data, we examined relationships among predictors of motor vehicle injury/fatality outcomes for younger (35–54 years) and older (65 years and older) drivers. Using the Precede-Proceed Model of Health Promotion as an organizing framework, we classified variables into person, vehicle and environment domains and conducted a multinomial logistic regression.Significant risk factors for older driver injuries were impact crashes at 1–3 o’clock (OR = 1.65; CI: 1.05–2.59), 7–9 o’clock angles (OR = 2.59; CI = 1.45–4.63), and driving with one passenger (OR = 2.25; CI: 1.58–3.20). Previous other motor vehicle convictions were significantly associated with reduced risk of injury (OR = 0.55; CI = 0.34–0.90). The 7–9 o’clock angle (OR = 3.06; CI: 1.83–5.12), and driving in daylight hours were risk factors for fatality among older drivers.Many risk factors (e.g., female gender, non-seatbelt use, rollover crashes, and vehicle body type), and protective factors (e.g., number of lanes and non-airbag deployment) were relevant for younger and older drivers. Findings showed relevant factors for drivers from both age groups, with some pointing to older adults, and set the stage for further research to develop injury and fatality prevention programs.     

Three decades of workload control research: a systematic review of the literature

The workload control (WLC) concept has received much attention in the past three decades; however, a comprehensive literature review has not been presented. In response, this article provides a systematic review of the conceptual, analytical, empirical and simulation-based WLC literature. It explores the evolution of WLC research, determines the current state-of-the-art and identifies key areas for further study. The research finds that the field has evolved substantially. Early research focused on theoretical development and experimental testing of order release strategies; order release was then integrated with other planning stages, e.g., the customer enquiry stage, making the concept more suitable for customised manufacturing and leading to a comprehensive concept which combines input and output control effectively; recent attention has focused on implementing the resulting concept in practice and refining theory. While WLC is well placed to meet the needs of producers of customised products, future research should include: conducting further action research into how WLC can be effectively implemented in practice; studying human factors that affect WLC; and feeding back empirical findings to simulation-based WLC research to improve the applicability of WLC theory to real-life job shops.     

Resource extension functions: properties, inversion, and generalization to segments

The unified modeling and solution framework, presented by Desaulniers et al. (Fleet Management and Logistics. Kluwer Academic, Boston, pp 57–93, 1998), is applicable to nearly all types of vehicle-routing and crew-scheduling problems found in the literature thus far. The framework utilizes resource extension functions (REFs) as its main tool for handling complex side constraints that relate to a single vehicle route or crew schedule. The intention of this paper is to clarify which properties of REFs allow important algorithmic procedures, such as efficient representation of (partial) paths, efficient cost computations, and constant time feasibility checking for partial paths (= segments) and their concatenations. The theoretical results provided by the paper are useful for developing highly efficient solution methods for both exact and heuristic approaches. Acceleration techniques for solving resource-constrained shortest-path subproblems are a key success factor for those exact algorithms which are based on column generation or Lagrangean relaxation. Similarly, those heuristic algorithms which are based on resource-constrained paths can benefit from efficient operations needed to construct or manipulate segments. Fast operations are indispensable for efficient local-search algorithms that explore edge-exchange or node-exchange neighborhoods. Efficiency is crucial, since these operations are repeatedly performed in many types of metaheuristics.     

A survey of semiconductor supply chain models Part II: demand planning,inventory management,and capacity planning

Part I of this three-part series described semiconductor supply chains from the decision-making and functional perspectives, using this as a framework to review the industrial engineering and operations research literature on the problems arising in these supply chains. Part I then reviewed the literature on Strategic Network Design, supply chain coordination, sustainability and simulation-based decision support. This paper, Part II, reviews the areas of Demand Planning, Inventory Management and Capacity Planning in semiconductor supply chains. Part III concludes the series by discussing models to support Master Planning, Production Planning and Demand Fulfilment in this industry.     

Reliability-based evaluation of automotive wind noise quality

This paper proposes an efficient method for the reliability analysis of a vehicle body–door subsystem with respect to an important quality issue—wind noise. A nonlinear seal model is constructed for the automotive wind noise problem and the limit state function is evaluated using finite element analysis. A multi-modal adaptive importance sampling (AIS) method is developed for reliability analysis at system level, to improve the efficiency of the Monte Carlo simulation. The method can easily handle implicit and time-dependent limit-state functions, with variables of any statistical distributions. Existing analytical as well as simulation-based methods are also investigated to solve the car door wind noise problem. It is demonstrated through this industrial application problem that the multi-modal AIS method is superior to existing methods in terms of efficiency and accuracy. A generalized framework for reliability estimation is then established for series system reliability problems with large numbers of random variables and complicated, implicit limit states.     

Cognitive failures as predictors of driving errors, lapses, and violations

Dual-process models from the cognitive literature have proposed a taxonomy of cognitive failures in everyday activities, and this novel approach was applied to understanding driver behaviour. This framework was used to examine whether categories of cognitive failure would explain driving errors, driving lapses, and driving violations in a sample of undergraduates at a large urban university. Two types of cognitive failure were examined, one associated with missing affective information and the other associated with a failure to engage effortful processes to override an automatic response. Alexithymia was used as an indicator of missing affective information, and attention regulation, reactivity, and impulsivity were used as indicators of override failure. Relevant demographic variables included gender and hours typically driven. Override failures were significantly associated with driving behaviour in the correlational analyses. In the regression analyses, attention regulation predicted driving errors, and gender, attention regulation, and impulsivity predicted driving violations. The implications of this work include the potential application to driver training, to users of informatics devices (e.g., GPS, cellular phones, messaging systems), and for individuals diagnosed with attention and/or impulsivity problems.     

Getting the driver back into the loop: the quality of manual vehicle control following long and short non-critical transfer-of-control requests: TI:NS

ABSTRACT

Specific vehicle automation use-cases such as traffic jams will be the first level 3 functions on the market. When the ‘traffic jam pilot’ nears its limits in non-critical situations, control needs to be handed back to the driver, enabling appropriate situation awareness (SA) and vehicle handling. According to previous research, operational vehicle stabilisation can be achieved within a transfer-of-control (TOC) of a few seconds in simple traffic environments, but tactical level decisions benefit from longer hand-over times. To date, the effects of non-critical TOCs have not been studied using set time frames. To investigate the impact of short (unplanned, 5 seconds) and long (planned, 50 seconds) TOC requests, while playing/not playing an engaging tablet game, a simulator experiment was conducted with 16 participants. Comparisons of the 60-second-period of manual driving following automation suggest better longitudinal vehicle control as well as more appropriate SA following the long TOC request, and automation periods without the game. However, following no engaging game, lateral performance was worse during the first 10 seconds of manual driving. Control-level visual search patterns did not change with TOC time or the game. Future research needs to consider support for drivers’ SA maintenance and readiness to drive following high automation.     

Driving with pets and motor vehicle collision involvement among older drivers: A prospective population-based study

ObjectiveDistracted driving is a major cause of motor vehicle collision (MVC) involvement. Pets have been identified as potential distraction to drivers, particularly in the front. This type of distraction could be worse for those with impairment in the cognitive aspects of visual processing. The purpose of this study is to evaluate the association between driving with pets and rates of motor vehicle collision involvement in a cohort of older drivers.MethodsA three-year prospective study was conducted in a population-based sample of 2000 licensed drivers aged 70 years and older. At the baseline visit, a trained interviewer asked participants about pet ownership, whether they drive with pets, how frequently, and where the pet sits in the vehicle. Motor vehicle collision (MVC) involvement during the three-year study period was obtained from the Alabama Department of Public Safety. At-fault status was determined by the police officer who arrived on the scene. Participants were followed until the earliest of death, driving cessation, or end of the study period. Poisson regression was used to calculate crude and adjusted rate ratios (RR) examining the association between pet ownership, presence of a pet in a vehicle, frequency of driving with a pet, and location of the pet inside with vehicle with any and at-fault MVC involvement. We examined whether the associations differed by higher order visual processing impairment status, as measured by Useful Field of View, Trails B, and Motor-free Visual Perception Test.ResultsRates of crash involvement were similar for older adults who have ever driven with a pet compared to those who never drove with their pet (RR = 1.15, 95% CI 0.76–1.75). Drivers who reported always or sometimes driving with their pet had higher MVC rates compared to pet owners who never drive with a pet, but this association was not statistically significant (RR = 1.39, 95% CI 0.86–2.24). In terms of location, those reporting having a pet frequently ride in the front of the vehicle had similar rates of MVC involvement compared to those who did not drive with a pet in the front. A similar pattern of results was observed for at-fault MVCs. This association was not modified by visual processing impairment status.ConclusionThe current study demonstrates a positive but non-significant association between frequently driving with pets and MVC involvement. More research is needed, particularly on restraint use and whether the pet was in the car at the time of the crash, to help characterize the public safety benefit of regulations on driving with pets.     

Simulation-based optimisation of logistics distribution system for an assembly line with path constraints

The design of logistics distribution system for an assembly line with given layout is usually constrained by various factors such as the vehicles for the distribution of assembly components and the paths on the shop floor for the vehicle movement. Since design optimisation of these production systems with such constraints is hard to solve by mathematic approaches, simulation-based approach is mostly used. In this paper, simulation-based analysis and optimisation of a logistic distribution system is performed combined with a heuristic algorithm to obtain a solution that most suits the practical requirements of an assembly line. The proposed simulation-based method builds a model that satisfied various complicated real-world problems at the work site of an investigated factory. Meanwhile, a dynamic dispatching method for vehicle movement is also presented which can adjust the control strategy and decision parameters dynamically in the process of simulation and contributes significantly to form a more fruitful design scheme. Experimental results show that proposed method performs better than the analytical model-based approach and some other methods, and the dynamic vehicle dispatching method presented performs better than some existing common strategies used in literature.     

基于控制策略数据的ACC切入场景下HMI设计研究

目的 通过基于控制策略数据的智能汽车自适应巡航控制功能的人机界面设计研究、实践和评估,为自动驾驶汽车中的智能功能HMI设计提供参考方法和指导.方法 首先通过大量查阅文献资料,了解ACC功能在切入场景下的运行逻辑和控制策略参数.然后结合驾驶员操作及认知任务,抽离出汽车HMI中进行展示的数据信息.最后通过信息架构设计、基础元素设计和多屏互动方案设计等设计分析和实践,对切入场景下汽车多屏HMI进行设计.结果 对ACC功能切入场景下的设计方案进行了可用性测试评估,结果显示ACC的HMI设计方案逻辑清晰,有较高的可用性.结论 基于控制策略数据的HMI设计,能有效帮助设计人员进行复杂汽车智能功能的人机交互设计,提高HMI可用性.     

Exploring precrash maneuvers using classification trees and random forests

Taking evasive actions vis-à-vis critical traffic situations impending to motor vehicle crashes endows drivers an opportunity to avoid the crash occurrence or at least diminish its severity. This study explores the drivers, vehicles, and environments’ characteristics associated with crash avoidance maneuvers (i.e., evasive actions or no evasive actions). Rear-end collisions, head-on collisions, and angle collisions are analyzed separately using decision trees and the significance of the variables on the binary response variable (evasive actions or no evasive actions) is determined. Moreover, the random forests method is employed to rank the importance of the drivers/vehicles/environments characteristics on crash avoidance maneuvers. According to the exploratory analyses’ results, drivers’ visibility obstruction, drivers’ physical impairment, drivers’ distraction are associated with crash avoidance maneuvers in all three types of accidents. Moreover, speed limit is associated with rear-end collisions’ avoidance maneuvers and vehicle type is correlated with head-on collisions and angle collisions’ avoidance maneuvers. It is recommended that future research investigates further the explored trends (e.g., physically impaired drivers, visibility obstruction) using driving simulators which may help in legislative initiatives and in-vehicle technology recommendations.     

Reliability analysis of automotive body-door subsystem

This paper proposes an efficient method for the reliability analysis of a vehicle body-door subsystem with respect to one of the important quality issues—the door closing energy. The developed method combines optimization-based and simulation-based approaches and is particularly applicable for problems with highly non-linear and implicit limit state functions. The proposed approach consists of two major parts. In the first part, an optimization-based method is used to search for the most probable point (MPP) on the limit state. This is achieved by using an adaptive response surface constructed through an optimal symmetric Latin hypercube design of experiments. In the second part, a multi-modal adaptive importance sampling method is proposed using the MPP information from the first part as the starting point. It is demonstrated through numerical examples that the proposed method is superior to existing methods in terms of efficiency and accuracy. The proposed method is illustrated for application to the reliability estimation with respect to the door closing energy problem. A generalized framework for reliability estimation is then established for problems with large numbers of random variables and complicated limit states.     

A holistic approach for assessing traffic safety in the United Arab Emirates

Traffic accident and fatality rates can be utilized as indicators of traffic safety, but cannot reflect the overall status of traffic safety in a country. This paper uses a holistic perspective approach to investigate traffic safety in the United Arab Emirates (UAE). Initially, 12 potential items were selected to investigate the issue of traffic safety in the country. The investigation included data collection and analyses from official police reports, survey among road-users and interview of traffic safety experts. Based on data analysis and interpretation, the main factors affecting traffic safety in the UAE along with their level of deficiency were identified. The study revealed that the main factors contributing to traffic safety in the UAE are driving behaviour, awareness, education and training, infrastructure, vehicle, law enforcement, coordination and quality of resources. Among these factors, a major deficiency was found in the “driving behaviour”, a minor deficiency in “vehicle safety”, and a moderate deficiency in the others. Based on the deficiency level of the factors recommendations were proposed to improve the status of traffic safety in the country.     

A multi-level model on automated vehicle acceptance (MAVA): a review-based study

Abstract

Automated vehicle acceptance (AVA) is a necessary condition for the realisation of higher-level objectives such as improvements in road safety, reductions in traffic congestion and environmental pollution. On the basis of a systematic literature review of 124 empirical studies, the present study proposes MAVA, a multi-level model to predict AVA. It incorporates a process-oriented view on AVA, considering acceptance as the result of a four-stage decision-making process that ranges from the exposure of the individual to automated vehicles (AVs) in Stage 1, the formation of favourable or unfavourable attitudes towards AVs in Stage 2, making the decision to adopt or reject AVs in Stage 3, to the implementation of AVs into practice in Stage 4. MAVA incorporates 28 acceptance factors that represent seven main acceptance classes. The acceptance factors are located at two levels, i.e., micro and meso. Factors at the micro-level constitute individual difference factors (i.e., socio-demographics, personality and travel behaviour). The meso-level captures the exposure of individuals to AVs, instrumental domain-specific, symbolic-affective and moral-normative factors of AVA. The literature review revealed that 6% of the studies investigated the exposure of individuals to AVs (i.e., knowledge and experience). 22% of the studies investigated domain-specific factors (i.e., performance and effort ­expectancy, safety, facilitating conditions, and service and vehicle ­characteristics), 4% symbolic-affective factors (i.e., hedonic motivation and social influence), and 12% moral-normative factors (i.e., perceived benefits and risks). Factors related to a person’s socio-demographic profile, travel behaviour and personality were investigated by 28%, 15% and 14% of the studies, respectively. We recommend that future studies empirically verify MAVA using longitudinal or experimental studies.     

A method to assess the driver mental workload: The driving activity load index (DALI)

Intelligent transport system (ITS) can induce improvement in road safety, as long as acceptability by user?s population and adequacy with driving task demands have been checked. The Human-Centred Design process allows taking into account the diversified driver?s population needs and functional abilities in addition to the complexity of the driving context, for setting up design recommendations and for the development of methodologies for safety assessment. Evaluation of the driver?s mental workload is an important parameter, complementary to objective ones such as control of the vehicle and driver?s visual strategies. This paper describes experiments conducted in the framework of the European project AIDE aiming at validating the DALI (Driving Activity Load Index), a tool set up to allow the evaluation of driver?s mental workload while driving, with or without the support of in-vehicle systems. The main results and conclusion from the conducted analysis are presented.     

On the training and testing of entry-level commercial motor vehicle drivers

This study examined the effectiveness of 3 different training types on commercial motor vehicle (CMV) drivers’ skill levels. The training types included a conventional 8-week certified course, a conventional 8-week certified course with approximately 60% of driving time spent in a CMV driving simulator, and a Commercial Driver's License (CDL) test focused short course. Participants’ scores on the Division of Motor Vehicles (DMV) road and range tests were assessed. In addition to their DMV scores, participants replicated DMV road and range driving tests in an instrumented vehicle and the CMV driving simulator. Results indicated no training group differences in DMV road tests. There were differences between training groups on DMV range tests and real truck and simulator versions of the DMV road and range tests; on these tests conventional- and simulator-trained participants generally scored higher than CDL-focused participants. However, all groups performed higher in the real truck than in the simulator for both road and range tests. These findings indicate the need for a minimum standard of entry-level CMV driver training as well as support of the use of a driving simulator for training entry-level drivers; however, testing using a simulator does not appear to be feasible with current technology.