A human factors perspective on automated driving

ABSTRACT

Automated driving can fundamentally change road transportation and improve quality of life. However, at present, the role of humans in automated vehicles (AVs) is not clearly established. Interviews were conducted in April and May 2015 with 12 expert researchers in the field of human factors (HFs) of automated driving to identify commonalities and distinctive perspectives regarding HF challenges in the development of AVs. The experts indicated that an AV up to SAE Level 4 should inform its driver about the AV's capabilities and operational status, and ensure safety while changing between automated and manual modes. HF research should particularly address interactions between AVs, human drivers and vulnerable road users. Additionally, driver-training programmes may have to be modified to ensure that humans are capable of using AVs. Finally, a reflection on the interviews is provided, showing discordance between the interviewees’ statements – which appear to be in line with a long history of HFs research – and the rapid development of automation technology. We expect our perspective to be instrumental for stakeholders involved in AV development and instructive to other parties.     

Cooperation between drivers and automation: implications for safety

The purpose of this paper is to develop a recently proposed framework of human–machine cooperation (Hoc, J.M., 2001. Towards a cognitive approach to human–machine cooperation in dynamic situations. International Journal of Human–Computer Studies, 54, 509–540) and apply it to the domain of in-car automation. Previous models of automation (e.g. Sheridan, T.B. and Verplanck, W.L., 1978. Human and computer control of undersea teleoperators. Cambridge, MA: MIT Man–Machine Systems Laboratory) delineate the roles of human and machine in a task-based manner and primarily from the viewpoint of machine requirements. However, with increasing arguments that automation should support the human operator rather than replace them (e.g. Young, M.S., Stanton, N.A., and Harris, D., 2007. Driving automation: learning from aviation about design philosophies. International Journal of Vehicle Design, 45(3), 323–338), Hoc's (2001) framework offers a means of modelling the interaction from the perspective of teamwork–that is, from the viewpoint of human requirements. In the present context, the framework describes how both the driver and the automation can be considered as separate agents pursuing their own goals while trying to facilitate a common task, but who may interfere with each other positively (e.g. mutual control) or negatively (mutual conflict). Vehicle automation, as an area of fervent research in ergonomics at present, provides the opportunity to explore the framework and use it to interpret current and emerging research findings. It is suggested here that many of the psychological problems underlying the introduction of in-car automation are a result of suboptimal communications between human and machine, and the framework is used to propose directions for future research in this area.     

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.     

Human–machine interaction theories and lane departure warnings

A technological race toward always ever-increasing automation is engaged, a race in which automation users are hardly considered. This explains why more automation is not necessarily for the best. A variety of human–machine interaction (HMI) theories co-exist. First, those HMI theories and predictions for lane departure warnings systems (LDWS, assisted driving automation) are described. Second, a synthesis of the key questions currently addressed by empirical data on LDWS is offered. Lastly, a new model of human–machine cooperation modelling is proposed. The model has been inspired by previous theories and empirical data collected with LDWS-assisted driving. Interestingly, automation such as LDWS seems to avoid the ‘ironies of automation’ with no negative effects on human performance. However, a major issue to be addressed is poor automation acceptance. While the focus was set on LDWS to offer a complete overview for this type of device, the model may be extended to other warning assistance devices.     

Human behaviour with automated driving systems: a quantitative framework for meaningful human control

Abstract

Automated driving systems (ADS) with partial automation are currently available for the consumer. They are potentially beneficial to traffic flow, fuel consumption, and safety, but human behaviour whilst driving with ADS is poorly understood. Human behaviour is currently expected to lead to dangerous circumstances as ADS could place human drivers ‘out-of-the-loop’ or cause other types of adverse behavioural adaptation. This article introduces the concept of ‘meaningful human control’ to better address the challenges raised by ADS, and presents a new framework of human control over ADS by means of literature-based categorisation. Using standards set by European authorities for driver skills and road rules, this framework offers a unique, quantified perspective into the effects of ADS on human behaviour. One main result is a rapid and inconsistent decrease in required skill- and rule-based behaviour mismatching with the increasing amount of required knowledge-based behaviour. Furthermore, the development of higher levels of automation currently requires different human behaviour than feasible, as a mismatch between supply and demand in terms of behaviour arises. Implications, discrepancies and emerging mismatches this framework elicits are discussed, and recommendations towards future design strategies and research opportunities are made to provide a meaningful transition of human control over ADS.     

LED高速高精度自动测试分拣设备研制

LED芯片制备、封装中的自动测试与分拣关键设备仍然依赖国外产品.通过分析LED自动测试与分拣设备实现高速高精度的关键问题,介绍了全自动LED高速高精度自动测试分拣设备研制,包括基于PCI总线技术的高速测试单元和自动测试与分拣机械系统.通过实际测量与试验,工作节拍小于150 ms,包括50 ms测试时间,整机效率达到24 K/h.     

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.     

Car driver attitudes,perceptions of social norms and aggressive driving behaviour towards cyclists

The interaction of car drivers and cyclists is one of the main causes of cycle incidents. The role of attitudes and social norms in shaping car drivers’ aggressive behaviour towards cyclists, is not well understood and merits investigation. A sample of 276 drivers completed an online questionnaire concerning their attitudes towards cyclists, attitudes towards risky driving, perception of social norms concerning aggressive driving towards cyclists, and the frequency with which they engage in such aggressive driving behaviours. The results showed that attitudes towards cyclists, as well as social norm perceptions concerning aggressive driving towards cyclists, were associated with aggressive driving towards cyclists. Negative attitudes towards cyclists were more pronounced in non-cyclists than cyclists and their association with aggressive driving behaviour was stronger in cyclists than non-cyclists. The perception of social norms concerning aggressive driving towards cyclists had a stronger association with aggressive driving in non-cyclists than cyclists. Attitudes towards risk taking did not affect aggressive driving towards cyclists. These findings can inform campaigns that aim to improve cyclist and car driver interaction on the roads, making them safer to use for cyclists.     

Psychological constructs in driving automation: a consensus model and critical comment on construct proliferation

As automation in vehicles becomes more prevalent, the call for understanding the behaviour of the driver while driving an automated vehicle becomes more salient. Although a variety of driver behaviour models exist, and various psychological constructs have been said to be influenced by automation, an empirically testable psychological model of automated driving has yet to be developed. Building upon Stanton and Young's model of driving automation, this article presents an updated model of interrelated psychological constructs. The proposed model was created based upon a systematic literature search of driving automation papers and a subsequent quantification of the number of reported links between a selected set of psychological constructs. A secondary aim of this article is to reach consensus in the use of psychological constructs regarding driving automation. Henceforth special attention is paid to resolving the issue of construct proliferation.     

Objective and subjective evaluation of motor priming and warning systems applied to lateral control assistance

Previous research has shown that a device called “motor priming” (MP) was more effective than other lane departure warning systems. MP prompts drivers to take action by means of small asymmetric oscillations of the steering wheel. The first objective of this experiment was to provide a deeper understanding of MP mechanisms through a series of comparisons with other haptic and auditory systems. The results suggest that much of the improvement in recovery manoeuvres observed with MP is due to the motor cue (proprioceptive pre-activation of the gesture). Other factors, such as delivering the signal directly to the hands (stimulation of response effectors) or using the tactile modality rather than auditory warning, play a lesser role. This supports the hypothesis that MP devices directly intervene at the motor level, in contrast to more traditional warning systems, which only improve situation diagnosis. The second objective was to assess drivers’ acceptance of the assistance devices. A dissociation between efficiency and acceptance of the devices was observed: drivers globally preferred auditory warning to MP. The combination of auditory warning and motor priming appeared to be a good compromise to achieve both effectiveness and acceptance. This experiment illustrates the relevance of simulator studies when dangerous situations are the main targets of the investigation.     

An automated machine for inducing optimal biosynthesis of resveratrol from peanut kernels

Abstract

Resveratrol is a type of phytoalexin, which is produced by plants in response to injury or attack. Higher yields of resveratrol are obtained through soaking, slicing, and bioreacting. Currently, no automated processing systems for producing more resveratrol and stilbenoid have been developed. This study presents a comprehensive and complete automated machine for producing more resveratrol consistently. It is comprised of a soaker, a slicer, and a bioreactor for controlling the conditions of the environment and a grinder. Several parameters are considered to optimize the whole system. The temperature, humidity, and CO₂ concentrations were 29.5 °C, 85%, and 815–850 ppm, respectively. From experiment results, the germination rate of the soaked peanut kernels reached 91.1% at 29.5 °C. The integrity ratio of the peanut kernels slicer was 95.2%, and the throughput of the slicer was 4.2 kg/h. High-performance liquid chromatography analysis showed that the quantities of resveratrol produced from mechanically treated peanut slices match those from manually treated peanut slices. Furthermore, scanning electron microscopy showed that the mechanically produced peanut slices exhibited smoother surface structures and more uniform thickness.     

A conceptual framework of autonomous and automated agents

Recent human-automation interaction research has confused concepts of automation and autonomy and has critiqued theories of automation in human-systems in terms of aspects of autonomy. This situation has led to inappropriate expectation for design and misdirected criticism of design methods. The situation is not new and has origins in historical human factors research. I differentiate the concepts of automation and autonomy with a new framework of agents. The framework is complemented by observations on characteristics of automated vs. autonomous systems, identification of error and failure modes, and formulation of a matrix of design constraints dictating possible applications of each type of agent. Discussion is provided on levels of automation, which have also been criticized in the literature, along with coverage of types of autonomy. A definition of autonomy is mutated throughout the paper to a form with utility for engineering. In general, demands of automated agents on the human-task-environment system should be absent from design of autonomous agents and design of automated systems is always automation-centric despite best efforts at human-centred approaches. Key requirements of design for autonomy include: agent viability in a target context, agent self-governance in goal formulation and fulfilment of roles, and independence in defined tasks performance.     

Effect of consecutive driving on accident risk: A comparison between passenger and freight train driving

This study combined driver-responsible accidents with on-board driving hours to examine the effect of consecutive driving on the accident risk of train operations. The data collected from the Taiwan Railway Administration for the period 1996–2006 was used to compute accident rates for varied accumulated driving hours for passenger and freight trains. The results showed that accident risk grew with increased consecutive driving hours for both passenger and freight trains, and doubled that of the first hour after four consecutive hours of driving. Additional accident risk was found for freight trains during the first hour due to required shunting in the marshalling yards where there are complex track layouts and semi-automatic traffic controls. Also, accident risk for train driving increased more quickly over consecutive driving hours than for automobile driving, and accumulated fatigue caused by high working pressure and monotony of the working environment are considered to be the part of the reason. To prevent human errors accidents, enhancing safety equipment, driver training programs, and establishing a sound auditing system are suggested and discussed.     

The relationship of dangerous driving with traffic offenses: A study on an adapted measure of dangerous driving

Using data from three different samples and more than 1000 participants, the current study examines differences in dangerous driving in terms of age, gender, professional driving, as well as the relationship of dangerous driving with behavioral indicators (mileage) and criteria (traffic offenses). The study uses an adapted (Romanian) version of the Dula Dangerous Driving Index (DDDI, Dula and Ballard, 2003) and also reports data on the psychometric characteristics of this measure. Findings suggest that the Romanian version of the DDDI has sound psychometric properties. Dangerous driving is higher in males and occasional drivers, is not correlated with mileage and is significantly related with speeding as a traffic offense, both self-reported and objectively measured. The utility of predictive models including dangerous driving is not very large: logistic regression models have a significant fit to the data, but their misclassification rate (especially in terms of sensitivity) is unacceptable high.     

The association between driving anger and driving outcomes: A meta-analysis of evidence from the past twenty years

Through the use of meta-analysis, this study investigated the relationships between driving anger and five types of driving outcomes (aggressive driving, risky driving, driving errors, near misses and accidents). The moderating effects of three variables (age, study publication year, and participants’ country of origin) on these relationships were also examined. A total of 51 studies published over the past two decades met the inclusion criteria for the meta-analysis. The results showed that driving anger significantly predicted all three types of aberrant driving, with zero-order correlations of 0.312, 0.243, and 0.179 with aggressive driving, risky driving and driving errors, respectively. The correlations between driving anger and accident-related conditions, though at relatively weaker levels, were still statistically significant. Tests for effects of the moderating variables suggested that driving anger was a stronger predictor of risky driving among young drivers than among old drivers. Also, the anger–aggression association was found to decrease over time and vary across countries. The implications of the results and the directions for future research are discussed.     

A meta-analysis of the effects of texting on driving

Text messaging while driving is considered dangerous and known to produce injuries and fatalities. However, the effects of text messaging on driving performance have not been synthesized or summarily estimated. All available experimental studies that measured the effects of text messaging on driving were identified through database searches using variants of “driving” and “texting” without restriction on year of publication through March 2014. Of the 1476 abstracts reviewed, 82 met general inclusion criteria. Of these, 28 studies were found to sufficiently compare reading or typing text messages while driving with a control or baseline condition. Independent variables (text-messaging tasks) were coded as typing, reading, or a combination of both. Dependent variables included eye movements, stimulus detection, reaction time, collisions, lane positioning, speed and headway. Statistics were extracted from studies to compute effect sizes (r_c). A total sample of 977 participants from 28 experimental studies yielded 234 effect size estimates of the relationships among independent and dependent variables. Typing and reading text messages while driving adversely affected eye movements, stimulus detection, reaction time, collisions, lane positioning, speed and headway. Typing text messages alone produced similar decrements as typing and reading, whereas reading alone had smaller decrements over fewer dependent variables. Typing and reading text messages affects drivers’ capability to adequately direct attention to the roadway, respond to important traffic events, control a vehicle within a lane and maintain speed and headway. This meta-analysis provides convergent evidence that texting compromises the safety of the driver, passengers and other road users. Combined efforts, including legislation, enforcement, blocking technologies, parent modeling, social media, social norms and education, will be required to prevent continued deaths and injuries from texting and driving.     

What determines the take-over time? An integrated model approach of driver take-over after automated driving

In recent years the automation level of driver assistance systems has increased continuously. One of the major challenges for highly automated driving is to ensure a safe driver take-over of the vehicle guidance. This must be ensured especially when the driver is engaged in non-driving related secondary tasks. For this purpose it is essential to find indicators of the driver’s readiness to take over and to gain more knowledge about the take-over process in general.     

基于互信息与支持向量回归的盾构掘进载荷预测方法研究

掘进载荷是盾构施工中的重要控制量,直接关系着施工安全与效率。通过对掘进载荷影响因素的分析,建立了一种基于工程实测数据分析的掘进载荷特征选择及预测方法。首先,对工程实测数据进行极值归一化预处理,以降低不同参数间量纲和量级的差异产生的支配性影响;其次,通过参数分析和基于互信息的特征选择选取主要的影响参数作为输入;最后,通过支持向量回归（support vector regression,SVR）建立掘进载荷的预测模型,并结合天津地铁9号线盾构施工工程案例检验其预测表现。结果表明,所建立的掘进载荷预测方法能够在工程实测数据包含的众多影响参数中筛选出少量关键特征,实现对掘进载荷的合理预测。研究结果可以为盾构掘进参数的调控提供参考,也为具有众多参数的工程实测数据的分析提供一种思路。     

A behavioral economic analysis of texting while driving: Delay discounting processes

The purpose of the present study was to examine an impulsive decision-making process underlying texting while driving from a behavioral economic perspective. A sample of 108 college students completed a novel discounting task that presented participants with a hypothetical scenario in which, after receiving a text message while driving, they rated the likelihood of replying to a text message immediately versus waiting to reply for a specific period of time. Participants also completed a delay discounting task in which they made repeated hypothetical choices between obtaining a larger amount of money available after a delay and an equal or lesser amount of money available immediately. The results show that the duration of the delay is a critical variable that strongly determines whether participants choose to wait to reply to a text message, and that the decrease in the likelihood of waiting as a function of delay is best described by a hyperbolic delay discounting function. The results also show that participants who self-reported higher frequency of texting while driving discounted the opportunity to reply to a text message at greater rates, whereas there was no relation between the rates of discounting of hypothetical monetary rewards and the frequency of texting while driving. The results support the conclusion that texting while driving is fundamentally an impulsive choice.