1 Introduction

The consequences of vehicle automation are potentially far reaching;however,the body of the literature regarding consumer preferences and public opinion is now in its formative stages(Table 1 summarizes these recent studies[1–6])and relatively little is known with reliability.

The objective of this study is to identify novel aspects of consumer preferences and public opinion regarding highly automated cars(levels 2 through 4 under the NHTSA taxonomy[7]).For instance,we investigate consumers’relative prioritization of various prospective benefits of automated cars and their preferences for travel at higher rates of free-flow speed than at present;neither of these issues are addressed in the prior literature.To do this,we designed a survey protocol that included both attitudinal questions and a stated-preference component.The survey was administered to a moderate-sized sample(n=370),which was generated according to a sampling plan based on age/gender quotas(the sampling strategy is described in detail in Sect.3).Therefore,the results can be interpreted as being nationally representative along these dimensions;however,larger sample(n＞＞1000)follow-on surveying will be required to have greater confidence in the findings.

The remainder of this paper is organized as follows.Section 2 describes the survey protocol that we designed and executed in this study.Section 3 then presents and discusses the empirical results,and Sect.4 summarizes and concludes this paper.

2 Overview of survey protocol

The survey protocol comprised three distinct modules,which respondents completed in the following sequence:

·Demographic information.

·Hypothetical stated-preference scenarios.

·Attitudinal questions.

The survey protocol was reviewed and approved by the SUNY New Paltz Human Research Ethics Board and pilottested with college staff as respondents prior to the main fieldwork.The survey fieldwork was performed via webbased computer-aided self-administration(CASI).The fieldwork was undertaken by the market-research firm Qualtrics,which maintains market-research panels to which researchers can purchase access.The present survey had a straightforward sampling plan(a nationally representative sample in terms of age and gender),which served to control costs;a more-complex quota plan(e.g.,introducing a requirement for a minimum number of respondents that do not hold a driving license)would have had corresponding impacts on resource requirements for this study.Respondents received incentives in the form of‘…points[that]can be pooled and later redeemed in the form of gift cards,skymiles,credit for online games,etc’.1Personal communication from Qualtrics’S.J.Campbell,4/14/2016. The value of the incentives are variable(depending in part on whether a particular respondent is in a difficult-to-reach segment of the sample quota)and were approximately equivalent to$1/respondent for completing this survey.

·The person riding in a driverless car should have the choice of whether to leave a large distance behind the car ahead of it,if they wish to reduce the possibility of rear-ending the car ahead,even if this makes traffic congestion worse:48%.

招考模式还是相对注重学业，虽然政策和考试形式不断改革，但是学业成绩比重依旧很大，学生成绩是能否进入优秀学校的重要考核指标。而学校又背负着能否培养出优秀学生的压力，成绩就成为了学校的追求。同时，学生升入重点初中的比例影响到学校招生和生源，所以学生的成绩又影响到教师个人的考核，这种种原因就造成了为了提升学生成绩，学生课业越来越重的现象。

Data were collected in four batches of approximately equal sizes:

1.April 25,2016.

2.May 9,2016.

3.May 17,2016.

4.May 19,2016.

The survey questionnaire was identical across the four batches,with the exception of the numerical values in the stated-preference module.The logic for varying the numerical values in the stated-preference module is to increase the variability in the dataset,which increases the robustness of the regression parameters and minimizes the opportunity for collecting stated-preference data that do not provide useful information in model estimation (see Sect.3).

For the purposes of communicating with respondents within the survey,we elected to consistently use the term‘driverless car’rather than ‘automated car.’

2.1 Design of stated-preference module

总而言之，高中信息技术科教师必须要肩负起提高教学质量的责任，根据实际情况，在课堂教学中合理应用新教学模式，这也是一种最佳的选择。不同的教学方式对学生学习情况，以及课堂教学效果都会产生不同的影响。因此，高中信息技术科教师在选择新的教学方式时，必须要对教学效果做出准确的评估和及时的收集。在这个前提下对教学方式做出适当的调整，只有这样才可以提高教学质量，同时还可以提升学习成绩。

· Drive a ‘normal’car or take a bus(see two paragraphs below).

· A ‘semi-driverless’car.2The text describing the driverless cars is as follows:

· A ‘completely driverless’car.

·Take a commercial air flight.

花奴打了电话后，小虫被罚了款后放了。从派出所出来，小虫不知道玉敏在等他，带一帮乡党去大排档喝酒，喝到半夜才回来。小虫喝高了，满嘴酒气，舌头像短了一截，对玉敏说，我……不会……放过她的。光脚不怕穿鞋的，不把钻戒要回来，老子决不罢休！玉敏看他醉醺醺的样子，什么也没说，只是紧紧地搂着小虫，一股温暖在心头荡漾。

The stated-preference module consisted of 10 replications for each respondent.During each replication,the cost and duration and maximum speed of each of the options in the listing above were varied according to a pre-defined D-efficient design prepared by the research team[8].

本次报告显示,手术切除和注射碘酊治疗口腔粘液腺囊肿治愈率均达到95%以上,临床疗效确切,两种治疗方法没有显著性差异。10例复发病例(手术切除治疗4例,注射碘酊治疗6例),其原因可能是囊壁处理不彻底,部分囊壁残留导致囊肿复发。

The questionnaire branched on the basis of whether or not the respondent holds a driver’s license.In the statedpreference module,licensed respondents were presented with the option to drive a ‘normal’(i.e.,not automated)car if they wish,whereas unlicensed respondents were instead provided the option to take an inter-city bus.Also,respondents who do not hold a driving license were only presented with the option of using a ‘fully automated’car(i.e.,NHTSA level 4),whereas respondents holding a driving license were also presented in specific scenarios with the option of using a ‘partly automated’car(NHTSA level 2/3).The questionnaires were otherwise identical,with the exception of a question about how fast drivers drive on Interstate Highways(which was not asked of unlicensed respondents and is not analyzed in this paper).

Footnote 2 continued

Carmakers are now experimenting with driverless cars,which

use sensors and computers to ‘drive’themselves.It works a bit

like an advanced form of Cruise Control,but you tell a

driverless car the destination address of your journey and it is

able to brake by itself,change lanes,and make turns to get you

there.In the different scenarios,you will see two types of

获得正中矢状切面的判断标准：透明隔腔，胼胝体，第四脑室，小脑蚓部，小脑幕，后颅窝池6个中线重要结构中获得大于等于三个，即认为为获得正中矢状切面。正中矢状切面上重要结构显示的标准：透明隔腔显示为不规则无回声区；胝体显示为紧贴透明隔腔上方的均匀低回声带；小脑蚓部显示为桔子瓣样的稍高回声，甚至可以观察到原裂和次裂；第四脑室显示为尖端指向小脑蚓部的三角形无回声区；小脑幕显示为小脑蚓部后方略上方的条状高回声；颅后窝池显示为小脑蚓部后方与枕骨之间的无回声区。

‘‘driverless cars’’:

1. A ‘‘semi-driverless’’car can drive itself for nearly all of the journey(you just need to enter the destination address).However,you must keep your hands on the steering wheel and be ready to take control at any time in case something goes wrong to avoid accidents.

2. A ‘‘completely-driverless’’car will drive itself for the entire journey,after you tell it your destination.You are free to do whatever you wish as you travel with no need to ever keep your hands on the wheel,as the car avoids accidents on its own.

3 Results

Table 2 presents descriptive results of the sample regarding socio-demographics and spatial characteristics of their residence.Pluralities reported being married(42%)and living in a suburban location(46%);70%reported household income of under$75,000/year,approximately two-thirds(68%)of respondents reported living without children in their household,and 90%reported having a full(i.e.,excluding learner’s permits)driving license.

3.1 Prioritization and attitudinal results

After the stated-preference module(see Sect.3.2),the survey concluded with a set of attitudinal questions.The first three of these questions asked about speed selection when driving ‘manually’(not discussed further in this paper),preferencesforprogramming instructionsof autonomous cars,and prioritization of their bene fits.In order to minimize primary effects for the questions consisting of agreement with statements or prioritization,the order in which these statements were presented was randomized(i.e.,different for different respondents).In Tables 3 and 4,color coding is used for ease of interpretation(darkest green for the largest percentages and darkest red for the smallest percentages,excluding ‘Unsure’responses).

‘Visiting relatives’was chosen as the hypothetical activity,because in comparison with other activities that were considered by the research team(e.g.,business travel,tourism)it was felt that visiting relatives would likely be a type of long-distance travel that is familiar to a larger share of respondents.

3.1.1 Prioritization of automated cars’programming instructions

Respondents were presented the following paragraph of text which describes the trade-off between safety and congestion.Respondents then indicated which of the following responses best describes their view:Driverless cars will need to be programmed with instructions of how to follow behind other cars.Following closely can reduce the severity of traffic congestion,but this could increase the risk of rearend crashes.Which of these statements best describes yourview ofhow driverlesscarsshould be programmed:

·Driverless cars should be programmed to follow closely behind the car ahead of it in traffic,in order to reduce traffic congestion,even if this increases the possibility of rear-ending the car ahead:9%.

转基因食品的研发层面：在研发过程中，所有研究资料均应妥善保管，研究人员应当记录研发中的各种情况，尤其是研究结果的去向，更应该详细记载，以备查阅。

Only complete responses(i.e.,no partially completed surveys)were provided in the final dataset.No surveys completed in less than 4 min were included in the final dataset.The average time to complete the survey was 14 min and 43 s.

加强区域流域水资源水环境承载能力和经济社会用水结构研究，分析区域流域虚拟水贸易状况，通过政府政策和水价调整，发挥市场配置资源的重要作用，调整产业布局和结构，水资源丰富区域可多生产耗水量大产品，整体贸易以输出水量为主；缺水地区以生产低耗水量产品为主，整体贸易以输入水量为主；达到人与自然和谐共生。

·Driverless cars should be programmed to leave a large distance behind the car ahead of it,in order to reduce the possibility of rear-ending the car ahead,even if this makes traffic congestion worse:42%.

Hainbuch最新推出的MAXXOS六棱锥形（非圆锥形）心轴按照严格的制造要求和工艺可靠性设计，满足了特定领域用户对更高性能和更可靠流程心轴不断增长的需求。由于其夹具是六角锥形，因此可实现最大的转矩传递。与传统的Mando T211心轴相比，MAXXOS T211反涨心轴夹具转矩传递率高达155%，弯曲刚度高达57%，其带六边锥形底座的涨套与锥形夹具完全匹配，可实现最大的切割性能。

3.1.2 Prioritization of benefits from automated cars

A simple majority of respondents(51%)chose the‘highest possible level of safety’as their top priority,more than three times the proportion that selected any other item as their top priority.The most frequently cited second priority was congestion reduction,and at the opposite end of the scale,the most frequently cited item as respondents’lowest priority is the ability to send an unoccupied driverless car to perform errands.The item cited least frequently as the top priority was being able to do other activities while inside a car(9%);this is consistent with findings from Piao et al.[2](see Table 2).

·Being able to read,sleep,send text messages,or do other activities inside the car besides driving,while the car does the driving.

·Being able to send a driverless car to pick up or drop off packages,groceries,or children,without a human driver inside the vehicle.

·Having the highest possible level of safety in a driverless car.

·Having traffic congestion reduced,so that traffic moves more smoothly even when there are many cars on the road.

In the stated-preference module,respondents were presented with the task of deciding how to travel to‘see relatives that live in another part of the country’and were presented the following options:

为了让学生阅读能够更有成效性，每个月进行班级阅读摘记评比。阅读摘记内容，学生根据自己的课外阅读进行好词（2—3个词）、佳句（1—2句）、读后感（1—2句）等三个方面的摘记。每个月进行阅读摘记评比并点评（学生点评，老师总结），每次评比选出5名学生各奖励1本绘本书。这个成效是非常明显的，最明显的是后进生。后进生一开始写话只能根据图画写一句话，到期末能够根据图画表达并加上自己的情感，最大的进步在于学生敢于表达，并大胆运用课堂上没教过但是生活中常用的字词。

随着药物溶栓、经皮冠状动脉介入治疗、冠脉旁路移植术等再灌注治疗技术日益完善且广泛普及,心脏再灌注损伤的干预治疗备受重视.目前,已有观点多集中于心肌梗死后缺血再灌注引起的氧化应激损伤及其产生的强烈炎症反应.这些损伤反应的最终结果是大片细胞的坏死和进一步的心肌细胞凋亡.

·When there are few other cars on the road,being able to travel much faster(higher speed)than drivers are allowed to drive today.

Table 3 shows results from a question in which respondents were asked to prioritize among five prospective benefits of automated cars:

3.1.3 Attitudinal results

It can be seen that only a small minority of respondents(9%)indicated that congestion reduction should be prioritized over safety.The large majority felt that either the choice of how to make this trade-off should rest with the occupant of an automated car(48%)or that automated cars should be programmed to prioritize safety over congestion reduction.

这篇文章就写得明白晓畅，可谓晓之以理，动之以情。开头先叙王朝以孝治天下的背景，可谓晓之以理，中叙皇帝对自己的天高地厚之恩以及父母年迈体弱的具体情况，又引用李密报刘的典故，可谓动之以情。整篇文章流畅自然，没有古奥文词，没有隐僻典故，读后给人一种情真意切、字字从肺腑中淌出的感觉。其写给友人的书札如《答叶中孚书》就写得极为风趣诙谐：

Table 4 shows results from a set of questions designed by the research team to investigate attitudes thought to affect one’s views regarding automated cars.In the interests of space,we comment on a subset of the results presented in Table 4.

Respondents indicated strong preferences for preferring to travel by car than public transportation,yet a majority indicated that they agreed that they care about environmental issues.Fully 65%of respondents reported enjoying driving.

Majorities expressed willingness to pay more(51%)and accept slower travel(54%)in exchange for greater comfort while traveling.Nearly two out of five respondents(39%),however,indicated that speed is the most important factor in how they travel.

3.2 Stated-preference results

Table 5 contains results from the estimation of two specifications of mode choice models,using the data from the stated-preference survey.

本组72例分为4大类，即永存左上腔61例，永存左上腔伴右上腔缺如1例，主动脉弓左后无名静脉8例，左无名静脉胸腺内走行2例。合并心内畸形17例，合并心外畸形7例，合并单脐动脉3例，其中1例同时合并右脐静脉；其余41例未合并心内心外畸形。41例永存左上腔存在双上腔静脉，1例永存左上腔合并右上腔缺如，8例主动脉弓后左无名静脉，左无名静脉胸腺内走行2例，属先天上腔静脉单一变异。

Alternative speci fic constants and alternative speci fic travel time parameters were estimated,with generic parameters estimated for journey cost and maximum travel speed[9,10].The multinomial logit results are included for completeness;the mixed logit with panel effects model is preferred as it accounts for the fact that stated-preference responses are not each independent from one another,because each respondent performed a set of ten scenarios.Taking this ‘panel effect’into account improved goodness of fit(McFadden’s rho-squared)and accounts for one dimension of bias in the parameter estimates,thereby yielding parameter estimates that are closer to the theoretical‘true’values.

All parameters for travel time and cost have the expected(negative)sign,indicating that,ceteris paribus,respondents were less likely to select alternatives that were more expensive or took additional time.These diagnostic results provide a measure of confidence that the data appear reasonable and suitable for subsequent analysis.To the authors’knowledge,the result that travelers appear to(positively)value the maximum speed at which they would travel in an automated car during their journey,independently from the duration of their journey,is a novel finding;it will have consequences for how automated cars are programmed to operate on both arterial streets[12]and freeways[13].This apparent expression of consumer preference is somewhat in tension with,for instance,the suggestion by Anderson et al.[11,p.30]that automated cars might‘enable lower peak speeds(improving fuel economy)but higher effective speeds(improving travel time).’

评审方式：改变评审方式。吸收各行业专家，采取各行业联合评审，根据专家自身专业优势，分章节进行主审和整体内容评审相结合的方式，这样能充分发挥各专家的专长，也能更好地整合各个专家的观点。

A counterintuitive result from the mode choice analysis is that the parameter for travel time for the‘Semi-driverless car’option is smaller than for the ‘Completely driverless car.’This is contrary to a priori expectations,as it has been theorized by researchers[13,14]that travelers in fully automated cars may have lower values(disutilities)of travel time,as being disengaged from the driving task would allow the vehicle occupant to focus on other productive or leisurely activities.Further analysis into this issue is needed to identify whether the result reported here is anomalous,or indicative of a misunderstanding on the part of researchers regarding people’s preferences for the various technologies of automated cars.

4 Conclusions

In this study,we present results from a moderate-sample survey of the public’s priorities for automated cars and preferences for their specific attributes in the context of a stated-preference survey.The context of the stated-preference survey is long-distance travel(specifically to visit relatives),and a novel result is that travelers appear to value both their ‘effective’or average speed and their maximum speed(i.e.,their‘cruising speed’under free-flow conditions).Resultsregarding the public’spriorities demonstrate that safety appears to clearly be a higher priority than congestion reduction,a finding which has consequences for the car-following distances thatare programmed into the control algorithms of automated cars.Documenting this finding is important,as decisions must be made in the near future by driving-algorithm designers,public-sectorregulators,and ultimately thejudiciary regarding the guidelines for acceptable automated drivingbehavior instructions.

Acknowledgements The authors wish to thank the anonymous referees for helpful feedback on earlier drafts.An earlier version of this study was presented at the 2017 Transportation Research Board conference(Poster#17-3648);the authors thank attendees for helpful questions/feedback.Lustgarten thanks SUNY New Paltz’s SURE program for financial support,and Le Vine acknowledges support of the University Transportation Research Center,Region 2(Grant#49997-53-25,titled:Empirical Aspects of Autonomous Cars).The usual disclaimer applies:any errors in this paper are the authors’sole responsibility.

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License(http://creativecommons.org/licenses/by/4.0/),which permits unrestricted use,distribution,and reproduction in any medium,provided you give appropriate credit to the original author(s)and the source,provide a link to the Creative Commons license,and indicate if changes were made.

References

1.Schoettle B,Sivak M(2014)Public opinion about self-driving vehicles in China,India,Japan,the U.S.,the U.K.,and Australia.http://deepblue.lib.umich.edu/bitstream/handle/2027.42/109433/103139.pdf

2.Piao J,McDonald M,Hounsell N,Graindorge M,Graindorge T,Malhene N(2016)Public views towards implementation of automated vehicles in urban areas.Transport Research Procedia,vol. 14. http://www.sciencedirect.com/science/article/pii/S2352146516302356

3.J.D.Power Inc.(2016)U.S.tech choice study.http://www.jdpower.com/resource/us-tech-choice-study

4.Zmud J,Sener IN,Wagner J(2016)Revolutionizing our roadways:consumer acceptance and travel behavior impacts of automated vehicles.TexasTransportation Institute.http://d2dtl5nnlpfr0r.cloudfront.net/tti.tamu.edu/documents/TTI-2016-8.pdf

5.Weinstein D(2016)Forecasting consumer adoption of autonomous vehicles and the impact on total vehicle sales.In:Paper presented at 5th annual automated vehicles symposium,July 19th–21st 2016,San Francisco

6.Kreuger R,Rashidi T,Rose J(2016)Adoption of shared autonomous vehicles:hybrid choice modeling approach based on stated-choice survey.In:Paper presented at the 95th annual meeting of the transportation research board,January 2016,Washington DC

7.NHTSA(2013)U.S.department of transportation releases policy on automated vehicle development.http://www.nhtsa.gov/About+NHTSA/Press+Releases/U.S.+Department+of+Transportation+Releases+Policy+on+Automated+Vehicle+Development

8.Rose JM,Bliemer MCJ(2009)Constructing efficient stated choice experimental designs.Transp Rev 29(5):587–617

9.Ben-Akiva M,Lerman S(1985)Discrete choice analysis.MIT Press,Cambridge

10.Louviere JJ,Hensher DA,Swait JD(2000)Stated choice methods:analysis and application.Cambridge University Press,Cambridge

11.Anderson JM,Nidhi K,Stanley KD,Sorensen P,Samaras C,Oluwatola OA(2016)Autonomous vehicle technology:a guide for policymakers.Santa Monica,CA:RAND Corporation.http://www.rand.org/pubs/research_reports/RR443-2.html

12.Le Vine S,Liu X,Zheng F,Polak J(2016)Automated cars:queue discharge at signalized intersections with‘Assured-Clear-Distance-Ahead’driving strategies.Transp Res Part C Emerg Technol 62:35–54

13.Le Vine S,Kong Y,Liu X,Polak J(2016)Vehicle automation,legal standards of care,and freeway capacity.http://dx.doi.org/10.1007/s11116-017-9825-8

14.Malokin A,Mokhtarian PL,Circella G(2015)How do activities conducted while commuting influence mode choice?Testing public transportation advantage and autonomous vehicle scenarios.In:94th annual meeting of the transportation research board,January 2016,Washington,DC