外刊精读：无人驾驶汽车反映了当今AI的局限性

Driverless cars illustrate the limits of today’s AI

无人驾驶汽车反映了当今AI的局限性

IN MARCH Starsky Robotics, a self-driving lorry firm based in San Francisco, closed down. Stefan Seltz- Axmacher, its founder, gave several reasons for its failure. Investors’ interest was already cooling, owing to a run of poorly performing tech-sector IPOs and a recession in the trucking business. His firm’s focus on safety, he wrote, did not go down well with impatient funders, who preferred to see a steady stream of whizzy new features. But the biggest problem was that the technology was simply not up to the job. “Supervised machine learning doesn’t live up to the hype. It isn’t actual artificial intelligence akin to c-3PO [a humanoid robot from the “Star Wars” films]. It’s a sophisticated pattern- matching tool.”

今年3月，位于旧金山的无人驾驶卡车公司“星空机器人”（Starky Robotics）宣告关门。创始人斯蒂芬·塞尔兹·阿克斯马赫（Stenfan Seltz-Axmacher）给出了失败的几个原因：一系列科技公司上市后表现不佳，加之货运业务衰退，投资者的兴趣以及开始降温。他写道，自己的公司重视安全性，但缺乏耐心的资方对此反应不佳，他们宁愿看到技术花哨的新功能被源源不断地开发出来。但最大的问题是，技术根本无法胜任。“有监督的机器学习达不到现在的热炒的程度。这不是类似于c-3PO(《星球大战》电影中的人形机器人)的真正的人工智能。这无非是一种先进的模式匹配工具。”

IPOs：首次上市

Recession：经济衰退，经济萎缩

Whizzy adj.采用先进技术的

Hype：n./v.

n. （电视、广播等中言过其实的）促销广告，促销讨论

marketing/media hype 夸张的促销╱媒体广告

v. 夸张地宣传（某事物）

Humanoid n.仿真机器人；类人动物

Go down well with： 受...欢迎；与...相处融洽

Trucking business： 货运业

Live up to: 不辜负，做到

【语法分析】

IN MARCH Starsky Robotics, a self-driving lorry firm based in San Francisco, closed down.

1.主干：Starsky Robotics  closed down。

2.同位语：a self-driving lorry firm based in San Francisco

3.同位语中的定语：based in San Francisco修饰a self-driving lorry firm

Policing social media, detecting fraud and defeating humans at ancient games are all very well. But building a vehicle that can drive itself on ordinary roads is—along with getting computers to conduct plausible conversations—one of the grand ambitions of modern AI. Some imagined driverless cars could do away with the need for car ownership by letting people summon robotaxis at will. They believe they would be safer, too. Computers never tire, and their attention never wanders. According to the WHO, over a million people a year die in car accidents caused by fallible human drivers. Advocates hoped to cut those numbers drastically.

监管社交媒体，发现欺诈行为，在古老的游戏中击败人类——这些固然都很好,但是制造一种能够在普通道路上自主行使的汽车（以及让计算机与人开展听上去合理的对话）是现代AI的宏伟目标之一。有人认为无人驾驶汽车可以让人们随时召唤机器人出租车，这样就不需要拥有汽车了。他们也相信这些车会更安全。电脑永不疲倦，也永远不会分神。根据世卫组织的数据，每年有超过一百万人死于易犯错的人类驾驶员引发的车祸。无人车的拥护者们希望能大幅削减这些数字。

Police v.

1.巡查；维护治安

2.监督；管制

Detect：vt. 察觉；发现；探测

Defeat：n./v.

n. 失败；战胜

v. 击败；挫败（某人）；阻止达成（目的）；反对（动议或提议）；（使）无法理解；（使）无效

Plausible adj.

1.有道理的；可信的

2.巧言令色的；花言巧语的

Robotaxis：机器人出租车

Advocate v./n.

v. 拥护；支持；提倡

n. 1.拥护者；支持者；提倡者;2.辩护律师；出庭辩护人

【语法分析】

But building a vehicle that can drive itself on ordinary roads is—along with getting computers to conduct plausible conversations—one of the grand ambitions of modern AI.

1.主干：Building a vehicle  is one of the grand ambitions.

2.插入语补充解释前述内容：—along with getting computers to conduct plausible conversations—

And they would do it soon. In 2015 Elon Musk, the boss of Tesla, an electric-car maker, predicted the arrival of “complete autonomy” by 2018. Cruise, a self-driving firm acquired by General Motors in 2016, had planned to launch self-driving taxis in San Francisco by 2019. Chris Urmson, then the boss of Waymo, a Google subsidiary widely seen as the market leader, said in 2015 that he hoped his son, then 11 years old, would never need a driving licence.

它们本来很快就要做到了。电动汽车制造商特斯拉的老板埃隆·马斯克在2015年预测，“完全自主”将在2018年到来。通用汽车在2016年收购的无人驾驶公司克鲁斯（Cruise）曾计划2019年之前在旧金山推出无人驾驶出租车。2015年，被广泛视为行业领头羊的谷歌子公司Waymo的时任老板克里斯·厄姆森（Chris Urmson）表示，希望自己11岁的儿子永远不需要驾照。

Predict v.预测，预言，预报

Autonomy n. 自治，自治权

Subsidiary adj./n.

adj. 1.辅助的；附带的；次要的;2.附属的；隶属的

n. 附属公司；子公司

Be acquired by: 被收购

【语法分析】

In 2015 Elon Musk, the boss of Tesla, an electric-car maker, predicted the arrival of “complete autonomy” by 2018.

1.主干：Elon Musk predicted the arrival.

2. the boss of Tesla, an electric-car maker两个同位语，均描述Elon Musk,

But progress has lagged. In 2018 a self-driving car being tested by Uber, a ride-hailing service, became the first to kill a pedestrian when it hit a woman pushing a bicycle across a road in Arizona. Users of Tesla’s “Autopilot” software must, despite its name, keep their hands on the wheel and their eyes on the road (several who seem to have failed to do so have been killed in crashes). The few firms that carry passengers, such as Waymo in America and WeRide in China, are geographically limited and rely on human safety drivers. Mr Urmson, who has since left Waymo, now thinks that adoption will be slower and more gradual.

但是进展慢了下来。2018年，网约车公司优步测试的一辆无人车在亚利桑那州撞到了一名推着自行车过马路的女性，成为无人车致⾏⼈死亡的第一例。特斯拉的“自动驾驶”（Autopilot）软件虽名为“自动”，但用户依然须将手放在方向盘上并看路（几个看起来没能做到的⼈在事故中丧生）。极少数几家载客的公司，例如美国的Waymo和中国的文远知行，覆盖的地理范围都不大，并且依赖⼈类安全驾驶员。已离开Waymo的厄姆森现在认为这项技术的推⼴速度会更慢、更渐进。

Lag v./n.

v. 1.缓慢移动；发展缓慢；滞后；落后于;2.给（管道等）加防冻保暖层

n. 延迟滞后链路聚合组一般是指网络延迟

Autopilot：n. [航] 自动驾驶仪（等于automatic pilot）

Gradual：adj. 逐渐的，平缓的

Ride-hailing:打车服务

Keep on :继续进行，持续不断

【语法分析】

In 2018 a self-driving car being tested by Uber, a ride-hailing service, became the first to kill a pedestrian when it hit a woman pushing a bicycle across a road in Arizona.

1.主干：a self-driving car became the first  to kill a pedestrian.

2. 定语：分词being tested by Uber 修饰 a self-driving car

3.时间状语：when it hit a woman pushing a bicycle across a road in Arizona

Black swans and bitter lessons

黑天鹅和血泪教训

Self-driving cars work in the same way as other applications of machine learning. Computers crunch huge piles of data to extract general rules about how driving works. The more data, at least in theory, the better the systems perform. Tesla’s cars continuously beam data back to headquarters, where it is used to refine the software. On top of the millions of real-world miles logged by its cars, Waymo claims to have generated well over a billion miles-worth of data using ersatz driving in virtual environments.

无人驾驶汽车的工作原理与其他机器学习应用是⼀样的。计算机处理大量数据以提取有关如何驾驶的⼀般规则。至少从理论上讲，数据越多，系统的性能越好。特斯拉的汽车不断将数据传回总部，用来完善软件。Waymo声称，它的汽车除在现实世界中⾏驶了数百万英里外，也使模拟驾驶在虚拟环境中生成了远超过十亿英里的数据。

Crunch n./v./adj.

n. 1.压碎声；碎裂声;2.紧要关头；困境；症结；令人不快的重要消息;3.（突发的）不足，短缺；（尤指）缺钱

v. 1.嘎吱嘎吱地嚼;2.（使）发出碎裂声;3.（在路上）行进发出响声;4.（快速大量地）处理信息；数字捣弄

eg：number crunching

adj. （会议、体育竞赛等）至关重要的，最后一线胜机的

Extract n./v.

n.1.摘录；选录；选曲；节录;2.提取物；浓缩物；精；汁

v. 1.提取；提炼;2.索取，设法得到（对方不愿提供的信息、钱财等）;3.选取；摘录；选录;4.（用力）取出，拔出;5.获得，得到（某种感觉或品质）

Beam n./v.

n. 1.光线；（电波的）波束；（粒子的）束;2.（建筑物的）梁;3.（体操运动的）平衡木;4.笑容；眉开眼笑

v. 1.笑容满面；眉开眼笑;2.发射（电波）；播送;3.照射；射出光（或热）

Headquarter n. 总部；指挥部；司令部

Refine v. 1.精炼；提纯；去除杂质;;2.改进；改善；使精练

The problem, says Rodney Brooks, an Australian roboticist who has long been sceptical of grand self- driving promises, is deep-learning approaches are fundamentally statistical, linking inputs to outputs in ways specified by their training data. That leaves them unable to cope with what engineers call “edge cases”—unusual circumstances that are not common in those training data. Driving is full of such oddities. Some are dramatic: an escaped horse in the road, say, or a light aircraft making an emergency landing on a highway (as happened in Canada in April). Most are trivial, such as a man running out in a chicken suit. Human drivers usually deal with them without thinking. But machines struggle.

澳⼤利亚机器⼈专家罗德尼·布鲁克斯（Rodney Brooks）长期以来⼀直对无人驾驶技术的辉煌未来持怀疑态度。他说，问题在于深度学习的方法从根本上说是⼀种统计方法，根据由训练数据规定的方式把输⼊与输出联系起来。这使它们无法应付工程师们说的“边缘案例”，即训练数据中不常见的异常情况。驾驶充满了这类怪事。其中⼀些是“大场面”，⽐如路上有一匹脱缰的马，或者是轻型飞机紧急降落在高速公路上（4月份在加拿大发生）。大多数不是什么⼤事，例如一个⼈穿着⼩鸡造型的服装跑出来。⼈类驾驶员应对这种情况一般都不假思索，机器却步履维艰。

Grand：adj. 宏伟的；豪华的；极重要的

Promise：n. 许诺，允诺；希望

Sceptical adj.怀疑的

Grand adj.

adj.1.壮丽的；堂皇的；重大的;2.（用于大建筑物等的名称）大;3.宏大的；宏伟的；有气派的;4.傲慢的；高高在上的;5.极好的；快乐的；美妙的；出色的;6.（对上层社会的人的称呼）大

Input n.输入；投入

Output n.输出

Oddity n. 奇异；古怪；怪癖

Trivial adj.不重要的；琐碎的；微不足道的

Struggle v./n.

v.1.奋斗；努力；争取;2.艰难地行进；吃力地进行;3.斗争；抗争；4.搏斗；扭打；挣扎脱身；5.[ V ] ~ (with 争夺；辩论。

n.1.斗争；奋斗；努力；2.搏斗；扭打；（尤指）抢夺，挣扎脱身；3.难事。

【语法分析】

The problem, says Rodney Brooks, an Australian roboticist who has long been sceptical of grand self- driving promises, is deep-learning approaches are fundamentally statistical, linking inputs to outputs in ways specified by their training data.

1.主干：The problem is deep-learning approaches （that）are fundamentally statistical.省略了that

2.同位语：says Rodney Brooks, an Australian roboticist who has long been sceptical of grand self- driving promises 描述The problem

 3.定语：linking inputs to outputs in ways specified by their training data. 修饰 linking inputs to outputs in ways specified by their training data.

One study, for instance, found that computer-vision systems were thrown when snow partly obscured lane markings. Another found that a handful of stickers could cause a car to misidentify a “stop” sign as one showing a speed limit of 45mph. Even unobscured objects can baffle computers when seen in unusual orientations: in one paper a motorbike was classified as a parachute or a bobsled. Fixing such issues has proved extremely difficult, says Mr Seltz-Axmacher. “A lot of people thought that filling in the last 10% would be harder than the first 90%”, he says. “But not that it would be ten thousand times harder.”

例如，一项研究发现，当车道标记被雪部分覆盖时，计算机视觉系统就无法计算。另一项研究发现，某些贴纸可能会导致汽车将“停车”标志误认为是时速限制45英⾥的标志。即使是未被遮挡的物体，如果从不寻常的方向观察也会迷惑计算机，⽐如一篇论文提到摩托车被归类成了降落伞或雪橇。塞尔兹-阿克斯马赫说，解决此类问题极其困难。“很多⼈认为完成最后10%要⽐前90%更难，”他说，“但没想过要难一万倍。”

Obscure adj./vt./n.

adj. 昏暗的，朦胧的；晦涩的，不清楚的；隐蔽的；不著名的，无名的

vt. 使…模糊不清，掩盖；隐藏；使难理解

n. 某种模糊的或不清楚的东西

Sticker n. [轻] 尖刀，贴纸；[胶粘] 粘着剂

Misidentify v. 错认；错误识别；错误鉴定

Parachute n. 降落伞

Bobsled n./v.

n. 双人（或四人）大雪橇；雪橇；雪橇比赛（等于 bobsleigh）

v. 滑大雪橇；乘大雪橇（同 bobsleigh）

Mary “Missy” Cummings, the director of Duke University’s Humans and Autonomy Laboratory, says that humans are better able to cope with such oddities because they can use “top-down” reasoning about the way the world works to guide them in situations where “bottom-up” signals from their senses are ambiguous or incomplete. AI systems mostly lack that capacity and are, in a sense, working with only half a brain. Though they are competent in their comfort zone, even trivial changes can be problematic. In the absence of the capacity to reason and generalise, computers are imprisoned by the same data that make them work in the first place. “These systems are fundamentally brittle,” says Dr Cummings.This narrow intelligence is visible in areas beyond just self-driving cars. Google’s “Translate” system usually does a decent job at translating between languages. But in 2018 researchers noticed that, when asked to translate 18 repetitions of the word “dog” into Yoruba (a language spoken in parts of Nigeria and Benin) and then back into English, it came up with the following: “Doomsday Clock is at three minutes to twelve. We are experiencing characters and dramatic developments in the world, which indicate that we are increasingly approaching the end times and Jesus’ return.”

杜克大学人类与自主实验室主任玛丽·“丫头”·卡明斯（Mary “Missy” Cummings）说，人类能够更好地应对这些奇怪的情形，因为他们可以用对世界运行方式“自上而下”的推理，来指导感官收到的“自下而上”的信号模糊或不完整的情况。AI系统大多缺乏这种能力，从某种意义上说，它只能用一般的大脑工作。尽管它们在自己的舒适区十分称职，但即使是微不足道的改变也可能带来问题。由于缺乏推理和归纳能力，计算机工作的最初始数据也会禁锢它们。卡明斯博士说：“这些系统从根本上就是脆弱的。”在无人车以外的领域，这种智能的狭隘也显而易见。谷歌的“翻译”系统在语言互译方面通常都做得不错。但在2018年，研究人员注意到，当要求把连续18个“dog”译成约鲁巴语（尼日利亚和贝宁的部分地区使用的一种语言）再译回英语时，它给出了这样的译文：“末日钟还有三分钟就要到12点。我们正在经历世界上的人物和戏剧性发展，这表明我们越来越接近末日和耶稣的归来。”

Top-down adj. [计] 自顶向下；组织管理严密的

Bottom-up adj. [计] 自底向上的；从细节到总体的

Competent adj.1.足以胜任的；有能力的；称职的;2.合格的；不错的；尚好的;3.有决定权的

Problematic：adj. 问题的；有疑问的；不确定的

Generalise：v.概括，归纳，总结

Brittle adj.1.硬但易碎的；脆性的;2.脆弱的;3.尖利的；刺耳的

Repetition n.1.重复；重做；重说;2.重做的事；重说的话

Gary Marcus, a professor of psychology at New York University, says that, besides its comedy value, the mistranslation highlights how Google’s system does not understand the basic structure of language. Concepts like verbs or nouns are alien, let alone the notion that nouns refer to physical objects in a real world. Instead, it has constructed statistical rules linking strings of letters in one language with strings of letters in another, without any understanding of the concepts to which those letters refer. Language processing, he says, is therefore still baffled by the sorts of questions a toddler would find trivial.

纽约大学的⼼理学教授Gary Marcus表⽰，除了颇有些喜剧价值外，误译凸显了谷歌的系统并不理解语言的基本结构。它全然不懂动词或名词之类的概念，更不用说理解名词指的是现实世界中的物理对象了。相反，它构造了统计规则，将一种语言的字母字符串与另一种语言的字母字符串联系在一起，但对这些字母所指的概念一无所知。因此，他说：“哪怕幼⼉也能轻易解答的问题也会让语言处理系统困惑不已。”

Highlight：vt. 强调, 突出, 使显 n. 最精彩的部分, 最重要的事情。

例句：The President highlights the importance of his visit to China.

Baffle：vt. 使困难, 使为难。

例句：The absence of clues baffled the police.

Toddler：n. 学步的小孩；幼童装。

例句：he toddler pointed to the toy he wanted.

Trivial：adj. 不重要的，琐碎的；琐细的。

例句：A trivial misunderstanding caused a breach between them.

How much those limitations matter varies from field to field. An automated system does not have to be better than a professional human translator to be useful, after all (Google’s system has since been tweaked). But it does set an upper bound on how useful chatbots or personal assistants are likely to become. And for safety-critical applications like self-driving cars, says Dr Cummings, AI’s limitations  are potentially show-stopping.

这些限制到底有多紧要因领域而异。毕竟自动化系统并不一定要胜过专业译员才有用处（谷歌的系统之后有所优化）。但这确实也给聊天机器⼈或个⼈助⼿到底能做到什么程度设置了上限。而对于无人车等性命攸关的应用， Cummings博士说：“AI的局限性可能会成为拦路虎。”

Tweak：n. 拧；扭；焦急 vt. 扭；用力拉；开足马力。此处翻译为谷歌对系统进行“优化”。

例句：After that, it's all about tweaking your thinking.

Upper bound：上界,最大值。

Researchers are beginning to ponder what to do about the problem. In a conference talk in December Yoshua Bengio, one of AI’s elder statesmen, devoted his keynote address to it. Current machine- learning systems, said Dr Bengio, “learn in a very narrow way, they need much more data to learn a new task than [humans], they need humans to provide high-level concepts through labels, and they still make really stupid mistakes”.

研究人员开始思索如何解决问题。AI领域的资深活动家约书亚·本希奥（Yoshua Bengio）在12月的⼀次会议中专门就此发表了主题演讲。本希奥博士说：“当前的机器学习系统以⼀种非常狭隘的方式学习，它们需要比⼈类多很多的数据来学习新任务，它们需要⼈类通过标签提供高级概念，但仍然犯下非常低级的错误。”

Ponder：vt. 仔细考虑；衡量。vi. 沉思；考虑。

例句：She pondered his marriage proposal for weeks.

Beyond deep learning

超越深度学习

Different researchers have different ideas about how to try to improve things. One idea is to widen the scope, rather than the volume, of what machines are taught. Christopher Manning, of Stanford University’s AI Lab, points out that biological brains learn from far richer data-sets than machines.Artificial language models are trained solely on large quantities of text or speech. But a baby, he says, can rely on sounds, tone of voice or tracking what its parents are looking at, as well as a rich physical environment to help it anchor abstract concepts in the real world. This shades into an old idea in AI research called “embodied cognition”, which holds that if minds are to understand the world properly, they need to be fully embodied in it, not confined to an abstracted existence as pulses of electricity in a data-centre.

不同的研究⼈员对如何改进现状有不同的想法。一种想法是扩大机器训练的范围，而不单是数量。斯坦福大学AI实验室的克里斯托弗·曼宁（Christopher Manning）指出，相比机器，生物大脑学习的数据集要丰富得多。人工语言模型仅用大量的文本或语音训练。他说：”但婴⼉可以依靠声音、语调或追踪父母的眼神，以及丰富的物理环境来帮助自己在现实世界中把握抽象概念。这涉及到AI研究中⼀个叫做“具身认知”的古老观点，它认为如果⼈们要正确地理解世界，就需要充分沉浸其中，而不仅限于抽象的存在，如数据中心里的一些电脉冲。

Anchor：n. 锚；人安全感的物(或人) vt. 抛锚；担任(电视节目等的)主持人；

vt. & vi. (把…)系住, (使)固定。此处引申翻译为“把握，理解”。

例句：They lay at anchor outside the harbour.

Embodied cognition：具身认知（Embodied cognition），也称“具体化”（embodiment），是心理学中一个新兴的研究领域。具身认知理论主要指生理体验与心理状态之间有着强烈的联系。生理体验“激活”心理感觉，反之亦然。简言之，就是人在开心的时候会微笑，而如果微笑，人也会趋向于变得更开心。

Embody:vt. 表现, 象征, 具体表现；包括; 包含。

例句：The new model of car embodies many improvements.

Biology offers other ideas, too. Dr Brooks argues that the current generation of AI researchers “fetishise” models that begin as blank slates, with no hand-crafted hints built in by their creators. But “all animals are born with structure in their brains,” he says. “That’s where you get instincts from.”

生物学也提供了一些思路。Brooks 博士认为，当前一代AI研究⼈员“迷恋”始于空白状态的模型——创建者没有手工置⼊任何提示。但是所有动物的大脑都有天生的结构，他说：“本能就是从这里来的。”

Dr Marcus, for his part, thinks machine-learning techniques should be combined with older, “symbolic AI” approaches. These emphasise formal logic, hierarchical categories and top-down reasoning, and were most popular in the 1980s. Now, with machine-learning approaches in the ascendancy, they are a backwater.

Brooks 博士则认为，机器学习技术应与较早的“符号AI”方法结合使用。这些方法强调形式逻辑、层次类别和自上而下的推理，在1980年代最受欢迎。现在，随着机器学习方法的兴起，它们已成为一潭死水。

Hierarchical：adj. 分等级的。

例句：They point out clearly that their relationship with students is cooperative, rather than hierarchical.

Ascendancy：n. 优势；支配地位。

例句：We have had ascendancy over the enemy in the battle.

Backwater：n. 死水, 滞水, 回水；闭塞的地方, 不受外界事件或新思想影响的地方。

例句：I couldn't permit someone of your abilities to moulder away in a backwater.

But others argue for persisting with existing approaches. Last year Richard Sutton, an AI researcher at the University of Alberta and DeepMind, published an essay called “The Bitter Lesson”, arguing that the history of AI shows that attempts to build human understanding into computers rarely work. Instead most of the field’s progress has come courtesy of Moore’s law, and the ability to bring ever more brute computational force to bear on a problem. The “bitter lesson” is that “the actual contents of [human] minds are tremendously, irredeemably complex…They are not what should be built in [to machines].”

但其他⼈则主张坚持现有方法。去年，艾伯塔和DeepMind的AI研究⼈员理查德·萨顿（Richard Sutton）发表了⼀篇名为《苦涩的教训》（The Bitter Lesson）的文章，认为AI的历史表明，试图将⼈类的理解建构到计算机中的尝试很少行得通。相反，该领域的⼤多数进步都得益于摩尔定律，以及不断引⼊更多蛮力计算来解决问题的能力。血泪教训是：“［⼈类］思想的实际内容极度、无可救药地复杂……这不是应该置入［机器］的东西。”

Moore’s law：摩尔定律是由英特尔（Intel）创始人之一戈登·摩尔（Gordon Moore）提出来的。其内容为：当价格不变时，集成电路上可容纳的元器件的数目，约每隔18-24个月便会增加一倍，性能也将提升一倍。换言之，每一美元所能买到的电脑性能，将每隔18-24个月翻一倍以上。这一定律揭示了信息技术进步的速度。尽管这种趋势已经持续了超过半个世纪，摩尔定律仍应该被认为是观测或推测，而不是一个物理或自然法。

Brute：adj. 无理性的；残忍的 n. 畜生；残暴的人。

例句：He is a brute to his children.

Tremendously：adv. 极大地;极端地;极其;非常。tremendous adj.

例句：Setting off firecrackers makes a tremendous noise and causes serious air pollution.

Irredeemably：无法补救地，不能挽救地。irredeemable adj.

例句：The applicant was irredeemably incompetent.

Away from the research labs, expectations around driverless cars are cooling. Some Chinese firms are experimenting with building digital guide rails into urban infrastructure, in an attempt to lighten thecognitive burden on the cars themselves. Incumbent carmakers, meanwhile, now prefer to talk about “driver-assistance” tools such as automatic lane-keeping or parking systems, rather than full-blown autonomous cars. A new wave of startups has deliberately smaller ambitions, hoping to build cars that drive around small, limited areas such as airports or retirement villages, or vehicles which trundle slowly along pavements, delivering packages under remote human supervision. “There’s a scientific reason we’re not going to get to full self-driving with our current technology,” says Dr Cummings. “Thisless ambitious stuff—I think that’s much more realistic.”

在研究实验室之外，⼈们对无人车的期待正在降温。一些中国公司正尝试在城市基础设施中构建数字导轨，以减轻无人车的认知负担。同时，成熟的汽车制造商现在更喜欢谈论“驾驶员辅助”工具，例如自动车道保持停车系统，⽽不是完全自主驾驶的汽车。新一波创业公司则有意缩减野⼼，希望制造能在狭⼩有限的区域（如机场或退休社区）里行驶的汽车，或在人行道上缓慢移动、在⼈员的远程监督下递送包裹的车辆。“出于科学原因，我们不打算使用现有的技术实现完全无人驾驶，”卡明斯博士说，“我觉得这些不那么宏伟的目标要现实得多。”

Infrastructure：n. 基础设施; 基础结构。

例句：bizonal" service area for leisure and business office areas, including shopping, office services, community services, improving infrastructure systems.

Incumbent：adj.&n. 在职的，任职者。

例句：Any (not either ) of the three opposition candidates still in the race would make a better president than the incumbent.

full-blown：adj. 盛开的,张满的,成熟的，完全的。

例句：Our Africa editor Richard Hamilton says efforts are going on to avert a full blown civil war.

感兴趣的同学可以扫描下方二维码，阅读《滴滴出行：由苹果公司支持的公司目标是100万辆自动驾驶出租车》这篇短文。

推荐阅读

投稿邮箱：[e-mail]

主题；篇幅；体例

本文责编 ✎ 金钟罩

本期编辑 ✎ Ben