評論
打印
Science and technology
科技版塊
Self-driving cars
自動駕駛汽車
Hands off the wheel
雙手離開方向盤
Teaching autonomous cars to drive with computer games
用電腦游戲教自動駕駛汽車駕駛
WHEN DRIVING, Clara-Marina Martinez makes a note of any unusual behaviour she sees on the road.
駕駛過程中，克拉拉-瑪麗娜·馬丁內(nèi)斯會記下她在路上看到的任何不尋常的行為。
She then feeds these into machine-learning algorithms, a form of AI, which she is helping develop for Porsche Engineering, a division of the eponymous German sports-car company.
然后，她將這些數(shù)據(jù)輸入機器學習算法，這是人工智能的一種形式，她正在幫助德國保時捷的子公司開發(fā)這種算法。
Those algorithms are intended to produce a system reliable enough for a car to drive itself.
這些算法旨在產(chǎn)生一個足夠可靠的系統(tǒng)，使汽車能夠自動駕駛。
Such a fully autonomous car, known in the industry as Level 5, should be able to complete an entire journey without any intervention from the driver, and cope with all situations on the road.
這在業(yè)內(nèi)被稱為5代的全自動汽車，應(yīng)該能夠在沒有司機任何干預(yù)的情況下完成整個旅程，并應(yīng)對道路上的所有情況。
But this is proving hard to achieve, and many attempts to do so are being scaled back.
但事實證明，這很難實現(xiàn)，許多這樣做的嘗試都在縮減。
Last year, for instance, Uber, a ride-hailing service, sold off its unit developing self-driving cars.
例如，去年叫車服務(wù)公司優(yōu)步賣掉了開發(fā)自動駕駛汽車的部門。
Autonomous vehicles are touted as being not just convenient but potentially safer.
自動駕駛汽車被吹噓說不僅方便，而且可能更安全。
However, just as people take time to learn how to drive safely, so do machines.
然而，正如人們需要時間來學習如何安全駕駛一樣，機器也是如此。
And machines are not as quick on the uptake.
而且機器的理解速度也沒有那么快。
The RAND Corporation, an American think-tank, calculates that to develop a system 20% safer than a human driver, a fleet of 100 self-driving cars would have to operate 24 hours a day, 365 days a year, and cover 14bn kilometres.
美國智庫蘭德公司計算出，要開發(fā)一個比人類司機安全20%的系統(tǒng)，一支由100輛自動駕駛汽車組成的車隊必須一年365天每天24小時運行，行駛140億公里。
At average road speeds, that would take about 400 years.
在平均路速下，這將需要大約400年的時間。
Carmakers such as Porsche therefore accelerate the development process using simulators.
因此，保時捷等汽車制造商使用模擬器加快了開發(fā)過程。
These teach software about hazards only rarely encountered in reality.
這些模擬器教給軟件的是現(xiàn)實中很少遇到的危險。
Dr Martinez and her colleagues employ “game engines”, the programs that generate photorealistic images in computer games, to do this.
為了做到這一點，馬丁內(nèi)斯博士和她的同事們使用了“游戲引擎”，即在電腦游戲中生成逼真圖像程序。
These are used to create virtual worlds through which the software can drive.
這些是用來創(chuàng)建虛擬世界的，軟件可以通過這些虛擬世界來驅(qū)動。
Objects in these virtual worlds are assigned their physical characteristics (ie, buildings are hard, people are soft) so that the sensors in vehicles, such as cameras, radar, lidar (a form of radar that uses light) and ultrasound transceivers respond in the appropriate way.
這些虛擬世界中的物體被賦予了它們的物理特征（即，建筑物是堅硬的，人是柔軟的），因此車輛中的傳感器，如相機、雷達、激光雷達（一種用光的雷達）和超聲波收發(fā)器都會以適當?shù)姆绞阶龀龇磻?yīng)。
Once the software has been trained, it is tested in real autonomous vehicles by re-creating those situations on a test track.
一旦軟件經(jīng)過訓練，它就會在真實的自動駕駛汽車上進行測試，在測試軌道上重現(xiàn)這些情況。
How quickly, if ever, all this will translate into reality remains to be seen.
這一切究竟會以多快的速度轉(zhuǎn)化為現(xiàn)實，需要我們拭目以待。
Both regulators and customers will need to overcome scepticism that a software driver really can be safer than a wetware one.
監(jiān)管機構(gòu)和客戶都需要克服疑慮，即軟件驅(qū)動程序真的可以比人腦更安全。
From Porsche’s point of view, though, there is one other pertinent question.
不過，從保時捷的角度來看，還有另一個相關(guān)的問題。
Given that much of the reason for owning a sports car is for owners to show off what they perceive to be their driving skills, just how big a market will there be for a version where software takes those bragging rights away?
鑒于買跑車的主要原因是車主炫耀他們的駕駛技術(shù)，如果一個軟件剝奪了這些吹噓、炫耀的權(quán)利，它的市場還會有多大呢？
譯文由可可原創(chuàng)，僅供學習交流使用，未經(jīng)許可請勿轉(zhuǎn)載。
