Dying Is Easy. Making A Self-Driving Car Is Hard. Tesla & Waymo At The Crossroads.
Tesla is in the process of rolling out firmware Version 9, which includes features which are supposed to unlock more of the self-driving potential backed into every car it builds. Nobody has been more adamant or passionate about the benefits of autonomous driving than Elon Musk, who has criticized negative news reports about crashes involving the use of Autopilot, claiming they will influence people not to use the system and thus drive less safely.
With nearly 40,000 people killed on American roads every year — often as the result of human error — Musk sees Autopilot as a vital tool for reducing carnage both in the US and around the world. “It’s really incredibly irresponsible of any journalists with integrity to write an article that would lead people to believe that autonomy is less safe,” Musk said during the Q1 2018 earnings call. “Because people might actually turn it off, and then die.”
While V9 is being downloaded and installed now, certain features are being held back — at least for a little while — according to a Musk tweet early yesterday morning.
V9 moving to wide release now. Holding back Autopilot drive on navigation for a few more weeks of validation. Extremely difficult to achieve a general solution for self-driving that works well everywhere.
— Elon Musk (@elonmusk) October 5, 2018
Indeed, it is. Waymo intends to begin offering commercial ride-sharing services in the Phoenix area soon, but its self-driving cars still have difficulty crossing an intersection near the facility where they are stored when not in use. Phoenix residents are complaining long and loud about Waymo cars that stop unexpectedly or maneuver in unexpected ways. They are also software limited to prevailing speed limits — something human drivers seldom pay more than cursory attention to.
Additionally, Waymo vehicles are tied to locations heavily and thoroughly mapped, something Tesla’s Autopilot-equipped vehicles do not worry about. Tesla’s Autopilot system is instead focused on deep learning and striving to drive like a human does, but with better vision and unbreakable attention.
Chip in a few dollars a month to help support independent cleantech coverage that helps to accelerate the cleantech revolution!
Tesla has just released its first quarterly safety report, a report Elon promised during the Q1 earnings call earlier this year. But not everyone found it all that helpful. According to a report in Wired, the report says drivers using Autopilot “registered one accident or crash-like event for every 3.34 million miles driven in which drivers had Autopilot engaged.” Tesla drivers who elected not to engage Autopilot registered one such event every 1.92 million miles.
NHTSA says a crash occurs every 492,000 miles of driving, but NHTSA has no data for “crash-like event.” The implication is that someone not driving a Tesla and using Autopilot is 7 times more likely to be involved in a crash, but critics complain the safety report offers no insight into the severity of the crashes, whether anyone involved was injured, what may have caused the crashes, or where and when they happened. In other words, the report is mostly an overview with little supporting detail.
Additionally, while “one accident or crash-like event for every 3.34 million miles driven in which drivers had Autopilot engaged” looks better than “an automobile crash every 492,000 miles,” the Autopilot miles were presumably driven primarily on highways, not in city traffic, and a crashes-per-mile statistic on just those types of roads was not provided — and may not be available.
Wired goes on, saying, “A more rigorous statistical analysis could separate daytime versus nighttime crashes, drunk drivers versus sober, clear skies versus snow, new cars versus clunkers, and so on. More context, more insight.” David Friedman is a former NHTSA official who now directs advocacy for Consumer Reports. He says, “It’s silly to call it a vehicle safety report. It’s a couple of data points which are clearly being released in order to try to back up previous statements, but it’s missing all the context and detail that you need.”
Why The Push For Autonomous Cars?
It’s one thing for Tesla, Waymo, and other companies to want to bring self-driving cars to market. They all point to the decrease in traffic injuries and fatalities that autonomous technology can make possible, but are we to assume that is the motivating factor in the push for cars that drive themselves?
Cruise Automation, the self-driving division of General Motors, is valued at nearly $15 billion even though it has only a few cars undergoing real-world testing at present. Uber — which makes nothing — is valued at $72 billion according to Statista.com. China’s Didi Chuxing is worth $56 billion. Both are worth more than Tesla? Morgan Stanley analysts Adam Jonas says Waymo is worth 10 times Tesla’s self-driving system.
That’s nuts. The answer, of course, is that analysts predict ride hailing will replace private car ownership in the near future. If that happens, the industry will generate profits totaling hundreds of billions of dollars every year. Saving lives is a worthy goal, but harvesting profits is what really gets people’s attention.
This is strictly a personal observation, which you are free to ignore or belittle if you like, but if the objective is truly to slash the number of deaths and injuries from car crashes, I imagine software that detects drunk driving could instantly cut the accident rate by half. Yet no one is pursuing such systems. Cadillac has just gotten a nod from Consumer Reports because its Super Cruise tracks eye movements. How hard would it be to let similar inputs identify drunk or impaired drivers and disable their vehicles after gently guiding them to the side of the road? {Answer: Not that hard.}
I’m talking to you specifically, Elon. Surely, with all the sensors and software in your cars, it should be a simple matter to detect evidence of impaired driving. Is it because so many people drive while inebriated? Are manufacturers afraid a significant number of people will decline to buy cars that will tattle on them if they stop off for a few quick ones on the way home? Is it because there is no financial reward associated with doing something that should be as simple as falling off a log? Or is it really that much harder than it seems? Inquiring minds want to know why drunk driving continues to be such a deadly scourge after all these years and despite the astonishing advances in artificial intelligence and sensors in the past few years.
Are Regulations The Answer?
ArsTechnica did a story recently that asked why federal and state regulators are so reluctant to oversee autonomous cars and self-driving systems. Mary Cummings, an engineering professor at Duke, says, “I don’t think there should be any driverless cars on the road. I think it’s unconscionable that no one is stipulating that testing needs to be done before they’re put on the road.” [Technically, testing is being done, which makes this comment a little confusing, but let’s move on.]
ArsTechnica used the example of airplanes, which are tightly regulated by the FAA. Getting approval for a new aircraft takes 8 years on average and costs upwards of $100 million. Are airplane crashes any more lethal than car crashes? Dead is dead. “Partly that’s because federal regulators don’t want to slow the introduction of a technology that could save a lot of lives in the long run. Partly it’s because they believe that liability concerns give companies a strong enough incentive to behave responsibly. And partly it’s because no one is sure how to regulate self-driving cars effectively.”
Bryant Walker Smith, a legal scholar at the University of South Carolina, says, “there’s no consensus on what it means to be safe or how we go about proving that.” Self-driving advocates argue that regulations would slow down the implementation of technology that can save lives. But if all we have are glossy overviews like the first Tesla safety report, how can we have confidence that the systems are truly working as advertised? It seems some standardized reporting requirements that all companies need to comply with would not be too much to ask for.
The Bottom Line
When all is said and done, self-driving technology depends on billions of miles of experience. In that regard, Tesla is light years ahead of the competition no matter what Consumer Reports has to say on the matter. If you are wondering whether to buy shares in Tesla, bear that in mind. One could argue its lead in autonomous technology itself justifies its current stock price. Everything else is just icing on the cake. [Not stock-picking advice and please consult an investment professional rather than throwing cash down on companies that impress us.]
Have a tip for CleanTechnica? Want to advertise? Want to suggest a guest for our CleanTech Talk podcast? Contact us here.
Latest CleanTechnica.TV Video
CleanTechnica uses affiliate links. See our policy here.
