Rumors of an Apple iCar are becoming more nuanced, with various reports last week that Apple’s Special Project group has been negotiating with officials of The Contra Costa Transportation Authority (CCTA), custodians of the GoMentum Station — part of the 5,000-acre decommissioned U.S. Navy Concord naval weapons station located some 30 miles northwest of San Francisco. Apple is reportedly interested in scheduling use of the facility’s 20 or so miles of paved roadways and streets that several auto manufacturers such as Daimler and Honda (YouTube video) have been using as a military-guarded secure testing range for autonomous vehicles.
The Guardian’s Mark Harris reported Friday that the Apple car project appears to be further along than many had suspected, and that correspondence obtained by The Guardian under a public records act request included Apple engineer Frank Fearon noting that the company wants to gain understanding of “timing and availability for the [GoMentum] space, and how they would need to coordinate around other parties who would be using [it].”
That seems to indicate that Apple’s car project is strongly oriented toward driverless vehicles, which was probably inevitable, but more confirmation doesn’t bring gladness to my heart.
According to metrics cited by Ford Motor Company, 50 percent of young people polled have no fear about the introduction of self-driving cars, and 53 percent actually say they would be more comfortable travelling in a robot-driven vehicle than one driven by themselves.
These are profoundly distressing factoids for automobile enthusiasts, and the looming prospect of roads dominated by robot cars certainly has the car culture community freaked out. This summer many car magazine column inches have been dedicated to analyzing and speculating about what autonomous vehicles portend for the future of our sport, hobby and nexus of enthusiasm, and the prognosis is generally negative.
Consequently, it’s encouraging to hear members of the Apple community expressing apprehension about the autonomous car bandwagon-jumping that’s going on these days.
Earlier this month, the formidable Mac360 commentator Bambi Brennan weighed in with a column entitled: “Sorry, Apple. Sorry, Google. Im Not Going To Buy Your Self-Driving Cars,” expressing her skepticism about the relentless stream of news stories and commentaries about the wonders of self-driving cars and trucks and the benefits they’ll bring to mankind, affirming “I dont buy it now. I don’t want to buy it in the future.” Bravo Ms. Brennan. Me neither.
Some reasons to shun self-driving vehicles are obvious, at least for those of us of generations not totally baked into tech-immersion space. First and foremost let me join with Ms. Brennan in a rousing chorus of “I like to drive.” Personally, my various automotive enthusiasms predate my !interest in information technology by about three decades, and are much more deeply engrained. As automobile Magazine editor Mike Floyd affirmed recently, “We drive for pleasure, to enjoy the experience, to appreciate the craftsmanship of a finely tuned automobile.”
And we have plenty of company in our autonomous autos skepticism. Of course not all of us, even among the over-30 generations, is an automobile aficionado by a long shot. However in their latest periodic survey of levels of receptiveness to the concept of self-driving cars, including preferences among levels of vehicle automation, preferences for interacting with, and overall concern about riding in self-driving vehicles among 505 licensed drivers over age 18, University of Michigan Transportation Research Institute (UMTRI) researchers Brandon Schoettle and Dr. Michael Sivak — respectively a Project Manager and a Research Professor in UMTRI’s Human Factors Group at Ann Arbor, Michigan, found among other things that:
• The most frequent preference for vehicle automation was for no self-driving capability, followed by partially self-driving vehicles, with completely self-driving vehicles being the least preferred choice. Overall, just 15.6 percent of respondents expressed preference for completely autonomous vehicles as opposed to 43.8 percent who say they want no self-driving capability at all and 40.6 percent who would be willing to accept partial self-driving vehicle capability.
• Concern for riding in self-driving vehicles was higher for completely self-driving vehicles than for partially self-driving vehicles.
• Respondents overwhelmingly (96.2 percent) want to be able to manually control even completely self-driving capable vehicles when desired.
• Preferences were generally divided between touchscreens or voice commands to input route or destination information for completely self-driving vehicles.
• Most respondents prefer to be notified of the need to take control of a partially self-driving vehicle with a combination of sound, vibration, and visual warnings.
Schoettle and Sivak found levels of concern for riding in completely self-driving vehicles in this 2015 study similar to those observed in their previous survey on the topic conducted in June 2014, and conclude that currently, as in the previous study, concern about riding in completely self-driving vehicles remains high, although they found younger respondents more accepting of self-driving cars than older drivers. Interestingly, female respondents expressed stronger opposition to self-driving cars than did males, with 47.3 percent of females favoring no self-driving technology at all.
On the other hand, Car and Driver magazine’s editor Eddie Alterman cited a Washington Post report a while back that today’s hard-texting, IT-obsessed youth are inclined to dismiss driving as seriously lame, with only about 30 percent of 16-year-olds having even bothered to acquire driving licenses according to 2008 research. That’s a “distressing” statistic says Alterman, and as a lifelong automobile aficionado this writer concurs. When I was sixteen, getting a driver’s license was a near-universal rite of passage. I took my driving test and passed six days after my sixteenth birthday.
Secondly, driverless cars are heavily dependent on mapping and road marking/signage, which will highly restrict their usefulness and dependability outside major thoroughfares, to say nothing of markings and signs disappearing under heavy snow and ice accumulation on highways and behind high snowbanks in much of the U.S. and nearly all of Canada for weeks or even months at a time, as happened where I live last winter. Computer systems that could be programmed to cope adequately with those conditions and other forms of extreme weather, not to mention potholes, or the ad hoc nature of construction zone signage, or a whole spectrum of often unpredictable variables, is far beyond the limitations of present technology in operating systems, cameras and sensors. For example, the back-up camera in my Ford F-150 quickly becomes obscured by deposits when driving on salty, dusty, and wet roads, as would any optical sensor for self-driving systems.
Also, computer systems are only as good as the programming humans code into them. Automobile Magazine’s Preston Lerner cites Stanford University mechanical engineering professor and director of the Center for Automotive Research at Stanford (CARS) Christian Gerdes observing: “You’re not going to eliminate human error. You’re just shifting it from the driver to the programmer,” adding: “the more research we do with human beings, the more impressed I am with our eyes and ears.”
Prior to joining Stanford, Professor Gerdes was the project leader for vehicle dynamics at the Vehicle Systems Technology Center of Daimler-Benz Research and Technology North America. Daimler is another major automaker working on autonomous cars, (and testing at GoMentum) but also semi-autonomous heavy highway trucks. However as for robot 18-wheelers roaming the highways, Daimler says that’s a long time in the future, if ever. As Daimler Trucks North America president and CEO Martin Daum told Automobile mag, “The human brain is still the best computer money can’t buy.”
Thirdly, robot cars will be hacked, as some non-robot Fiat-Chrysler’s Jeep models were recently. The hack-proof computer system has yet to be invented. As Ms. Brennan astutely observes, “If the U.S. government and major corporations and banks around the world cannot protect their customers data from hackers, how will an internet connected computerized driverless car avoid being hacked?”
Autonomous car advocates extol a supposed enhancement of highway safety with robot controlled vehicles, and that’s a commendable objective, but at what cost in terms of human freedom, even if a statistical safety improvement is achieved? There is life skills value in learning how to drive well. The problem is that admittedly too few people do, no thanks to the bizarre disconnect from reality reflected in virtually all state and provincial driver’s license qualification bureaucracies, which do a reasonably good job of teaching the rules of the road, and how to parallel park, but in most cases not a whit about vehicle handling and control in challenging road conditions.
The biggest road safety issue these days is driver distraction, both from portable devices and from in-car infotainment and control systems. The problem, as Automobile Magazine’ Arthur Saint Antoine elegantly parsed it in a recent column entitled “The Next Extinction: Drivers” (July 2015), is “every arrogant, self–absorbed twit who willingly puts the lives of others at risk because he’s too busy checking the ‘likes’ on his Facebook page to leave for his scheduled appointment on time,” and once behind the wheel “behaves as if cars are autonomous already, casually abandoning his legal and moral duty — watching the road –- to stuff his face into his precious smartphone.”
There’s more. Robocars will presumably be programmed to never break the law, which means they will be unable to break posted speed limits under any circumstances. Advocates contend this will make for a safer highway environment. I emphatically disagree.
Consider the following scenario. You are stuck on a two-lane behind someone doggedly determined to drive five or ten miles per hour below the limit. Perhaps a large truck or a RV. Finally your robocar pulls out to pass the moving roadblock, but can proceed in the lane facing oncoming traffic at no greater rate of speed than the legal limit, since that is the letter of the law in many jurisdictions, including the one I live in, even though it’s idiotic and a formula for creating dangerous situations in real world circumstances.
I want to spend as little time as possible in the oncoming traffic’s lane, and use the V8 thrust in my current rides to accelerate past cars I overtake and tuck back into my own lane with dispatch, even though that typically means short bursts of exceeding the speed limit — sometimes by a substantial margin — which I consider the safest passing practice notwithstanding the unqualified “Speed Kills” notion. In circumstances like the one described, speeding combined with skilled vehicle control is the safest modality.
In her article, Ms. Brennan ponders safety and ethics issues associated with driverless cars, such as a child running into the street from between two parked cars — a circumstance an alert human driver could have anticipated as possible based on visual analysis, and been prepared for.
Then there’s the matter of applied ethics in rock-and-a-hard-place circumstances where all potential outcomes are bad, but some are more moral than others, such as the hypothetical of a large oncoming truck swerving into your lane leaving you with four no-win alternatives; a) collide head on with the truck likely killing you and all occupants in your vehicle and possibly people in the truck as well, b) swerve onto the sidewalk to your left killing an elderly pedestrian, c) swing to the right and kill two children, or d) steer into a tree or other solid object killing only yourself and your passengers. We might deduce that It’s generally more ethical to harm fewer people than more, or have one person die instead of three or five, but do you trust a drone driver’s computer program to make a best-of-a-bad-lot decision in a split second in such instances where almost infinite combinations of variable circumstances may obtain? Seriously?
And how about product liability, with the onus placed on large corporations to intentionally program car-control computers’ responses to make life-or-death ethical decisions in right now highway situations, and the prospect of mega-million dollar lawsuits arising therefrom, which can be expected to put a damper on autonomous car deployment? Autonomous vehicle product liability lawsuits will involve huge amounts because in the case of autonomous cars, the manufacturer (or programmer?)rather than the driver will be blamed. Human rights lawyers would have a field-day with user-adjustable ethics algorithms potentially weighting of different values to be applied in crash incidents, with manufacturers providing preference settings that could potentially predispose a hierarchy of gender, age, ethnicity and so forth in making determinations about who lives or dies in unavoidable crashes.
While there is an apparently near consensus assumption afoot in the tech community that self-driving cars becoming a consumer reality is both inevitable and will be a good thing, a white paper by the University of Michigan Transportation Research Institute’s Schoettle and Sivak investigated several important issues associated with autonomous autos. And while it’s not the central focus of the paper, entitled: “Potential Impact of Self-Driving Vehicles on Household Vehicle Demand and Usage (Schoettle and Sivak), Report No. UMTRI-2015-3, February 2015,” the report observes that there are still vast technological gaps that must be bridged before self-driving vehicles are turned loose on streets and highways.
Schoettle and Sivak, citing U.S. light-duty vehicle statistics, note that it currently takes 25 years before the proportion of cars 25 years old or older declines to 13.5 percent of vehicles in use, meaning that self-driving cars will be sharing roads for decades to come with cars driven by humans and quite possibly not “connected.” “There will likely be at least a several-decade-long period during which conventional and self-driving cars would need to interact,” say the coauthors. “Furthermore, to the extent that some people may want to drive only conventional vehicles, this overlapping period might last indefinitely.” It will as long as I have any say in the matter.
I acknowledge that we foot-draggers are bucking a trend, and that resistance to popular autonomous car aspirations may ultimately be futile, but while human drivers may be destined to eventually go the way of human elevator-operators, the steering wheel won’t be pried from our hands for a long time yet.