Pull out any issue of Popular Science from the past 50 years, and you’ll likely find a story predicting that we would be living in a world of self-driving cars any decade now. (You can add in recent long-form pieces by other national media that push that Jetsons-tinged future even harder.)
The self-driving scenario is definitely much closer to reality now, as automakers gird their technologies toward that ultimate purpose. As we at Motor Trend have tested repeatedly, the technology is indeed here in nascent stages and is improving rapidly. The combined brainpower of Detroit, Stuttgart, Tokyo, and Silicon Valley will not be denied.
But the truth is we are still a long way from a fully self-driving society, for several very key reasons that have nothing to do with our ability to create the technology. Here is the cocktail party checklist of the interrelated barriers we face:
Infrastructure: Autonomous vehicles need roadways that are well-marked and in good shape. There are 4.12 million miles of road in America, according to the Federal Highway Administration, of which 2.68 million miles are paved. How bad are our roads? According to the FHA, 42.1 percent of Connecticut’s federal-aid highway miles are in “poor or mediocre condition.” Traffic-choked California is close behind, with 35.1 percent in terrible shape. Fixing that takes …
Money: Repairing infrastructure is the janitorial service of the industrial age. Those jobs ain’t sexy, but they are expensive. To merely keep our roads in their current lousy condition—replete with crumbled shoulders, degraded markings, and unfilled potholes—costs more than $100 billion every year. To invest in upgrading roads to autonomous-worthy standards would cost substantially more. Given the current antitax sentiment in America, I don’t see many people willingly opening their checkbooks for this. Perhaps the creation of private, autonomous-only roads would solve some of the problem. Would you pay an extra $50 for your self-driving car to platoon from L.A. to San Francisco at 125 mph on a private road adjacent to the 75-mph lanes piloted by humans? Some might, but given the lack of success of toll roads in many states, who in the private sector is willing to make that risky investment? And even if we find the hundreds of billions of dollars to make our roads worthy of autonomous cars, we have uncontrollable situations prompted by …
Weather: So far, no autonomous technology has proven capable of driving in snowy conditions that obscure the road markings self-driving cars require to navigate. The alternatives? Installing sensors in the road (read Money, above) or vastly improving mapping and navigation data—down to the inch. Ford has made progress with “self-locating by deduction,” but that still might not be enough to satisfy the …
Lawyers: As much as we laud Volvo for stating that they will assume the legal liability if one of their self-driving cars gets in an accident, other automakers might not be as beneficent. And until the tort lawyers and insurance industry sort out the at-fault versus no-fault circumstances, this technology grinds to a halt with the first fatal accident. And when you have a situation involving fatalities, you also have the situation of …
Ethics: How do you teach a computer to be self-aware, to acknowledge every possibility in the driving world? Will a computer-controlled car willingly sacrifice itself (and its driver) if it is presented with this impossible situation: to run over children mingling around their stalled bus in the middle of a blind corner, or to drive off the adjacent cliff? No executive has yet properly answered that question, saying it’s a theoretical situation. But that is the sort of theory that must be embraced and programmed into every car—and the autonomous cars’ occupants must be willing to accept the uncomfortable outcome generated by a software coder. Which leads us to …
Humanity: While we are in this lengthy transition phase, how autonomous cars interact with erratic, even impaired humans remains one of the biggest hurdles facing autonomous-car programmers. Some argue that a computer can more accurately predict unpredictable human behavior than another human; the default autonomous response so far is “slow down.” But people want autonomous cars to get them to their destination faster. Merely replacing America’s 250 million human-controlled cars with autonomous ones—crawling in the same mosh pit—should not be the ultimate goal.
Look, this advancement is going to happen. The ability to create autonomous vehicles is not at issue. At issue is how to incorporate 21st century technology into a world that is still mired in the 20th. And that will take time.