xml:space="preserve">
xml:space="preserve">
Advertisement
Advertisement

The driverless future is still a long way off

WARNING. Police released a video that shows an Uber robot car running into a woman across a highway with her bicycle in Tempe Ariz. Sunday night.

Much public attention has been paid to self-driving and driverless vehicles, especially after an Uber self-driving vehicle recently killed a pedestrian in Arizona. While both terms have often been used interchangeably, they are different. “Self-driving” means some level of autonomy or automation, but the driver is expected to drive under certain circumstances and road conditions. The vehicle may prompt driver involvement. “Driverless” means that there is no human intervention except to input the destination; the vehicle does the rest. A driverless vehicle would have no steering wheel or pedals. Self-driving vehicles are already in use.

Cristina Perez Hesano, an attorney for the daughter and husband of 49-year-old Elaine Herzberg, responded Thursday to inquiries by The Associated Press by saying only that the matter "has been resolved."

In general, except for high-end vehicles, the technologies may best be described as “driver assistance,” such as adaptive cruise control that maintains safe distances. The level of automation will grow with innovation; right now we are at about level 2 on a scale of 1 to 5. However, driverless vehicles are many years off, 10-20, perhaps even more. With regard to light and heavy-duty vehicles, truck drivers will not be losing their jobs anytime soon.Technological feasibility continues to evolve, but driverless passenger and commercial vehicles would have to learn every possible driving scenario, and the legal, public investment and consumer acceptance obstacles are many. It’s obvious that those who can’t or won’t drive would be likely users of driverless passenger vehicles, personal and shared-ride, and as the population ages, there will be more such people.

Advertisement

Though the death of a pedestrian from a self-driving SUV has shaken the autonomous car industry, driverless cars still have remarkable potential to make streets and highways safer than they’ve ever been.

But many drivers will resist being herded into driverless vehicles, and how would you force them? Able-bodied drivers would likely not ride in a vehicle that always obeys the 55 miles-per-hour speed limit. According to a transportation colleague from Germany, automakers there are not seriously planning driverless vehicles. I’m guessing that BMW does not intend to be “the ultimate pod that slowly takes you from point A to point B” instead of “the ultimate driving machine.” Should driverless vehicles be programmed to break the speed limit? Would government or the insurance industry allow that? Would we be more or less liable in a crash? How should a driverless vehicle facing a crash scenario react? How would a driverless vehicle decide about the likely severity of damage when making its choice?

The hubris of the robot carmakers has fed the blind enthusiasm of government officials who have embraced the promise of the technology but largely ignored its perils.

Self-driving and driverless vehicles could provide significant safety benefits. The great majority of crashes — note, I don’t call them “accidents” — are caused by human error. It’s been estimated that connected vehicles, self-driving or driverless vehicles that communicate with each other, will reduce non-impaired crashes by 80 percent, saving many lives. Such a reduction in crashes will dramatically lessen the nonrecurring congestion that can slow our commutes to a crawl. But driverless vehicles would have to have flawless technology to indeed make them safer than human drivers.

Advertisement
Truly driverless cars — like this one from Google — are still many years away from being the norm on American roads.
Truly driverless cars — like this one from Google — are still many years away from being the norm on American roads. (Tony Avelar / AP)

Managing a road network with mixed driver-operated and fully self-driving/driverless/connected vehicles will be complicated, and the greatest reduction in crashes will not happen until just about all vehicles are so advanced. It takes time for the vehicle fleet to “turn over;” currently, the average vehicle on the road is 12 years old. The road infrastructure would also have to be improved greatly and be more actively managed.

A national online survey conducted at Morgan State University asked drivers what their acceptance of and willingness to pay for connected vehicle safety technologies would be. The survey showed that while consumer acceptance of such technologies was high, the expected prices were a significant obstacle. Let’s be clear, a fully self-driving or driverless, connected vehicle will be quite expensive, particularly during early phases of deployment. It’s been suggested that prices for safety technologies would come down over time, just as they did for computers and mobile phones. Computers and phones are no longer cheap; their capabilities and pricing have changed. While safety technologies do “trickle down” to modestly priced vehicles, when have motor vehicle prices experienced a downward trend?

While fully self-driving and driverless vehicles stimulate the imagination, there are uncertainties beyond the necessary technological innovations. Many legal, cultural and affordability issues have to be resolved before we see them abundantly on our roads.

Z. Andrew Farkas is director of the National Transportation Center and professor of transportation at Morgan State University. His email is andrew.farkas@morgan.edu.

Recommended on Baltimore Sun

Advertisement
Advertisement