Last year we wrote about self-driving vehicles in the distribution industry. Since then the technology has advanced by leaps and bounds. Autonomous vehicles of all kinds are now sharing the roads with us. And the odds that you’ll pass one during your daily commute are rising all the time.
But what about the news reports we’ve all seen? The first crash of a self-driving car made international headlines. So did the fatal crash of an electric car running on Tesla’s Autopilot system.
Are self-driving vehicles dangerous? We’ll take a closer look at these cases to find out.
In July of 2015, three Google employees sustained minor injuries while riding in a self-driving car.
In May of 2016, Joshua Brown of Ohio was killed in an accident while traveling in his Tesla Model S electric sedan. The vehicle’s Autopilot system was engaged at the time of the crash.
In March of this year, a self-driving SUV run by Uber was involved in a collision in Tempe, AZ. Uber’s vehicle hit a light pole, rolled onto its side, and bumped two cars stopped in traffic on the other side of the intersection.
These examples might make you worry that self-driving cars are more dangerous than cars driven by humans. But a closer look at the circumstances of each crash reveals the opposite.
In the rear-end collision that resulted in injuries, the Google car was not at fault. The other driver caused the crash by failing to brake at a traffic light.
In the tragic death of Joshua Brown, the National Transportation Safety
Board determined that he ignored the Autopilot system’s repeated warnings that he must put his hands on the steering wheel.
And human error caused the crash of Uber’s self-driving SUV, not the technology. The human driver of the second vehicle failed to yield the right of way.
In each of these cases, human error caused the collision, not the autonomous vehicle. But self-driving cars aren’t perfect. There is one case of a self-driving vehicle causing a crash. On February 14, 2016, a self-driving Google car collided with a municipal bus at low speeds. Thankfully, there were no injuries.
According to Google, the self-driving car caused the crash when it “decided” to merge into the next lane to avoid sand bags blocking the road. The program judged that the bus traveling in the next lane would slow and allow it to merge.
Meanwhile, the bus driver assumed that the car would remain in its own lane until the bus had passed. Because of this mutual misunderstanding, the car struck the side of the bus at about 2 MPH.
The number of collisions involving self-driving vehicles is tiny compared to the total number of accidents every year. In the first six months of 2016 alone, 2.3 million people were injured in traffic accidents, and 19,000 people were killed.
Distribution firms are so impressed with self-driving technology that they’re already using it on our roads. On October 25, 2016, Uber and Anheuser-Busch partnered to make the first commercial delivery by autonomous vehicle. A self-driving tractor-trailer traveled 120 miles and made a beer delivery. Meanwhile, the human driver stayed in the sleeper cab.
The possibilities for self-driving vehicles go beyond commercial applications. Officials in Colorado are using it to save lives. In the fall of 2017 Colorado will be adding a self-driving truck to its road crews.
A “crash truck” is a vehicle that follows behind workers while they perform road maintenance. This special truck is designed to shield workers from inattentive drivers who fail to stop.
Crash trucks can absorb the impact of cars speeding at up to 75 MPH. But driving one of these vehicles is still exceptionally dangerous.
To meet this need for greater road crew safety, vehicle firm Kratos Defense designed an autonomous crash truck. The system uses two vehicles. The first has a human operator who drives ahead of the crash truck and guides it forward using radio waves. This system ensures that road crews stay protected from speeding vehicles, and all without a human driver being put at risk.
Whenever a driver gets behind the wheel, he or she must make dozens of predictions about what other drivers are likely to do in a given situation. Self-driving cars use algorithms to accomplish this goal rather than intuition and common sense.
While humans and autonomous vehicles share the roads, uncertainly will always be in play. But the day might come when humans forgo driving altogether. If that happens, every vehicle will behave with the clockwork predictability of any other machine.
Will traffic accidents then become a relic of the past? With the current state of the technology, it’s impossible to know. But it is a hopeful idea.
Sources: Bloomberg.com, CNN Money, Theverge.com, Wikipedia, NYTimes.com, Newsweek.com, USAToday.com, Stuff.co.nz