The early attempts to preview what roads will be like in the future, once self-driving cars of the sort Google is working on dominate, are pretty exciting.
Research already shows that even partial automation of driving could significantly reduce the 30,000 traffic deaths that occur each year in the US alone. Cars with “forward collision warning systems” that alert drivers or automatically hit the brakes are involved in far fewer crashes, suggest several studies. Worldwide, 1.2 million people are killed every year by cars.
Writing for Technology Review, Will Knight outlines the possibilities of our driverless car future, while also addressing the fact that it might be many decades before fully autonomous cars are possible or accepted. Those benefits could include:
- 90% of automobile accidents are due to human error, leading to 30,000 deaths and $300 billion in damages in the US alone. Fully autonomous vehicles could slash those losses by huge margins.
- Autonomous vehicles traveling in high-speed “platoons” that reduce aerodynamic drag could reduce fuel consumption by 20%.
- Up to four times as many vehicles could travel on existing highways if all vehicles were automated. The Texas Transportation Institute says traffic congestion wastes 5.5 billion hours and 2.9 billion gallons of fuel each year.
- Self-driving cars incorporated into car-sharing services like Zipcar could affordably transform cars from a thing people own to a service they call up on demand.
Getting to the point that the road is completely or even just mostly full of autonomous vehicles will be difficult, notes Knight. The biggest challenge of all could be that the moment when the sensors and decision-making abilities of an autonomous vehicle are overwhelmed and the car requires human intervention is a difficult transition to make. Lulled into a sense of security by a car’s self-driving systems, human drivers can be slow to react when a traffic situation requires their attention.
“We may have this terrible irony that when the car is driving autonomously it is much safer, but because of the inability of humans to get back in the loop it may ultimately be less safe,” Clifford Nass of Stanford University told Technology Review.