“10 million autonomous vehicles predicted to be on the roads by 2020” Business Insider.
In the wake of the most recent Waymo autonomous vehicle (AV) crash in Chandler Arizona, the AV market is currently in a very tenuous state. Leaders in the AV industry such as Waymo, Tesla, General Motors, and Ford are racing to create a system that the consumer can trust and utilize, while the National Transit and Safety Board (NTSB) are challenged to thoroughly investigate and regulate this emerging new technology that is poised to have high adoption by 2022.
According to the Boston Consulting Group by 2030, 20% of all driving will be done by AV’s. They are evolving faster than you might think. In order for vehicles and new companies to survive, they must gain the trust of the American people. Unfortunately, with each and every incident that occurs the AV market hits the brake, rather than rides a smooth road to early adoption. For those of you who are not aware, AV’s are the future of an easier and more convenient transportation experience. The real question is, when will that be? Although there have been many positives predictions, in the eyes of the consumer right now, those positives do not outweigh the recent deaths from AV’s. Citizens are not ready to give up their autonomy to driverless cars just yet. The relationship between humans and AV’s is comparable to animals in nature. Through survival of the fittest, only the best companies will be able to adapt and meet the expectations of trust and safety that consumers expect.
“As the technology is developed, autonomous driving could provide driving opportunities for the physically challenged or enable the elderly to continue driving longer. This will be vital as many nations experience an aging population.” William Ford Jr.
There are three major issues that have to be addressed before AV’s will be adopted by society. First, companies cannot be testing these devices on public roads. It is unethical to be risking lives while you are testing a product that may or may not be able to mix with non- autonomous traffic. Next, the next issue is what safety guidelines and regulations will be implemented by the NTSB to deem vehicles safe enough to reduce the current 56% of Americans who will not ride in a driverless car. With there being nearly 1.3 million car crash deaths per year, we have accepted the fact that every time we get into a car there is a chance we may not get out. That does not have to be the case when interacting with AV technology. In a perfect world, there would be no deaths from AV’s but that is a little bit too unrealistic. There needs to safety standards in place to determine whether artificial intelligence has evolved enough to be put on our streets for everyone to use and benefit from.
Self-driving cars run into a problem- Cars with drivers
If we lived in a 100% autonomous world, coexistence wouldn't be a problem… but until we get there, we need to find a way to make robots and humans share the roads safely.
A few accidents have already happened, and the latest one happened to a Waymo's self-driving car. Waymo, formerly the Google self-driving car project, crashed last week in Chandler, Arizona. “The vehicle was at the wrong place at the wrong time,” says Seth Tyler, a spokesperson for the Chandler Police Department.
Even though more people seem to be getting used to the idea of sharing roads with fully robotic vehicles, they are still very far from being considered trustworthy by the masses. The goal is to reach 5 nines of safety (99.999%), but incidents like Waymo’s crash, or the similar Navia crash in Las Vegas, or the self-driving Uber, are definitely slowing down that process.
Sharing the Road
There is an interesting relationship between the different ways that humans and AV’s think and make decisions. Unlike computers, humans change their decision-making processes based off of experience and intuition; they can adapt to their surroundings. In contrast, computers have a set list of predetermined decisions that it chooses based off of calculations that it makes within seconds. This can bring up some issues. For example, if an AV has 3 options of escape: to the left a child playing on the sidewalk, to the right a tree that will kill you, and in front, a car stopped short, what decision does the AV make?
By using cameras, lidars, and radars, autonomous cars know how to operate the vehicle, sensing other cars on the road as objects. But they lack subtle contextual understanding since they don’t have the intelligence to identify specific situations, such as someone nodding or waving.
There are some possibilities out there that could help solve that problem. In Seattle, the plan is to transform Interstate-5 between Seattle and