Why self-driving cars haven’t disrupted the auto industry like they were supposed to
“The era of automated driving is nearly here,” Audi proclaimed in a January 2015 press release. After one of its A7 models completed a 550-mile self-driven journey from Silicon Valley, California to Las Vegas, Nevada, the company was seemingly ready to usher in a completely new age of self-driving transportation.
Just five years ago, carmakers and tech giants alike began large public relations campaigns touting their plans for self-driving cars. These cars, powered by tech breakthroughs, were positioned as virtual panaceas. They’d be capable of alleviating — if not outright eliminating — freeway congestion as well as bringing roadway fatalities down to statistical zero.
What’s more, they’d all be powered by electricity, further benefiting the environment as well as humanity — all by 2020.
Fast forward to today and talks of self-driving cars are a whisper of what they once were. After all, there are no self-driving cars being sold in showrooms and only a few thousand are currently undergoing testing in the U.S.
But car companies and tech innovators haven’t given up on self-driving tech. Instead, the lessons they learned over the last half-decade have forced them to shift focus on where and how self-driving technology is implemented.
Why buy an $85,000 self-driving car if it can only be driven inside your city and nowhere else?
What is autonomy?
Before getting into the current impacts autonomy is making on our lives, we ought to briefly cover the various capability levels of self-driving vehicles.
The International Society of Automotive Engineers (SAE) defines automated driving in levels 0 through 5, with each level increasing in technological power. Level 0, for example, encompasses relative common car features like automatic emergency braking and blind-spot monitoring, while Level 5 scales all the way up to the theoretical vehicle that could drive itself anywhere, under any conditions.
When we think realistically about the self-driving cars in our future, we're mostly thinking about Level 4 vehicles — that would encompass inventions like self-driving taxis, aka "robotaxis." These cars may or may not have driving controls, like a steering wheel and pedals, but would be able to drive themselves under certain conditions without needing humans to take over. But two factors currently stand in the way: technical complexity and cost.
Currently, Level 4 self-driving systems can cost upwards of $200,000 each in prototype form. And while mass production could eventually bring down the cost of the sensors and systems that enable the technology, asking everyday consumers to plunk down an extra $50,000 (or more) for a self-driving car isn’t feasible. Nor is it feasible for companies to yet produce a self-driving car that safely operates in any city or environment. Why buy an $85,000 self-driving car if it can only be driven inside your city and nowhere else?
With these concerns in mind, companies have looked for alternate ways to spread out the cost of automated driving systems and implement them in other products. Specifically, they have shifted focus to lower-speed, lower-risk vehicles and business models.
What kind of self-driving vehicles are currently on the roads?
The business model consumers might be most familiar with is ride-hailing robotaxis.
There are two current leaders in the robotaxi market: Waymo (Google’s self-driving tech arm) and General Motors’ brand Cruise Automation. Both operate with Level 4 autonomous vehicles.
Waymo launched its first commercial robotaxi service, called Waymo One, in Phoenix, Arizona, in December 2018. To date, Waymo One serves more than one thousand riders daily and has clocked 10 million autonomously-driven miles on public roads — and over 10 billion miles in simulation.
Like Uber or Lyft, riders can order up a car through an app 24/7. Unlike Uber or Lyft, however, the cars drive themselves — though there is a human safety driver behind the wheel. What’s more, Waymo One’s vehicles are all identical Chrysler Pacifica minivans.
Meanwhile, in 2017, Cruise said that it would launch its commercial robotaxi-based ride-hailing service in San Francisco some time in 2019. However, in late July of this year, Cruise CEO Dan Ammann publicly readjusted that timeline, pushing the launch timeline past the end of 2019.
“With such high stakes, our first deployment needs to be done right and we will only deploy when we can demonstrate that we will have a net positive impact on safety on our roads,” Ammann wrote on Medium.
Ammann is clearly addressing the elephant in the room: safety.
When carmakers and tech giants first began discussing autonomy, consumer cars were the primary focus. But since 2016, several Tesla drivers have been killed while the brand’s semi-autonomous Autopilot system was engaged. In 2018, a pedestrian was struck and killed by an Uber self-driving test vehicle in Tempe, Arizona.
Understandably, public concerns over the safety of self-driving technology prompted companies to pump the brakes on plans to sell self-driving cars to the public.
As a result, many other brands developing self-driving systems have shifted toward lower-risk modes of transport. It is in these markets that self-driving tech is rapidly growing and affecting our daily lives most — whether you notice it or not.
What’s the simplest way to limit risk? Go slow. That’s the driving idea behind another form of autonomous tech: driverless pods, which could either take the form of larger people transporters (think driverless buses or trams) and smaller goods-delivery pods.
One of the leading names in the self-driving transporter market is Local Motors. For years, Local Motors has been running trial tests of its driverless bus-like vehicle, Olli, around the globe.
These include transporting people around a beach in Australia; the campus of California State University at Sacramento and at California’s state fairgrounds; a business park in Rancho Cordova, California; and, most recently, around the Joint Base Meyer-Henderson Hall military base near Washington, D.C.
“The possibilities for how Olli can be used are endless, and its success in a variety of settings illustrates that,” Local Motors executive vice president Matt Rivett said. “Looking ahead, as regulations continue to evolve to allow for self-driving vehicles on public roads, Olli has the potential to not only serve the first and last mile, but to offer a safe, sustainable, responsive option for mobility.”
Another way to limit risk is to eliminate the human altogether — and not just from the driver seat. That’s where delivery pods come into play.
One of the leading companies in driverless pod technology is called Nuro, which has developed two compact vehicles called R1 and R2. The company is running trial programs with Kroger and Domino's locations in Arizona and Texas.
Nuro is like any grocery or pizza delivery service; customers order their goods on the respective company’s app. The Nuro pods, the size of a large desk, slowly and autonomously trundle their way from the store to the customer’s home. Upon the Nuro’s arrival, the app provides the customer with a passcode that they use to unlock the pod and retrieve their order.
Once the goods are retrieved, the all-electric Nuro silently and autonomously returns back to the store; since August 2018, the pods have successfully made thousands of deliveries.
Semi-trucks and vans
Although these are the least sexy form of automated driving, trucks and vans are growing most quickly — and also have the biggest chance of affecting our daily lives.
Programming a self-driving vehicle that can handle any situation safely is incredibly difficult. But designing one to perform a repeatable task on a closed property or lightly trafficked public road is much simpler. That’s where self-driving vans and semis come in — because they tend to spend a great deal of time running either the same repeatable routes or on open highways, these types of vehicles are perfect for autonomous tech.
Over the last six years, carmakers and tech companies have had to back off their utopian vision of self-driving vehicles.
Trucks like Volvo Trucks’ Vera prototype vehicle, which is designed to haul shipping containers, and the self-driving vans being utilized by Walmart to move goods from warehouses to fulfillment centers, are already improving how quickly and efficiently goods get delivered to our doors. These vehicles can drive all day and night without incident in a way that human drivers can't, enabling companies to fulfill shipping more quickly and reliably. And who doesn’t want to get their online order delivered more speedily?
Autonomy in your life today
No doubt, over the last six years, carmakers and tech companies have had to back off their utopian vision of self-driving vehicles. That’s because the initial concept of self-driving cars for quotidian customers has proven unrealistic.
That said, the technological and developmental leaps brands made have been refocused and reimplemented to improve our lives in other under-the-radar ways.
And the programs discussed here are just the beginning. For example, many other companies have plans for their very own ride-sharing robotaxi service in the future, including carmakers like Tesla and Hyundai as well as the ride-hailing heavies Uber and Lyft.
Local Motors’ Olli and Nuro’s delivery pods won’t be the only players in those spaces either. GM and Toyota have plans for people-mover and delivery pods as well, which will expand to airports as well as company and college campuses around the globe in short order.
Self-driving electric semi-trucks are widely expected to become the standard in shipping and trucking by the end of the next decade. Meanwhile, autonomous vans will be moving merchandise from warehouses to fulfillment centers in increasing numbers, getting our goods to us more quickly.
Autonomy might not quite resemble what we were promised just a few years ago. That doesn’t mean, however, that it’s not alive and well.