The standard response from the AV tech industry to the question of “Why self-driving?” is to save lives. Really?
“The closer you look, the less you see!”
Movie aficionados everywhere may recognize this line from the 2013 film Now You See Me. I’ve been thinking about those words a lot just recently, because the closer I look at all things “self-driving,” the less I see an obvious outcome. Read on if you want to delve further into that idea with me.
Let’s start at the very beginning; a very good place to start — why do we need “self-driving” technology at all? Be clear, I didn’t ask if it is cool, nor if videos of self-steering or self-braking cars create amazing PR to secure new investors. I asked: Why?
The best media trained CEOs in the AV tech industry subtly deflect “why?” by answering “when?” or “what?” I think we are all done hearing timelines for “self-driving” vaporware product launches in the next three years, while the myriad of “what?” ideas is summarized neatly by this piece in Forbes.
There are many competing visions of how AV technology could be applied and the last couple of years at CES have showered us with a future of flying robo-taxis, “self-driving” pizza delivery vans and even robo-dogs. Just pause for a moment to consider what percentage of the global population is sufficiently wealthy to use such products and we can see how far the tech industry has lost its way. Wouldn’t this talent and these resources be better directed to helping treat cancer or heart disease?
These vanity projects come from what I call the Trough of Enlightenment — the cesspit of truly stupid thinking, espoused by serial entrepreneurs with more Vision funding than sense, and drunk on the supply of virtually free money from central banks and sovereign wealth funds.
Most of these vain ideas suck and serve no purpose, other than to incinerate capital. Step forward enlightened early stage WeWork investors and most VCs backing the hundred or so lidar companies chasing “autonomy.” Some of these entrepreneurs would have created more value if they had actually just set fire to the funds.
The standard response from the AV tech industry to the question of “Why self-driving?” is to save lives. Really? I always thought it was about making money. Figures from the World Health Organization indicate that globally more than 1.3 million people die on our roads and highways each year — by far the highest number of fatalities are in India and China, together totaling more than 0.5 million. For the U.S. about 37,000 lives are lost annually, just a couple of percent of the global total.
Yes, there are too many lives lost, but we know the main causes are related to human failings such as distraction; fatigue; driving while intoxicated by alcohol or drugs; and speeding. We won’t succeed in reducing the number of fatalities by replacing human drivers with non-existent “self-driving” AVs. We will succeed by making human drivers into safer drivers using ADAS and camera-based driver monitoring — technology which is tested, validated and suitable for deployment at scale today.
If we are serious about road safety, we need to install this technology into every new vehicle with four or more wheels (cars, vans, trucks, coaches and buses) as soon as is practically possible on automotive OEM timescales. Silicon Valley tech companies and VC-funded start-ups have no meaningful role to play in this initiative. If you want to see decisive activity regarding road safety, you need to look to Europe, in particular the role played by Euro NCAP and the European Council.
Starting in 2022, to secure the maximum five-star rating from Euro NCAP’s testing process will require a vehicle model to have safety systems fitted as standard encompassing longitudinal speed assist (autonomous emergency braking), lateral lane support (lane keep assistance) and driver monitoring systems (almost certainly using 940nm IR computer vision to meet the testing protocols). Where Euro NCAP leads, other regional NCAPs will quickly follow as these testing standards are subsequently promoted by Global NCAP.
Under laws now passed by the European Parliament, no new vehicle model will achieve type-approval after mid-2024 without all of the aforementioned safety systems and intelligent speed assistance and accident data recorders too. From mid-2026 it will be against the law to sell a single new vehicle in Europe without all of these systems installed. Yes, Elon, that will include Tesla too.
NHTSA has not announced any equivalent plans to improve road safety standards in the U.S. However, the future of road safety is clear: Mandatory installation of ADAS and driver monitoring in all new vehicles.
Lawmakers and regulators should do that first and then let the market decide if “self-driving” anything is advantageous, or even desired. The tech industry may find a large percentage of the global population doesn’t care about AVs, any more than they care about flying cars and robo-dogs.
Now You See Me is a fun movie about magic and illusion — and the first rule of magic is always be the smartest person in the room. But there is only suffering in the tragedy of more than 1.3 million people losing their lives on our roads each year. So, over to you lawmakers and regulators to expedite the installation of ADAS and driver monitoring. Now that really would be smart — and a magical outcome too.
— Colin Barnden is principal analyst at Semicast Research.