Select Page

Could connectivity be a threat to autonomous cars?

by | Feb 28, 2022 | 5G, Connectivity, V2X

In his article “The Myth of 5G and Driverless Cars”, Doug Dawson, president of CCG Consulting, shares his thoughts on 5G and self-driving vehicles. We asked him about the roadblocks to autonomous cars. He addresses the various connectivity approaches and their limitations, such as consumer running costs and environmental interferences that may prevent cars from communicating.

This post was produced in collaboration with The5GExchange.com

Diana Blass and Jo Kassis Photo

A story by Diana Blass and Jo Kassis.

Diana is Kurrant Insights’ Editor-in-Chief and video director. She is passionate about technologies that have a real impact on people.

Jo is an IoT-focused content strategist. He creates in-depth reports to keep our readers up to date on the newest industry developments.

Autonomous cars are one of the most talked about use cases of 5G, but telecommunication consultant Doug Dawson, President at CCG Consulting has a different view. He says 5G and driverless cars are nothing more than a mess. And he tell us why…

What are the challenges when it comes to building out a 5G network for autonomous cars?

Smart cars have three different paths to take for communications, one is they can be on 5G. Another one is completely the opposite, where they just talk to whoever else they can see, car to car. The third approach is to put a supercomputer in the car, so it doesn’t need help from anybody.

You look around to all the folks developing autonomous cars and there’s somebody trying each of those three approaches. The problem of them using any spectrum is all these cars are going to be interfering with something else around. And there are certain places in the world that you just can’t interfere. We don’t want cars using frequencies to talk to each other when they’re in the waiting line at the airport because the airports do not want interference in their spectrum. And there’s a whole bunch of other places just like that.

Millions of miles of paved roads is a really giant number and to get really good spectrum along all those roads. Who’s going to pay for that? Are we going to charge each car $40 a month for the whole United States to pay for that cellular network? When you go to cellular companies’ annual reports, they still say one of the possible uses for 5G is smart cars, because they just can’t let go of that big dream of every car paying that money.

But we know that people don’t really do that. You know, every new car nowadays comes with that satellite service for the emergencies. They usually give you a one or two years free but 90% of the people don’t pay once it runs out. People do not like to buy a new subscription. That’s the human behavior part of this.

But the main one is technical. The amount of bandwidth you need and the latency it takes to get the signal from the data centers back to the smart car to decide what it needs to do next while its driving makes it almost impossible. So, I think we’re going to end up with no cellular coverage in the cars. It’ll be there for it’ll be there for GPS and to make a cell phone calls but the car will not rely on that.

5g and V2X autonomous cars on the highway

We also hear about V2X communication for autonomous cars. But are there other ones you can tell us about?

The big debate is do we use real light, infrared or radio waves to talk car to car? And they’re trying all three of those.

Light has a problem with the fog. Spectrum has a problem anytime you’re around other users at that same spectrum when you get interference. Because we don’t have any spectrum bands that are pure everywhere, there’s going to always be some places where there’s going to be too much interference. And then, infrared is probably the best, it’s the same thing that’s on your remote control at your house, but that doesn’t go very far. And so, each of the three of them have an issue for cars talking to each other.

But cars talking to each other seems to be an important part of the solution because, on the DC Beltway when 30 cars ahead of you hit their brakes when they’re all communicating with each other, your car gets ready to slow down, which is way better than what you do. You don’t see it until it is only five cars, but the smart cars will say, you know, somebody hit their brakes hard, and it’ll get right back to the chain. That’s an awesome improvement safety wise.

But we need to find which one of those three to do it, and we might use multiple ones. The problem with spectrum is that the FCC already cut that spectrum in half. And so, and most of the car maker manufacturers have already abandoned that research, a few of them are still trying it, but they didn’t scream all that hard when they lost the spectrum. Now they’re off trying, light and infrared and all three of those, including spectrum, doesn’t always work.

Spectrum is probably the trickiest of the three. Radio waves have their issues, particularly when they bounce off metal cars. So, the whole issue of cars communicating with each other in real time is one of the big problems they’re having. You know, they there are people in the car industry that are saying they may not solve that till we have true AI. I think we might have just reached the limit of what computers can do.

AI microchip to power autonomous cars

So, can we expect for network providers and telecoms to even play a role here?

They still have a role because what my smart car today does, when I pull into my house is connect to my Wi-Fi and I have no idea what it’s telling Subaru, I guess it tells everything I did on that trip. And so, the cars still want to communicate with the cloud, they just won’t do it in real time to make driving decisions.

You can download a cartoon for your kid to watch in the backseat while you’re driving. And so what will happen with 5G, which this function’s not enabled yet but it’s coming, is it’ll grab a giant channel and go download the movie in a minute when you have a good cell signal. But whether the cell companies can talk people into buying a subscription to do that is the big question.

The only reason they keep putting that statement is because they really want that to be a revenue stream and right now you don’t have that revenue stream. None of us are paying a cell company for our car and I don’t know what it’s going to take to get them to do that. It could come with the mobile Netflix subscription for five more dollars a month and a lot of people would buy that… but that’s not what they originally planned for 5G.

WATCH THE FULL VIDEO REPORT

Diana Blass and Jo Kassis Photo

A story by Diana Blass and Jo Kassis.

Diana is Kurrant Insights’ Editor-in-Chief and video director. She is passionate about technologies that have a real impact on people.

Jo is an IoT-focused content strategist. He creates in-depth reports to keep our readers up to date on the newest industry developments.

RELATED POSTS

ACCESS HUNDREDS OF DOCUMENTARIES ABOUT SMART CITIES AND IOT!

We deliver the information you need in a short and engaging format. Our weekly in-depth documentaries cover real Smart Cities projects in 10 minutes or less.