TruckingTruth Article: Self Driving Vehicles Are Coming Soon You Say? I Say Please Stop Clowning Us

Brett, are you a wizard?

The long, winding road for driverless cars

Brett, are you a wizard?

The long, winding road for driverless cars

Hey, look at that.

Yeah, there's little doubt this hype is purely to drive funding for all sorts of people.

Another interesting point is that a new highway infrastructure would be required to make driverless cars a realistic possibility, and yet why isn't anyone talking about doing that? Because it's easier for us to believe that supercomputers can somehow learn to run the world properly than it is to believe that someday politicians will figure out how to. Driverless cars seems like a more realistic fantasy than reasonable politics. At least there's hope for driverless cars, no matter how remote it may be.

Politicians 'paving the way' for a better road infrastructure? Now that's just stupid talk right there.

Good article Brett.

Funny, I just had a close call with a four wheeler yesterday that made me wonder how an autonomous truck would have reacted to. I just picked up a load at a paper mill that was 67K worth of huge paper rolls stuffed into a tridem axle trailer and so was grossing over 100K. Yeah, because Canada, f**k eh? Anyways, I was on a rural two lane passing through some small town. Speed limit is 56, and I usually cruise at 58. But when rolling through these towns I always back down a bit, to low 50s just to have an extra safety margin, especially loaded to the limit like I was...

As I crested a hill and started my descent, off-throttle, jakes doing their job, a four-wheeler backs out of a driveway on the right and straight into my lane and is stopped pointing the wrong way (he wanted to go in opposite direction). I did not see him prior because he was hidden by motorhome parked in the same driveway. I slam on the brakes and realize there's no way I'm stopping in time and start looking for a way out, I had a few seconds to make the decision. To the right was the aforementionned motorhome and a gathering of a handfull of people, probably friends and/or relatives. To the left, there was a minivan heading in the opposite direction. Further to the left, a ditch, more driveways with cars and a few people. Lose, lose, lose. So I just maintained by heading straight, hard on the brakes and braced for impact praying the four wheeler be gone when I got there...

Fortunately, the idiot in the four-wheeler managed to get his ****box in gear and move into the opposite lane literally half a second before I got there, still doing at least 35... The minivan driver managed to slow down enough to avoid impact as well. As I passed the motorhome I saw what was probably a dozen people standing there, their eyes and mouths wide open. As the minivan rolled by and disappeared in opposite direction I caught a glimpse of the family and kids inside. As I looked back in my left mirror, and those people on the opposite side of the street? Were a father mowing the lawn and kids playing in the driveway...

I was shaking so badly it took me at least 20-30 seconds to get the truck into right gear and start accelerating again. Plus another couple of hours to get the puke feeling out of my stomach...

Now, what would the AI driven truck have done? How do you program for such a situation and who decides on the ethics code? Imagine the tragedy and the news headlines If myself or the AI would have chosen the other options:

1. Go right, plow through the motorhome and kill 10-12 people.

2. Go in the opposite lane, wipe out the four wheeler who is now there AND the family in the minivan.

3. Go further left, into the ditch, certainly roll the 40 ton trailer right on top of the playing kids and the dad mowing the lawn.

As a former Programmer/IT guy I understand the challenges with autonomous driving technology. It would take technology/processing/AI power far more advanced than what we have today. But computer technology advances at an incredible pace. If you would have told me 15 years ago about what Siri in your smartphone could do, I would have laughed at the notion of a smartphone, nevermind Siri...

The problem is that where we do not know how far we need to go to get there. Forget specialized road infrastructure. Aint gonna happen fast enough. I think the key is how quickly we can develop advanced AI. It could take 20-50 years. Or it could happen in the next 5.

Here's an interesting read on AI if you havent seen it yet. It's long, in two parts but I think worthy of some of your weekend downtime:

The AI Revolution

Self Driving Vehicles Are Coming Soon You Say? I Say Please Stop Clowning Us

Brett - I am a frequent reader and Podcast subscriber, however, I rarely log in to comment. That being said - GREAT ARTICLE here. I will be sharing this on our Facebook and Twitter pages. It's becoming the trucker version of "So...how about that weather" small-talk. When it comes up after meeting a trucker or someone who knows I'm in trucking, I always try to straddle the fence as to not get deep into any conversation about it.

We already have a problem getting the youth into our industry. Now they think its a dead end career that can't even consider? (Even more so. Beat ya to it. ha ha ha, haters) Just think of the technology changes your average 20-40 has lived through! Why wouldn't they believe that all commercial vehicles will be self-driven in the next few years?

Again - thanks for the article, keep them coming.

- Jrod

Thanks Jrod!

There's a huge difference in the world of technology between what's possible and what's practicable. It has already been proven that super geniuses can make cars and trucks drive themselves, and sometimes under pretty intense conditions. It's very impressive!

But that's when the million and one questions begin, right?

What is the cost of scaling it?

What is the risk versus reward for this implementation?

How many levels of red tape will we have to jump through to get this legalized?

What will the American public think the first time a bug or virus or hacker causes a major catastrophe? Because it will happen.

I think if we were serious about building new forms of transportation we would be building a new infrastructure to support them, not just sitting around hoping super geniuses can figure out how to retrofit current vehicles to drive themselves within the existing infrastructure.

If you're going to let computers drive cars, they don't have to be limited to 60 or 70 mph. They would be able to run near the limits of traction, inches apart from each other. You wouldn't need red lights or gigantic highways with 8 lanes in each direction. You could build single lane highways for autonomous vehicles where they run 120+ mph inches apart. The efficiency would be mesmerizing.

And of course no one would have to actually own a vehicle. You would just press a button on your iPhone app and the closest available one would be there in minutes, if not seconds.

And yet no one talks of any of this. There's no serious push to put autonomous vehicles on the highway, any more than there's a series push to put electric vehicles on the highway. It's all just lip service and fund raising.

I'm a technology buff myself. I love it. I love rocketships and race cars and computers. But I also know enough about the worlds of big business and politics to know there would need to be a revolution in the laws and business strategies, and a major shift in the public perception of technology.

What's funny is that most people think technology is cool and they would welcome more of it. But what happened when something totally revolutionary came around like Google Glass? (I have a pair, by the way) People freaked! We saw a glimpse of Star Trek type technology right in front of us and it creeped people out.

I've heard plenty of people say they don't like using cruise control because it's creepy. It makes them feel like they don't have control of the vehicle. So if Google Glass and cruise control freak people out, how many families are going to load their tiny little children into self-driving cars and trust that things will be fine?

There's a million and one questions that would have to be answered.

Ok, I realized that prob just sounded like a Trucking Truth paid advertisement, but I promise it wasn't! lol Brett never pays me no matter how many nice things I say and social media platforms I share to.

Going contrarian. We already have Embark taking refrigerators from Texas to California, using humans in Ryder trucks for "the last mile". Sticking to the relatively uncomplex I-10 is a great proof of concept. Since the first few months have been successful I think more can only be expected.

My guess as to when autonomous gets truly disruptive of the status quo? Five years, tops.

In an industry where tiny efficiency gains are avidly pursued the cost per mile of autonomous vehicles will be irresistible. Watch JB Hunt, as they ordered some Tesla electrics (once Musk gets production going, always a challenge for him) and Tesla electric is just a software patch from autonomous.

As long as these autonomous vehicles have a chance of having a malfunction or of being hacked and used as weapons targeting schools, hospitals, government buildings, etc....there will have to be humans able to disengage the "autopilot" and take control of the vehicle.

Could you imagine the chaos if a hostile entity "commandeered" 20-30 tanker trucks and directed them at soft targets???

Then there is the point made previously that people just don't trust technology. Now this may eventually change, but it will take quite some time. I spent over a decade in the aviation industry. The technology has existed for over 30 years that could take off, fly enroute, and land an aircraft without human intervention, but you still have two pilots sitting up front. Why? Because machines break and nobody wants it to be them in the back of the plane when the computer goes dark.....

I don't know about autonomous, but I do believe Tesla is releasing an all electric semi with a 500 mile range in a couple years. They say in a thirty minute charge time you can have 450 mile range.

We already have Embark taking refrigerators from Texas to California, using humans in Ryder trucks for "the last mile"

Ok, now let's tell it like it really is about the Embark trucks:

Currently, the highway portion of the trip, where the truck drives itself, is overseen by a team of two drivers who switch off regularly so they can stay alert while monitoring the system, and an engineer; the expectation is that eventually those drivers won’t be needed at all.

So you're right, there's not a driver in the truck, there are two drivers and an engineer monitoring the truck. I'm always amazed to find there is always a driver (or two, and maybe an engineer) behind the wheel of these "autonomous" vehicles.

So let's get to the real reason there is all this hype around autonomous trucks - investment money:

They founded Embark, landed $17 million in venture backing, and set out to sign up as many partners as possible before giants such as Tesla or Volvo undercut their market. But crucially, Embark doesn’t want to sell any trucks. Instead, they hope to shed their fleet and become a software company. So when a multinational such as Electrolux buys another truck to ship its home appliances, they might procure it from Peterbilt–but they’ll buy a software kit from Embark to automate that truck.

Tech startups are drumming up attention so they can raise tons and tons of money for their company. Then they create some basic software, hype the heck out of it to make it look like the next big thing, then either sell the company outright or try to license the software for big money to other companies.

Another example of a stupid idea - "platooning" trucks. The theory is you take a series of autonomous trucks and run them nose to tail to make them more fuel efficient. The lead truck would signal to the trailing trucks when to speed up or slow down or whatever. As you guys know, wireless technology is very sketchy. You can't rely on it. It's 2017 and half the time it takes two or three tries to connect my bluetooth earpiece to my phone. So do you really think we can rely on software and wireless connections to safely let trucks communicate? Can we let them drive nose to tail for wind resistance? Of course not! It's a ludicrous idea.

Instead, why not do something that makes way more sense, is way more reliable, and accomplishes the same thing - truck trains like they have in Australia: