When you hear the words driverless vehicle, you might picture one of those proto types in YouTube videos. You know, a small hatchback with a big gizmo on the roof and sensors sticking out all over the place and nobody in the driver's seat. Or maybe you think of Tesla's getting closer but not quite there yet autopilot feature.

When you hear the words driverless vehicle, you might picture one of those proto types in YouTube videos. You know, a small hatchback with a big gizmo on the roof and sensors sticking out all over the place and nobody in the driver's seat. Or maybe you think of Tesla's getting closer but not quite there yet autopilot feature.

According to the Tesla website, autopilot, enhanced autopilot, and full self driving features do not make their vehicles autonomous. Drivers must be fully attentive with their hands on the wheel.

According to the Tesla website, autopilot, enhanced autopilot, and full self driving features do not make their vehicles autonomous. Drivers must be fully attentive with their hands on the wheel.

The promise of truly autonomous vehicles has attracted a lot of attention over the years and a lot of investor money.

The promise of truly autonomous vehicles has attracted a lot of attention over the years and a lot of investor money.

Meanwhile, Fiat has announced that it has joined a bmwled consortium to develop self driving car technology.

Meanwhile, Fiat has announced that it has joined a bmwled consortium to develop self driving car technology.

Madrapan SoftBank has announced they will invest two and a quarter billion dollars in General Motors.

Madrapan SoftBank has announced they will invest two and a quarter billion dollars in General Motors.

Self driving car unit called the Crews, and self driving cars are starting to show up on the roads in states like California and Arizona.

Self driving car unit called the Crews, and self driving cars are starting to show up on the roads in states like California and Arizona.

Wayima, formerly known as Google's self driving car project, will soon start offering rides to the public in self driving vans.

Wayima, formerly known as Google's self driving car project, will soon start offering rides to the public in self driving vans.

There's also still plenty of skepticism though, especially whenever there's an accident involving one of these test vehicles.

There's also still plenty of skepticism though, especially whenever there's an accident involving one of these test vehicles.

Well, this morning, a woman is in the hospital after sufferank critical injuries after she was hit by a driverless car in downtown San Francisco last night.

Well, this morning, a woman is in the hospital after sufferank critical injuries after she was hit by a driverless car in downtown San Francisco last night.

Of course, but with so much attention on cars, you might not know that one of the fastest growing areas in the push toward driverless vehicles is actually trucks, and not pickups, but box trucks and the massive eighteen wheelers on the highway that deliver goods all over the US. Several US companies, including Kodiak, Gaddick and Aurora, are now testing autonomous trucks on the open road in Texas. Bloomberg Transportation reporter Thomas Black has taking a ride in these trucks, and he's here to talk about what these companies are trying to do and the high bar they have to meet to succeed.

Of course, but with so much attention on cars, you might not know that one of the fastest growing areas in the push toward driverless vehicles is actually trucks, and not pickups, but box trucks and the massive eighteen wheelers on the highway that deliver goods all over the US. Several US companies, including Kodiak, Gaddick and Aurora, are now testing autonomous trucks on the open road in Texas. Bloomberg Transportation reporter Thomas Black has taking a ride in these trucks, and he's here to talk about what these companies are trying to do and the high bar they have to meet to succeed.

All these companies know that they have to be almost flawless, if not perfect, on the technology as far as safety goes.

All these companies know that they have to be almost flawless, if not perfect, on the technology as far as safety goes.

And later we hear from Chris Ermson, the CEO of Aurora, about why he believes his eighteen wheelers will be on the road without a driver as early as next year. I'm Westksova today on the Big Take. Are you ready to share the road with driverless trucks? Tom? I think we're all aware that driverless cars have been in production in various ways for a long time. But what I did not know is that there's not one, but three driverless truck companies right now.

And later we hear from Chris Ermson, the CEO of Aurora, about why he believes his eighteen wheelers will be on the road without a driver as early as next year. I'm Westksova today on the Big Take. Are you ready to share the road with driverless trucks? Tom? I think we're all aware that driverless cars have been in production in various ways for a long time. But what I did not know is that there's not one, but three driverless truck companies right now.

Access there are, and there's going to be more.

Access there are, and there's going to be more.

It's a technology that probably has been proven out and now it's proving its safety case, and then they're going to roll out that safety case to people, and then they're going to take drivers out and we're going to see these on the roads if their pathway turns out the way they think it's going to turn out.

It's a technology that probably has been proven out and now it's proving its safety case, and then they're going to roll out that safety case to people, and then they're going to take drivers out and we're going to see these on the roads if their pathway turns out the way they think it's going to turn out.

Tell us about the companies that are making these trucks and putting them on the road and exactly how they were what they're doing.

Tell us about the companies that are making these trucks and putting them on the road and exactly how they were what they're doing.

So you have to think about these companies as software companies, because they're not going to make the trucks. The trucks are going to be made by the manufacturers that have made trucks for decades and more. All these other truck makers are going to provide the vehicles. And these guys really are going to add obviously the software, and these are big on board computers.

So you have to think about these companies as software companies, because they're not going to make the trucks. The trucks are going to be made by the manufacturers that have made trucks for decades and more. All these other truck makers are going to provide the vehicles. And these guys really are going to add obviously the software, and these are big on board computers.

That's very important because you want.

That's very important because you want.

To be able to have the vehicles self sustain it doesn't need the Internet to be able to operate. And then the mechanical side is really about redundancy. You want to make sure if something failed, like the steering wheel, that the thing doesn't careen off and hurt some people.

To be able to have the vehicles self sustain it doesn't need the Internet to be able to operate. And then the mechanical side is really about redundancy. You want to make sure if something failed, like the steering wheel, that the thing doesn't careen off and hurt some people.

So they're going to have redundant.

So they're going to have redundant.

Actuators and motors on the three things that we use to drive a vehicle, the steering wheel, the brakes, and the gas.

Actuators and motors on the three things that we use to drive a vehicle, the steering wheel, the brakes, and the gas.

So tom of these are three different companies that are doing it, and I imagine they each have a slightly different approach to figuring out how you put a truck on the highway without anyone driving it.

So tom of these are three different companies that are doing it, and I imagine they each have a slightly different approach to figuring out how you put a truck on the highway without anyone driving it.

The approach behind how this is going to work with the software and the mechanics is somewhat similar. It's about having three basic types of sensors. You have cameras of course, and then light r and then you have radar. So you have the sensor package, and then the sensor package connects to the computer and the computer has to analyze all this data coming in and make the decisions and then those decisions are translated into the actuators, which are little things that can either turn or twist or whatever, and then it'll be attached to again those three of driving, the steering wheel, the gas, and the brake. That all is pretty much the same. Really, where they might differ is what part of the market.

The approach behind how this is going to work with the software and the mechanics is somewhat similar. It's about having three basic types of sensors. You have cameras of course, and then light r and then you have radar. So you have the sensor package, and then the sensor package connects to the computer and the computer has to analyze all this data coming in and make the decisions and then those decisions are translated into the actuators, which are little things that can either turn or twist or whatever, and then it'll be attached to again those three of driving, the steering wheel, the gas, and the brake. That all is pretty much the same. Really, where they might differ is what part of the market.

They want to go after.

They want to go after.

Got Tea, for example, wants to go after what they would call the middle market, connecting the warehouse to the store. And then you have Kodiak and Aurora, which all they want to do is take it from warehouse to warehouse, and those warehouses would be usually outside of the city, right along a major highway, so they don't have to go into city traffic.

Got Tea, for example, wants to go after what they would call the middle market, connecting the warehouse to the store. And then you have Kodiak and Aurora, which all they want to do is take it from warehouse to warehouse, and those warehouses would be usually outside of the city, right along a major highway, so they don't have to go into city traffic.

Tom, give us a little tutorial here. I think we're all familiar with cameras would be but what's the difference between the two other technologies radar and light ar.

Tom, give us a little tutorial here. I think we're all familiar with cameras would be but what's the difference between the two other technologies radar and light ar.

So radar they bounce off radio waves and if it hits something, it will echo back and that way they can see, Okay, there's something in that radio wave spectrum out there. Light Ar is essentially the same type of concept, except it's with light.

So radar they bounce off radio waves and if it hits something, it will echo back and that way they can see, Okay, there's something in that radio wave spectrum out there. Light Ar is essentially the same type of concept, except it's with light.

Some of them work better when it's raining, others need more light. So they complement each other.

Some of them work better when it's raining, others need more light. So they complement each other.

And then of course you have the cameras, which gives you a visual. They talk about being able to have sensors out to about one thousand meters, so you have these three main modes constantly sweeping and picking up information about every tenth of a second. They're going through a cycle and feeding the computer information, so they have a pretty good picture of what's going on around them.

And then of course you have the cameras, which gives you a visual. They talk about being able to have sensors out to about one thousand meters, so you have these three main modes constantly sweeping and picking up information about every tenth of a second. They're going through a cycle and feeding the computer information, so they have a pretty good picture of what's going on around them.

If you're riding down the road and you see a driverlest truck, does it look the same as a regular eighteen wheel erd? Does it have all kinds of stuff stuck on the outside.

If you're riding down the road and you see a driverlest truck, does it look the same as a regular eighteen wheel erd? Does it have all kinds of stuff stuck on the outside.

Aurora has a sensor bar that goes on top of the cab. Then it had these almost like cone shape things, what would it be like a paint can size maybe a little smaller sticking out on the top. And then Kodiak has taken a different tact. They're putting their center package in.

Aurora has a sensor bar that goes on top of the cab. Then it had these almost like cone shape things, what would it be like a paint can size maybe a little smaller sticking out on the top. And then Kodiak has taken a different tact. They're putting their center package in.

The area where the rear view mirror is.

The area where the rear view mirror is.

So it looks like a small toolbox you might have in your garage that's tipped up on its end, so it's bigger than the normal rear mirror, and it does have the mirror in there, but along with that has a bunch of sensors and right now the Kodiak. All of them have their names on them and some have autonomous truck written across them, so if you know what you're looking for, you can see them.

So it looks like a small toolbox you might have in your garage that's tipped up on its end, so it's bigger than the normal rear mirror, and it does have the mirror in there, but along with that has a bunch of sensors and right now the Kodiak. All of them have their names on them and some have autonomous truck written across them, so if you know what you're looking for, you can see them.

Time when you were reporting this story, you actually got to get inside the cab of these trucks and ride inside them. What was that like?

Time when you were reporting this story, you actually got to get inside the cab of these trucks and ride inside them. What was that like?

It was a lot of fun.

It was a lot of fun.

I'd love to do these type of things where you get to go see something.

I'd love to do these type of things where you get to go see something.

You have a safety driver.

You have a safety driver.

It's usually a driver that's had a ton of experience, and they picked the most safe drivers possible to recruit. And then on the passenger seat of the truck there will be another person who's taking all kinds of diagnostics of how the system's working and if it had to be engaged, and all these kinds of things that they do to build their safety case. Essentially, as you take off, the truck starts and the safety driver has his hands covered over the wheel, but you can tell he's about an inch or so away from it.

It's usually a driver that's had a ton of experience, and they picked the most safe drivers possible to recruit. And then on the passenger seat of the truck there will be another person who's taking all kinds of diagnostics of how the system's working and if it had to be engaged, and all these kinds of things that they do to build their safety case. Essentially, as you take off, the truck starts and the safety driver has his hands covered over the wheel, but you can tell he's about an inch or so away from it.

I was watching hands quite a lot.

I was watching hands quite a lot.

I wanted to see, Okay, when we do a turn or we do something, is he going to touch this or grab this? But during the trip I did not see that happen. I was impressed with that. And in our particular case, we pulled out of the terminal, drove up.

I wanted to see, Okay, when we do a turn or we do something, is he going to touch this or grab this? But during the trip I did not see that happen. I was impressed with that. And in our particular case, we pulled out of the terminal, drove up.

A service road.

A service road.

We're going up this service road and we come to a stop sign.

We're going up this service road and we come to a stop sign.

The truck turns left.

The truck turns left.

And one of the impressive things is that we were going over and overpass over the highway and it navigated a stop light that didn't have a protected left turn where you actually get the green arrow to turn and you don't have to worry about oncoming traffic because they have to stop. If you don't have that protected left you have to wait for all those cars to clear and then you have to go. And sometimes you might have to make a pretty quick decision if there's traffic and there's only a small space.

And one of the impressive things is that we were going over and overpass over the highway and it navigated a stop light that didn't have a protected left turn where you actually get the green arrow to turn and you don't have to worry about oncoming traffic because they have to stop. If you don't have that protected left you have to wait for all those cars to clear and then you have to go. And sometimes you might have to make a pretty quick decision if there's traffic and there's only a small space.

So this truck was able to navigate that.

So this truck was able to navigate that.

It had the green light, and it waited and a couple of cars came by, and it pulled out, got on the highway.

It had the green light, and it waited and a couple of cars came by, and it pulled out, got on the highway.

We drove on the highway a bit.

We drove on the highway a bit.

These things are going to drive a couple of miles per hour below the speed limit. They're going to stay in the right lanes, they're going to prod along, and they're probably not going to interact as much with vehicles as some other trucks do. When they're trying to gain a little bit on the traffic and so forth, they might be changing lanes more often. The machine doesn't care how long it takes to get to their destination. They're just going to pulog along. There was an instance where a vehicle was coming on the on ramp and the truck had a decision.

These things are going to drive a couple of miles per hour below the speed limit. They're going to stay in the right lanes, they're going to prod along, and they're probably not going to interact as much with vehicles as some other trucks do. When they're trying to gain a little bit on the traffic and so forth, they might be changing lanes more often. The machine doesn't care how long it takes to get to their destination. They're just going to pulog along. There was an instance where a vehicle was coming on the on ramp and the truck had a decision.

It could pull over and give it space, or it could slow down.

It could pull over and give it space, or it could slow down.

The truck decided to slow down because there were cars that were coming up on its left faster. One of the coolest experiences was at the four way stop to turn back on the service road and get back to the warehouse.

The truck decided to slow down because there were cars that were coming up on its left faster. One of the coolest experiences was at the four way stop to turn back on the service road and get back to the warehouse.

There was quite a few cars.

There was quite a few cars.

In fact, every one of the stop signs had a line of cars, so people had to navigate this correctly.

In fact, every one of the stop signs had a line of cars, so people had to navigate this correctly.

And take their turn.

And take their turn.

So the truck comes up, it stops, and they were kind of explaining to me what it's doing, what's checking everybody where its turn is. And the three cars went, and then the truck lunged forward, and after it lunched forward, it stopped, and then it did the turn, and they explained to me that the truck was gaining its space. In other words, it was saying, okay, it's my turn with that lunge, and then when no other car moved, the truck then proceeded. So these are the kind of things that they have to program into this to mimic how humans drive so that they can interact with vehicles that have humans behind the wheel.

So the truck comes up, it stops, and they were kind of explaining to me what it's doing, what's checking everybody where its turn is. And the three cars went, and then the truck lunged forward, and after it lunched forward, it stopped, and then it did the turn, and they explained to me that the truck was gaining its space. In other words, it was saying, okay, it's my turn with that lunge, and then when no other car moved, the truck then proceeded. So these are the kind of things that they have to program into this to mimic how humans drive so that they can interact with vehicles that have humans behind the wheel.

Did it feel kind of eerie to be in a truck that was driving itself?

Did it feel kind of eerie to be in a truck that was driving itself?

I would say it was actually mundane in a way because you do have the safety driver there. You know, he can grab the wheel at any time. It doesn't spook you in that sense. It is cool to watch when the truck makes a turn, and then after it completes the turn, the wheel has to wind back as the wheels straighten up, so that's when you can really see that, oh hey, he's not grabbing this wheel at all. I'll take a lot of video that because to me, that's where the action is right watching this truck steer itself.

I would say it was actually mundane in a way because you do have the safety driver there. You know, he can grab the wheel at any time. It doesn't spook you in that sense. It is cool to watch when the truck makes a turn, and then after it completes the turn, the wheel has to wind back as the wheels straighten up, so that's when you can really see that, oh hey, he's not grabbing this wheel at all. I'll take a lot of video that because to me, that's where the action is right watching this truck steer itself.

After the break. What these trucks have to do to persuade the public that they're safe. Tom, you've mentioned the names of these three companies. Tell us about them. These are American companies.

After the break. What these trucks have to do to persuade the public that they're safe. Tom, you've mentioned the names of these three companies. Tell us about them. These are American companies.

They're American companies.

They're American companies.

In fact, Gatique and Kodiak are based in Silicon Valley and Aurora is officially based in Pittsburgh, but their founder, Chris Ermson, is based in Silicon Valley, so they're tech companies. The interesting thing is most of these guys know each other. Google was an early mover in autonomous vehicle technology, and Chris Ermson and Don Burnett, who's the CEO of Kodiak, they both work there.

In fact, Gatique and Kodiak are based in Silicon Valley and Aurora is officially based in Pittsburgh, but their founder, Chris Ermson, is based in Silicon Valley, so they're tech companies. The interesting thing is most of these guys know each other. Google was an early mover in autonomous vehicle technology, and Chris Ermson and Don Burnett, who's the CEO of Kodiak, they both work there.

It's about software that's.

It's about software that's.

What makes the barrier to entry to this is somewhat low. In that sense, you don't have to set up a truck manufacturing facility to be able to get into the business.

What makes the barrier to entry to this is somewhat low. In that sense, you don't have to set up a truck manufacturing facility to be able to get into the business.

It really is about the software.

It really is about the software.

All three of them are testing these trucks in Texas.

All three of them are testing these trucks in Texas.

Why Texas, Texas was an early mover and setting up the regulations for this to happen. They passed legislation that set the groundwork to test and eventually operate autonomous.

Why Texas, Texas was an early mover and setting up the regulations for this to happen. They passed legislation that set the groundwork to test and eventually operate autonomous.

Vehicles on the roads.

Vehicles on the roads.

That gives a lot of certainty, So I think that's one of the things that was attractive. The other is that Texas is a very big freight market. In fact, it's the second biggest freight market behind California and it has a vast area. It has a big port in Houston, has several ports. Also, it has Dallas, which is a transportation hub because it's basically in the middle of the country and it has lots of wide open flat spaces.

That gives a lot of certainty, So I think that's one of the things that was attractive. The other is that Texas is a very big freight market. In fact, it's the second biggest freight market behind California and it has a vast area. It has a big port in Houston, has several ports. Also, it has Dallas, which is a transportation hub because it's basically in the middle of the country and it has lots of wide open flat spaces.

In time, you write that all of these ports and transport hubbs make a really good testing ground because these trucks can practice in real life conditions.

In time, you write that all of these ports and transport hubbs make a really good testing ground because these trucks can practice in real life conditions.

That's right.

That's right.

They start out with lots of simulation and doing things in a test area. This is how they got their start. They would be in confined area where they could test the vehicles and make sure the software was doing what it was supposed to and reading all the sensors correctly. Once they got to a certain skill level, then they had to take it to the road.

They start out with lots of simulation and doing things in a test area. This is how they got their start. They would be in confined area where they could test the vehicles and make sure the software was doing what it was supposed to and reading all the sensors correctly. Once they got to a certain skill level, then they had to take it to the road.

They've been doing.

They've been doing.

This for several years now and they're getting close to saying, hey, we're ready. In the case of Aurora, they've came out and made a pretty bold statement that said that our software is ready. It can handle any of the events that come up on the road. The truck can handle all the circumstances that it's going to phase. They've said that case is closed. Now we're going to present our safety case. Kodiak hasn't made that statement. They say that they're there, but they're doing things a little bit differently. That's kind of the progression I think we're going to see. They're testing them on the road right now, they still have the safety drivers. At some point they're going to have to take the safety drivers out. I think the people are going to want to see more information about how many times.

This for several years now and they're getting close to saying, hey, we're ready. In the case of Aurora, they've came out and made a pretty bold statement that said that our software is ready. It can handle any of the events that come up on the road. The truck can handle all the circumstances that it's going to phase. They've said that case is closed. Now we're going to present our safety case. Kodiak hasn't made that statement. They say that they're there, but they're doing things a little bit differently. That's kind of the progression I think we're going to see. They're testing them on the road right now, they still have the safety drivers. At some point they're going to have to take the safety drivers out. I think the people are going to want to see more information about how many times.

These safety drivers actually grab the wheel.

These safety drivers actually grab the wheel.

They don't publish those, they say they give that information to authorities, but I think at some point they're going to have to bring a lot of information to the public to say this is why we know these things are safe.

They don't publish those, they say they give that information to authorities, but I think at some point they're going to have to bring a lot of information to the public to say this is why we know these things are safe.

This question of safety you report is really the big one that trying to persuade both regulators and normal people driving on the road that these things are safe.

This question of safety you report is really the big one that trying to persuade both regulators and normal people driving on the road that these things are safe.

I think I can make as bold a statement to say that if they don't make the roadways safer, the industry will fail.

I think I can make as bold a statement to say that if they don't make the roadways safer, the industry will fail.

And why is that?

And why is that?

It's almost a taboo thing. We do not want machines that can hurt people. I think most people would agree with me on that, But that's a pretty low bar because on our roadways, forty thousand people die every year. That includes everything passenger vehicles. That is an alarmingly high number. And people have absorbed that information. They've decided they can take the risk, and they get in their vehicles every day and they drive. And as far as large trucks, five thousand people die in accidents that involve large trucks, and most of those people are the passenger car drivers just because the truck is much bigger, and the vast majority of the time the truck driver is not at fault. It's usually a motorist that has done something silly or something stupid and causes accidents. These autonomous vehicles are going to have to be safer. I think that's clear, or it's just not going to be accepted.

It's almost a taboo thing. We do not want machines that can hurt people. I think most people would agree with me on that, But that's a pretty low bar because on our roadways, forty thousand people die every year. That includes everything passenger vehicles. That is an alarmingly high number. And people have absorbed that information. They've decided they can take the risk, and they get in their vehicles every day and they drive. And as far as large trucks, five thousand people die in accidents that involve large trucks, and most of those people are the passenger car drivers just because the truck is much bigger, and the vast majority of the time the truck driver is not at fault. It's usually a motorist that has done something silly or something stupid and causes accidents. These autonomous vehicles are going to have to be safer. I think that's clear, or it's just not going to be accepted.

By the public.

By the public.

Aurora, Kodiak, Gatique, all these companies know that they have to be almost flawless, if not perfect, on the technology. As far as safety goes. There are going to be accidents. It's inevitable that there will be accidents, but the truck, the system has to be able to show that it either wasn't our fault or it was some kind of mechanical failure that does happen, but it wasn't because of the autonomous driving technology. That is important for these companies maintain almost a spotless record where they can say, you know what, our autonomous technology has never caused an accident.

Aurora, Kodiak, Gatique, all these companies know that they have to be almost flawless, if not perfect, on the technology. As far as safety goes. There are going to be accidents. It's inevitable that there will be accidents, but the truck, the system has to be able to show that it either wasn't our fault or it was some kind of mechanical failure that does happen, but it wasn't because of the autonomous driving technology. That is important for these companies maintain almost a spotless record where they can say, you know what, our autonomous technology has never caused an accident.

So how has the safety record of these test trucks been. So far?

So how has the safety record of these test trucks been. So far?

So far, so good.

So far, so good.

They haven't caused it an accident, that's for sure. They have been in accidents. There is a reporting system, a federal reporting system where they report any incident that has happened, no matter how minor. The cool thing is that they attached part of the police report to it in a section so you can see what happened.

They haven't caused it an accident, that's for sure. They have been in accidents. There is a reporting system, a federal reporting system where they report any incident that has happened, no matter how minor. The cool thing is that they attached part of the police report to it in a section so you can see what happened.

There was one.

There was one.

Case where a vehicle crossed over two lanes and hit the truck in the back, and it turned out that the driver admitted that he fell asleep in his car, lost over two lanes, but he wasn't hurt. It crumpled up his hood, he had to be towed, but the truck was able to carry on. I didn't see any of the accident reports where you would say, oh, the truck was that fault. Here it was the human essentially that made the mistake.

Case where a vehicle crossed over two lanes and hit the truck in the back, and it turned out that the driver admitted that he fell asleep in his car, lost over two lanes, but he wasn't hurt. It crumpled up his hood, he had to be towed, but the truck was able to carry on. I didn't see any of the accident reports where you would say, oh, the truck was that fault. Here it was the human essentially that made the mistake.

Right now, these trucks are being tested in Texas, but can they leave Texas and go into other states?

Right now, these trucks are being tested in Texas, but can they leave Texas and go into other states?

They can right now, the regulations have really been left up to the states for autonomous trucking. The federal government is likely to weigh in on this, but for right now, it's really up to the states to allow this.

They can right now, the regulations have really been left up to the states for autonomous trucking. The federal government is likely to weigh in on this, but for right now, it's really up to the states to allow this.

And there's a wide swath along the sun Belt, if you will, of states.

And there's a wide swath along the sun Belt, if you will, of states.

That are allowing this, from Arizona, New Mexico, Texas on out and even up into Oklahoma.

That are allowing this, from Arizona, New Mexico, Texas on out and even up into Oklahoma.

So, yes, it is happening.

So, yes, it is happening.

Time you said that safety concerns have to be satisfied before they can get final approval. How long until that happens. How long do you think before we start seeing autonomous eighteen wheelers all over the place in Texas and other states.

Time you said that safety concerns have to be satisfied before they can get final approval. How long until that happens. How long do you think before we start seeing autonomous eighteen wheelers all over the place in Texas and other states.

It's very interesting because in a state like Texas, there's probably no other hurdles for Aurora, Kodiak and some of these others to go driver out. It really kind of is up to them to say, Okay, we've proved our safety case.

It's very interesting because in a state like Texas, there's probably no other hurdles for Aurora, Kodiak and some of these others to go driver out. It really kind of is up to them to say, Okay, we've proved our safety case.

They have more or less a green light to go.

They have more or less a green light to go.

I'm sure they're going to have to get some sign off from state authorities when they want to go driver out, because it's a big step. What's holding them back is they know that they have to be really cautious about this because we talked about that if they had an accident in which the autonomous truck caused an accident, it would be a terrible blow for the whole industry.

I'm sure they're going to have to get some sign off from state authorities when they want to go driver out, because it's a big step. What's holding them back is they know that they have to be really cautious about this because we talked about that if they had an accident in which the autonomous truck caused an accident, it would be a terrible blow for the whole industry.

So when you look down the road, sorry for the pun, five years, ten years, do you think that the highways will be filled with more driverless trucks than trucks with people behind the wheel.

So when you look down the road, sorry for the pun, five years, ten years, do you think that the highways will be filled with more driverless trucks than trucks with people behind the wheel.

I would probably say no, just because I think this will be a slow rollout, slow and deliberate.

I would probably say no, just because I think this will be a slow rollout, slow and deliberate.

I could be wrong.

I could be wrong.

It doesn't have any any kind of production limitations. Again, it's really about software. So if it does take off and people embrace it and it really shows that it's safe, I could see it ramping up quite quickly. But I tend to think that it's going to be a slow rollout.

It doesn't have any any kind of production limitations. Again, it's really about software. So if it does take off and people embrace it and it really shows that it's safe, I could see it ramping up quite quickly. But I tend to think that it's going to be a slow rollout.

Tom, thanks for the ride along.

Tom, thanks for the ride along.

Yeah, it's been fun.

Yeah, it's been fun.

When we come back, the CEO of a driverlest trucking company talks about what it takes to put one of these vehicles on the road. Now, let's hear from someone who's working to put these trucks down the road. Chris Ermson is CEO of Aurora, one of the companies. Tom was talking about. Chris, what made you want to get into this business.

When we come back, the CEO of a driverlest trucking company talks about what it takes to put one of these vehicles on the road. Now, let's hear from someone who's working to put these trucks down the road. Chris Ermson is CEO of Aurora, one of the companies. Tom was talking about. Chris, what made you want to get into this business.

Honestly, back in the day, I thought it was cool. You know, I was a graduate student of the time in Carnegie Mellen and we'd been working on some very slow moving robots.

Honestly, back in the day, I thought it was cool. You know, I was a graduate student of the time in Carnegie Mellen and we'd been working on some very slow moving robots.

We're out in the desert in.

We're out in the desert in.

The auto Comma and the robot drove about thirty You send me to a second which is how fast you move if you have a walker? And my PhD advisor said, Hey, this is this competition to go drive robots across the desert at fifty miles an hour. And I thought that sounds amazing. And then over the last twenty years of working on it, you know, I've had the chance to work with amazing people. We work on a really challenging technical problem and a problem that has profound impact on society, and so that's kind of kept me with it over a couple of decades.

The auto Comma and the robot drove about thirty You send me to a second which is how fast you move if you have a walker? And my PhD advisor said, Hey, this is this competition to go drive robots across the desert at fifty miles an hour. And I thought that sounds amazing. And then over the last twenty years of working on it, you know, I've had the chance to work with amazing people. We work on a really challenging technical problem and a problem that has profound impact on society, and so that's kind of kept me with it over a couple of decades.

So those challenging technical problems, let me just talk about some of those. What is the biggest hurdle that you've had to try to puzzle through and overcome.

So those challenging technical problems, let me just talk about some of those. What is the biggest hurdle that you've had to try to puzzle through and overcome.

The joy of this problem is there's no one right that. There's a collection of problems, you know, around how do you see far enough down the road how do you model the way other people are going to behave on the road given what you do and how that interaction works and the complexities of that, And then how do you convince yourself that you know it works well enough that you could trust it out in the world and that when something breaks that it's going to respond in a way that's safe. And so it's that collection of all of those things, and then you layer on top of the technology, the business challenges, and the engaging with the public and policy makers, and you know, it's a fascinating space.

The joy of this problem is there's no one right that. There's a collection of problems, you know, around how do you see far enough down the road how do you model the way other people are going to behave on the road given what you do and how that interaction works and the complexities of that, And then how do you convince yourself that you know it works well enough that you could trust it out in the world and that when something breaks that it's going to respond in a way that's safe. And so it's that collection of all of those things, and then you layer on top of the technology, the business challenges, and the engaging with the public and policy makers, and you know, it's a fascinating space.

Where are you right now? How many trucks are out on the road, Where are they going?

Where are you right now? How many trucks are out on the road, Where are they going?

So we're on the road today in Texas. We drive between Fort Worth and Opasso and between Dallas and Houston. Every day. We're hauling loads for customers. We work with partners like FedEx, Wernerschneider, Hirschbach, Uber Freight, and a couple of others we can't talk about yet, and we have on order about thirty trucks that are out there.

So we're on the road today in Texas. We drive between Fort Worth and Opasso and between Dallas and Houston. Every day. We're hauling loads for customers. We work with partners like FedEx, Wernerschneider, Hirschbach, Uber Freight, and a couple of others we can't talk about yet, and we have on order about thirty trucks that are out there.

Now. Are these trucks that are actually out there delivering goods completely driverless or is there someone in the cab with their hands kind of hovering over the steering wheel.

Now. Are these trucks that are actually out there delivering goods completely driverless or is there someone in the cab with their hands kind of hovering over the steering wheel.

So today we have a team in the vehicles monitoring the system, but the vast majority of the time it's driving itself. By the end of next year, we expect to be at the point where we have vehicles operating on the road without people on board, so truly without human operators there.

So today we have a team in the vehicles monitoring the system, but the vast majority of the time it's driving itself. By the end of next year, we expect to be at the point where we have vehicles operating on the road without people on board, so truly without human operators there.

And what has to happen between now and then for you to feel comfortable or anyone else who might want to sign off on this, to say, yeah, we're totally fine with these things just going off and doing their thing.

And what has to happen between now and then for you to feel comfortable or anyone else who might want to sign off on this, to say, yeah, we're totally fine with these things just going off and doing their thing.

We've laid out a sequence of milestones. There were, you know, a bunch of things, but the three big ones were to get to feature complete, to get to what we're called the roor driver ready, and then to launch commercially. And so feature complete we achieved at the beginning of this year, and that really meant that all the parts are there, they work, but we haven't yet convinced ourselves that they're all the way correct and that we validated that we're safe to go. By the end of this year, when we hit this goal of e Royer Driver Ready, our expectation is at that point the parts that we control, the software, the way the sensors work each other, what's on the computer. All of that that we have done, all of the testing and all of the analysis to convince ourselves that if we had a truck that was ready to go, we'd be happy to put it's on the road and be safe. And then finally commercial launch will happen next year. And that's where we've done the work with our great automotive and truck partners to do the final testing and integration to make sure our stuff talks to their stuff the way it's supposed to. Everything is copasetic.

We've laid out a sequence of milestones. There were, you know, a bunch of things, but the three big ones were to get to feature complete, to get to what we're called the roor driver ready, and then to launch commercially. And so feature complete we achieved at the beginning of this year, and that really meant that all the parts are there, they work, but we haven't yet convinced ourselves that they're all the way correct and that we validated that we're safe to go. By the end of this year, when we hit this goal of e Royer Driver Ready, our expectation is at that point the parts that we control, the software, the way the sensors work each other, what's on the computer. All of that that we have done, all of the testing and all of the analysis to convince ourselves that if we had a truck that was ready to go, we'd be happy to put it's on the road and be safe. And then finally commercial launch will happen next year. And that's where we've done the work with our great automotive and truck partners to do the final testing and integration to make sure our stuff talks to their stuff the way it's supposed to. Everything is copasetic.

Earlier, I was talking to my colleague Thomas Black, who is saying that one of the challenges is that people know that there will be accidents on the road people are driving cars, but that the safety record, if it's driven autonomously almost needs to be perfect or else people will not have confidence. Is that like a bar that you can meet.

Earlier, I was talking to my colleague Thomas Black, who is saying that one of the challenges is that people know that there will be accidents on the road people are driving cars, but that the safety record, if it's driven autonomously almost needs to be perfect or else people will not have confidence. Is that like a bar that you can meet.

It's easy to overlook the status quota today right that forty two thousand Americans diner roads every year and want to quarter million people by globally, and this is a technology that really can drive that towards zero. And that's what we're focused on, and that work we do to demonstrate to ourselves that the vehicle is proficient, are profound. There's a few different ways we've explored this. So one is that the Department of Transportation has a taxonomy for how vehicles get into collisions, and so we've taken that taxonomy. They said, Okay, let's expose the euror driver to each of these kind of scenarios, and then let's create variations of them, tens of thousands of kind of near miss events or near collision events, and let's make sure all of them the eroor driver is behaving the way we'd want to. That's events that you or I we'd see if any one or two of these events in our lifetimes, and we're pushing the system through tens of thousands of them. And then we've actually looked at the real world implication of this. So the road that we're driving between Dawnson and Houston is I forty five. We've looked at all of the fatal accidents that involved trucks that happened between twenty eighteen and twenty twenty two, and we pulled the incident reports on those, and then we created simulations where we had the our driver operating the vehicle, and it was twenty nine of these that the ero driver could have been operating, and across all twenty nine of them, if the row driver had been operating, the collision just would not have occurred. And if you think about the impact for those families of what this would mean, right the not losing those loved ones, it's really the message home for me.

It's easy to overlook the status quota today right that forty two thousand Americans diner roads every year and want to quarter million people by globally, and this is a technology that really can drive that towards zero. And that's what we're focused on, and that work we do to demonstrate to ourselves that the vehicle is proficient, are profound. There's a few different ways we've explored this. So one is that the Department of Transportation has a taxonomy for how vehicles get into collisions, and so we've taken that taxonomy. They said, Okay, let's expose the euror driver to each of these kind of scenarios, and then let's create variations of them, tens of thousands of kind of near miss events or near collision events, and let's make sure all of them the eroor driver is behaving the way we'd want to. That's events that you or I we'd see if any one or two of these events in our lifetimes, and we're pushing the system through tens of thousands of them. And then we've actually looked at the real world implication of this. So the road that we're driving between Dawnson and Houston is I forty five. We've looked at all of the fatal accidents that involved trucks that happened between twenty eighteen and twenty twenty two, and we pulled the incident reports on those, and then we created simulations where we had the our driver operating the vehicle, and it was twenty nine of these that the ero driver could have been operating, and across all twenty nine of them, if the row driver had been operating, the collision just would not have occurred. And if you think about the impact for those families of what this would mean, right the not losing those loved ones, it's really the message home for me.

I imagine you spend a certain amount of your time just trying to explain this stuff to people, to build a kind of comfort level with the idea of this happening. What's the biggest missingconception you've come across when you start talking to people about a driverless truck.

I imagine you spend a certain amount of your time just trying to explain this stuff to people, to build a kind of comfort level with the idea of this happening. What's the biggest missingconception you've come across when you start talking to people about a driverless truck.

People have rational concern about the new and the uncertain. It's a perfectly normal behavior and expected and warranted. We're driving seventy thousand pound trucks down the road. You should have questions. And what we find is that people very quickly accept. So I've been working this space for twenty years and we've given folks rights in automated vehicles for a better part of that time. You know, you'll have the folks who are really excited about this, and so let's not talk about that they get in this is wonderful and amazing wow. Right, But you have the skeptics and they're like, I don't know, this must be smoking mirrors. I don't know that I'd ever trust it. And they get in the back of the vehicle, and you know, for the first five minutes they're kind of tense and they're like, you know, what's happening here, right, And then then they start to relax, and you know, invariably you get some kind of common or question of it. Just you know, it just drives, that's it. And then somewhere between ten and fifteen minutes they're checking their phones and so exposing people to this I think is one of the ways that you can help them understand. And part of my you know, amateur psychology here is that we're all used to driving in vehicles that other people drive. It's just a very common experience, and very quickly we kind of pattern match to that and relax around it.

People have rational concern about the new and the uncertain. It's a perfectly normal behavior and expected and warranted. We're driving seventy thousand pound trucks down the road. You should have questions. And what we find is that people very quickly accept. So I've been working this space for twenty years and we've given folks rights in automated vehicles for a better part of that time. You know, you'll have the folks who are really excited about this, and so let's not talk about that they get in this is wonderful and amazing wow. Right, But you have the skeptics and they're like, I don't know, this must be smoking mirrors. I don't know that I'd ever trust it. And they get in the back of the vehicle, and you know, for the first five minutes they're kind of tense and they're like, you know, what's happening here, right, And then then they start to relax, and you know, invariably you get some kind of common or question of it. Just you know, it just drives, that's it. And then somewhere between ten and fifteen minutes they're checking their phones and so exposing people to this I think is one of the ways that you can help them understand. And part of my you know, amateur psychology here is that we're all used to driving in vehicles that other people drive. It's just a very common experience, and very quickly we kind of pattern match to that and relax around it.

Chris, you've been at this for twenty years. I'm going to ask you a crystal ball question. When you look ahead ten years, what percentage of long haul trucks across the US are autonomous.

Chris, you've been at this for twenty years. I'm going to ask you a crystal ball question. When you look ahead ten years, what percentage of long haul trucks across the US are autonomous.

It's still going to be a small percentage because the market is just so big, right, It's an eight hundred billion dollar market today, and so this technology is going to roll out and grow and it's going to have an impact, And initially it's going to be about filling the shortage of drivers that are out there today. So in the US we're short about eighty thousand drivers. We expect by the end of the decade to be short about twice that one hundred and sixty thousand, and so automated driving is one way to compliment the human drivers that are out there today doing a really important and noble job.

It's still going to be a small percentage because the market is just so big, right, It's an eight hundred billion dollar market today, and so this technology is going to roll out and grow and it's going to have an impact, And initially it's going to be about filling the shortage of drivers that are out there today. So in the US we're short about eighty thousand drivers. We expect by the end of the decade to be short about twice that one hundred and sixty thousand, and so automated driving is one way to compliment the human drivers that are out there today doing a really important and noble job.

Chris really enjoyed talking with you. Thanks for taking the time.

Chris really enjoyed talking with you. Thanks for taking the time.

My pleasure. Really appreciate it.

My pleasure. Really appreciate it.

Thanks for listening to us here at The Big Take. It's a daily podcast from Bloomberg and iHeartRadio. For more shows from iHeartRadio, visit the iHeartRadio app, Apple Podcasts, or wherever you listen, and we'd love to hear from you. Email us questions or comments to Big Take at Bloomberg dot net. The supervising producer of The Big Take is Vicky Bergolina. Our senior producer is Katherine Fink. Our producers are Michael Falero and Mobero Raphael mcili is our engineer. Our original music was composed by Leo Sidrin. I'm West Kasova. We'll be back tomorrow with another big take.

Thanks for listening to us here at The Big Take. It's a daily podcast from Bloomberg and iHeartRadio. For more shows from iHeartRadio, visit the iHeartRadio app, Apple Podcasts, or wherever you listen, and we'd love to hear from you. Email us questions or comments to Big Take at Bloomberg dot net. The supervising producer of The Big Take is Vicky Bergolina. Our senior producer is Katherine Fink. Our producers are Michael Falero and Mobero Raphael mcili is our engineer. Our original music was composed by Leo Sidrin. I'm West Kasova. We'll be back tomorrow with another big take.