SEASON ONE: EPISODE ONE

AUTONOMOUS: WHO'S BEHIND THE WHEEL OF TOMORROW?

In this episode, we set out to answer a question - what does the move toward autonomy mean for the truck driver? To get that answer, Transport Topics Managing Editor, Features, Seth Clevenger, approaches two people at the forefront of developing this technology who are responsible for driving this movement despite representing two very different paradigms: one, the OEM engineer; the other, the Silicon Valley upstart. With his engaging interview style, the journalist provokes each to comment on the topic and answer the core question directly. The result, for you, is a fast-paced, novel look at what's really happening with automation in trucking.

Get full transcript here.

FEATURED GUESTS

Fred Andersky heads the controls business and government and industry affairs at Bendix Commercial Vehicle Systems, a major supplier of braking systems and active-safety technology.

Alex Rodrigues is CEO of Embark, a San Francisco-based startup that is developing self-driving truck technology and is testing it on the road today.

Fred Andersky

Alex Rodrigues

EP. 01

Brought to you by:

Episode Transcript

From Transport Topics in Washington D.C.. This is roadsigns.

This podcast is brought to you by International Truck. Now here is your host Seth Clevenger. Welcome to roadsigns new podcast series from Transport Topics. And this our first episode we're going to explore the road to autonomous trucking and discuss who will be behind the wheel the trucks of tomorrow automated driving technology could hold the potential to improve road safety boost productivity and may even begin to address the trucking industry's persistent driver shortage and high driver turnover rates. But all this also raises a very important question. If autonomous trucks are the future of this industry what is the future of the truck driver. We can start to answer that question by looking at two different pathways to automation and trucking. The baseline is the gradual expansion of driver assist technology. Building on the foundation provided by today's active safety systems at the same time several technology companies are focused on jumping straight to fully autonomous trucks but in limited applications one example that we'll get to in a bit is deploying driverless long haul trucks on certain freeways but continuing to rely on drivers to handle the more complex driving tasks. At the beginning and end of the journey course there are also many other factors that will shape the development of automated driving technology including the still emerging regulatory framework and public acceptance to get some more perspective on all this.

Guests

We've lined up two excellent guests later in the program we'll hear from Fred Andersky who heads the Controls business and government and industry affairs at Bendix, a major supplier of active safety systems that could serve as the building blocks of autonomy Fred will share his insights on the different pathways toward automation and look ahead at the evolution of driver assist ecology. But first we're thrilled to welcome a special guest who is working hard to make autonomous trucking a reality in the near future. Alex Rodriguez is the CEO of embark a San Francisco based startup that is developing self driving truck technology and is testing it out on the road today. So thank you very much for joining us Alex. Thanks for having me sir. You bet. Embark has been actively testing and operating self driving trucks mostly in the southwestern U.S. on Interstate 10. Today these trucks have a safety driver behind the wheel as a backup ready to take over when needed. But the goal is to eventually reach the point where your trucks can operate on the freeway without a driver on board. So just how much work remains before you can confidently post the safety driver out of the truck and secure approval from regulators to do so and describe the path they need to take to get to that point. Yes so surface ends up being as obviously as the million dollar question of this whole thing. And we're trying to make sure that we're doing it in a smooth responsible cadence I guess is the key here. And what are the pieces that need to go into it. At first I actually need to you need to do a whole lot of reliability testing. So going back and forth on 10 initially this isn't going to be a United States wide thing initially.

Where will the first autonomous trucks be driven?

This is going to be a single route in the southwestern United States as you mentioned. And so you end up we end up having to have a bunch of trucks out testing with safety drive areas to get to a point where we can show that they are reliable enough that they can safely do the whole driving task and we think that's probably going to take a couple of years just to do that initial testing just for that first route. And in terms of what other pieces we have to do we obviously also have to work with the regulators and with the OEM. And both of those are sort of ongoing and the the regulatory landscape is not set in stone yet but we spend a lot of time making sure that we and we work very closely with the regulators they know what we're doing and that they're doing things at a reasonable pace right. It will be bad just like you'll be back if the regulations were late and you couldn't deploy. Even though it safe it may be that if the regulations were early and we wrote the regulations around technology that hadn't been fully proven out yet. So we're working closely with them to make sure that the regulations line up with the technology and they end up coming out on the same timeline. Absolutely. That's very good guidance. And to be clear I mean we're talking millions of miles of testing right. I mean this is not something that happens overnight. This is this is lots of testing extensive on road testing to get to that point. Yeah that's right. We're talking doing probably more miles of testing than any individual human driver will drive in in Korea.

Deployment Models

For now let's go ahead and talk about Embark's deployment model for autonomous trucks. So to be clear you're designing and programming these are bing trucks to operate only on freeways which means that these trucks will need the hand off trailers to conventional trucks piloted by drivers for the remaining short haul into the city or on local routes to a distribution center. And these freight hand offs between autonomous and conventional trucks would take place at transfer hubs that will be built at key points along interstates. So Alex could you tell us a little bit more about this transfer hub concept and why you see this as the best way to deploy autonomous trucks in the relative near term yeah. So the way that we see the deployment of driver-less trucks and you have to be thinking not just about what's technically feasible but about what fits with the industry. And so there's been a lot of noise about a competing concept where you would have the drivers stay in the truck. But we think that while that's technically achievable it doesn't do anything for the big problems facing the industry today. So obviously those are the driver shortage. Those are in particular the shortage in long haul driving. And so we think that if you have a driver still inside the vehicle what you end up doing is creating a new class of driving which in my opinion.

Transfer Hub-Model

I imagine in most people's opinion is even harder than 11 hours and then sleeping and then 11 hours and then sleeping where this new class of driving would have to be you're just sitting in the back of a truck for a long period of time occasionally getting called upon to support it potentially driving or potentially On-Call for many more than the current 14 hours but still away from home all those problems so you could build the tech. Sure. But it wouldn't do anything to make a driver's life better or to make driving jobs more desirable. And so it's not really helping solve any of the problems facing the industry today. And by taking the transfer hub model what we're doing is breaking the route up so that you have what used to be one big long haul route becomes two local routes with an autonomous section in the middle. And this actually is both technically feasible and now starts to address the driver shortage where you've gone from a long haul route which could have been one of the 50000 that we don't have a driver to fill today. And we've actually created new local jobs which are much easier to fill and much more desirable. And we're getting the same freight move. And so that's a big part of the reason why we think you want to go with the transfer hubs model. Yes there are other ways to do it. From a technical perspective once you start thinking about how does this fit with the needs of the industry how does this fit with the needs of the drivers and it makes a lot more sense to try and allow drivers to stay local in their own city sure. You know as you said you know the driver is the real constraint that the industry is facing right now.

Making the driver's job easier

So this concept of transfer hubs is really aimed at finding a way to make that work for the driver and then make the job better for the driver. You know in looking at the technology side of this by restricting the scope of operations to the freeways you can introduce autonomous trucks much sooner. Right. You know it'll take a lot longer to develop the Thomas trucks that can go in theory anywhere that a truck today can go with a driver. Right. Yes. And it will take a lot longer because it's much much harder to do. From a technical perspective to drive in the city and also it's much harder from a logistical perspective. There's a lot of interaction with the customer pre trip post trip inspection bills of lading maybe loading in certain locations and there's a lot of detail at the beginning and the end of a trip that Israeli best suited to a person along with it being a much harder driving job whereas once you get out onto the interstate there's a lot less things that are going on at International Truck. We know that when it comes to the future we all have a lot of questions how do we take the chain off the supply chain. Will my fleet survive in an on demand world will elect to charge ahead of diesel with artificial intelligence play sensitivity to how the life we live come up how we might use it for a college. Gus what's next. Let's talk about it. Join international trend in driving transportation forward. And up next trucking when we think about plutonomy when we think about automation in this industry you know there's always the question of what will the emergence of autonomous trucks mean for truck driving jobs in the years ahead.

Don't Believe The Hype

So what are your thoughts Alex. You know how will the job change. And what would you say to somebody who's concerned that the truck driving job might just disappear altogether. Yeah. So I guess I would say a couple different things. The first one is I don't believe the hype. The media loves sensationalist stories about how there will be no driving jobs in two years from now which is just not in line with the reality of the technical development. And this is a gigantic industry with a huge variety of different driving tasks which are each individually very hard. And all of which are going to take a bunch of time to do and so I don't see for example city driving or even driving in snowy northern areas happening anytime in the next decade. So the first thing I think to say is don't believe the hype the media has their reasons for trying to make things overblown. But but it's just not realistic. The second thing that I would say is that there are a lot of as I said before parts of the driving task that I think are not well suited to automation that are are much better done by a person. And I don't think that's going to change anytime in the near future. So although we might see automation in chunks of the job that make sense like driving long interstate stretches on good quality roads. There are whole big chunks of the industry that you're unlikely to ever get to. For example I don't think we'll ever do hazmat.

Years Ahead

I don't think that we're ever going to be doing like last mile drop off or pick up for the vast majority of situations. The second piece is one, don't believe the hype it's going to be a lot slower than people say too. It's not going to be everything it's going to be sort of piecemeal, bit by bit and it's not going to cover the whole industry. And so I think that means that when you account for the big shortage that currently exists and for the huge amount of turnover that already exists in trucking that pretty much anybody who is presently in the industry and wants to continue in the industry is going to be able to find a job. And then the third thing I would say is that we really don't know the answer. So it's going to be slower. It's going to be less absolute. And also that we it's hard to predict the deployment of technologies. And so I think it's possible that the total number of drivers decreases as a result of automation 15-20 years from now. And it's also possible that it doesn't because for example right now we have 50000 routes that are unfilled. If we fill those in part using automation and in part using local drivers we actually have more driving jobs than we had before. And so there are a bunch of different variables and I don't pretend to know exactly how it's going to play out but I think it's going to play out in a slower smoother way than I would have you believe. Okay. Know that's that's very helpful perspective on what we're really looking at you know in the years and decades ahead.

Autonomy = Increase in Productivity

And you know we often think about the potential for Thomas trucks to improve safety. We maybe even think about the potential to reduce labor costs at some point in the future. But let's also take a moment to consider the potential for autonomy to improve productivity in the industry. So earlier this year an embarked truck completed a coast to coast test run along the length of itan from L.A. to Jacksonville Florida. And there was a safety driver behind the wheel who was still beholden to hours of service limits. So this run took the normal five days. But in theory if you reach the point where you can remove the driver and the truck can operate around the clock or close to it suddenly that journey can complete be completed in two days instead of five. So what are your thoughts on just how that could transform the supply chain someday. Yeah I think this is one of the gigantic pieces of automation that a lot of people don't talk about or don't consider. If you allow a vehicle to operate 24/7 you completely change how supply chains need to be set up right. We're talking about the time to get things from any place to any other place. It's more than 11 hours away suddenly dropping in half. We're talking about being able to achieve like when you say here two days instead of five. I for context that's now going live we're now talking closer to airfreight speeds than what it presently takes a tractor across the country. And so that's a radical again I was talking about earlier waterway's is going to change the industry.

Change in LTL

One of the things we might object, maybe, a lot of air freight switches to trucks. And that actually increased demand anyway. You're going to change the speed of things you're also going to change the layout of distribution centers right. Today the major LTL networks are all set up so that their bad DCs are all 11 hours apart so that you can drive from one switch in the middle and then drive back. And obviously if you go 24/7 that completely changes how you set up warehouses so you're going to see things moving a lot faster you're going to see things like Amazon shipping becoming cheaper and being able to get one day shipping instead of a two day shipping or today shipping instead of one week shipping. And we're talking about being able to radically improve the efficiency of warehouses supply chains and just in time manufacturing. All right. This will all be fascinating to watch moving forward. That's for sure. We certainly wish in a very interesting time for the transportation industry it's a fun time to be in this industry and for me to to write about it and report about it. Thanks again Alex for participating in the podcast. It was great to chat with you. Thanks for having me sir. Just at international track. We know that when it comes to autonomous transportation we all have a lot of questions. Who's in the driver's seat. Will autonomists make our world safer. How will the role of the driver connect the technology to supporting society. From French Trinca what's next. Let's talk about joint international truck and driving transportation forward. Up next company. Next up on road signs.

Fred Andersky, Automation Guru

We're going to bring in a guest who I think has really been one of the industry's leading voices on this topic of automation and trucking. So we're very pleased to welcome Fred Andersky who heads the controls business and government and industry affairs at Bendix a major supplier of breaking an active safety systems that I'm sure many of our listeners are using on the trucks today. So thank you very much for joining us Fred. It's good to be here with you. Let's go ahead and talk about some of the different pathways toward automation that we might see. You know first off you have you know the gradual step by step path of building on today's active safety technology to create advanced driver assist systems. And that appears to be the baseline for development. And then on the other hand you also have companies like Whammo and Uber and some of the tech startups that are trying to go straight to highly automated trucks. Fred what are your thoughts. You know you see the industry pursuing both of these pathways simultaneously in the years ahead. Or do you see one of these approaches clearly winning out. Well so let's let's set the let's set the ground groundwork here. You know so there's four things that are certain within life death taxes change and the failure or the failure of predictions to be accurate.

Far Future is Driverless

So as we think about that you know let's keep that in context as I talked this through a little bit you know over the next five to 10 years I really see both pathways making progress and you know both are going to move us forward in the future you know and it's really going to come down to the applications as to where they're going to be. You hear a lot of talk especially from companies like embarking on Waino and Hubler about this highway application which is really when you think about it one of the easiest applications to do versus what we see in kind of a local or a urban area. So that means then that that local or urban area is going to need help as well. Well my line though is I've got to admit I'm biased because I work for a developer of driver assistance systems. In the long run the stepping stone building on driver assistance systems approach when I think we both have to keep in mind that the real end result is we are going to get to antonymous vehicles either pathway we go eventually in the far future it's going to be driverless. The question is when. Let's go back to the stepping stone approach and you know this idea of building on where we are today with drivers this technology. Of course we already have a collision mitigation to automatically apply the brakes. We already have lane departure warnings to alert the driver when the truck starts to stray over the lane marking and we already have adaptive cruise control. So what do you see as the next stage of this progression. There's been a lot of movement toward automated steering. How do you see that rolling out. And you're you know when you think of the basic functions of a vehicle is to stop go and turn. And as these systems start getting smarter learning becomes another basic function.

System Fusion: A Thought for the Future

So today you know as you pointed out we're doing stop and go active cruise control intervened with the brakes and the acceleration to keep the vehicle safe following distance as we move down the path. You're absolutely right that steering is really going to be that next big approach. So you know we've talked in the past about the idea of sensor fusion centers working together and we see that today cameras LIDAR radars ultrasonic all these types of sensors working together in various applications. When I see moving forward that is kind of the second step in fusion which is what I call system fusion and that's where we take steering control working with other systems on the vehicle to move us forward. And I really see this kind of come across in for progression if you will or four phases. First of all it's going to kind of start with independent low level integration of Styr and SS so independently we're going to see things like tortoise's which will help. He's in theory really not a force that's needed to steer a vehicle. So make it easier in those slow speed situations or when backing up to park the truck at a loading dock some low level integrations will be like using the camera to see the lines in the road. So yes keep Unisys in lane center and types of functions.

Automation Doesn't Mean You'll Lose Your Job

Next we get into kind of the higher level integrations which will be things like systems working together when the adaptive cruise control working with steering control to give the driver a little bit more of a adaptive cruise with steering kind of a break type of situation where they can maybe do something else while the trucks moving down the road or we see blind spot detection working with steering to kind of give a side swipe crash mitigation. Third we move to the advanced automated apps these get into things like a full scale highway pilot type applications we've seen or what we also call as level 2 platooning. So maybe there is not a driver in that back truck that's tuning down the road. Then lastly or the fourth is what I call and what others call to convergence and this is where the steering control or the system fusion with steering control the sensor fusion all comes together to give us that autonomous driving function. So you know those two pathways as we talked about that you mentioned earlier they end up at that autonomous driving point. And that convergence of system fusion gets us there. What would you say to somebody who is concerned that this technology could eliminate jobs. Well you know I think like my first my first aspect on that is you know much as autopilot in planes has not eliminated the need for pilots. And let's face the fact pilots do cost money. I say you know personally I think that we're going to see drivers as an important part of that truck tractor for a long period of time. Change is coming upon us and nobody can predict exactly when that change is going to occur. So I think it's smart for drivers to be excited about the industry they're in. But I think it's also important for them to think about you know what their next act is going to be and how their role is going to evolve in the future.

How the Truck Driver's Job will change with Automation

Yeah any kind of preliminary thoughts on how that role might evolve. You know we think about what truck drivers do today. And you know how that job might change as some of the tasks become automated. What do you see the truck driver of the future doing. How will that job change compared with where we are today. Well I think in the short run as we see more technology coming onboard the drivers are going to learn to be smarter about how they utilize that technology in their day to day job. How is it going to help them and how is it going to help improve the driver experience if you will. In other words you know are they going to get the opportunity to be able to take more breaks on the road. Is there a 11 hour driving day going to expand to a 14 hour driving day because they're going to have time in the cab if you will to get other work done which is going to help make them more more productive overall. But as the road begins to change I think that they're going to see some lifestyle changes in their role that I think are going to be beneficial. I think they'll see more home time because they'll be able to get more stuff done up the road in a shorter period of time. And I think we're going to see the models changing in terms of transportation.

First steps of Automation:

You know I do think the parks and the Ubers and the way most have the right OCSE in terms of this first step being a level of automation out on the freeway or the interstate route which means that opens up the opportunity for more short haul types of things for drivers which again then means more long time. But I also do see the driver becoming important in terms of helping monitor what the systems are doing still being there and being ready to take over as necessary because the systems aren't going to be 100 percent foolproof. And as I mentioned all you need is a virus to get into that vehicle and shut it down and then you're going to need that driver you know to be able to get the last route. So I'm still on the side of driverless a very long time off, a long time away. But changes in the drivers role today that's going to help make them more concocted it and let's say more technically sophisticated in the future. Yeah this all be very fascinating to watch in the years ahead for for sure. And this has been a great conversation but I think that's a good place to leave it. Thanks again Fred for joining the program. We really appreciate your thoughts. Thanks Seth. As always a pleasure to work with you and Transport Topics. AD: Hi I'm Chris cochairs development chief engineer at international Truck. Thanks for listening to the first ever Transport Topics podcast. Interestingly we've heard a lot of thought provoking topics discussed here at International Truck we are driving important changes in our industry. Take autonomous transportation. We know that autonomous driving technology has the potential to transform the industry especially when it comes to safety and efficiency. We anticipate autonomous technology will boost these two factors to levels our industry has never seen before.

Evolution rather than Revolution

Not only will we see a shift in the role of the driver but we'll also see increased efficiencies for fleet owners and International Truck. We're taking a pragmatic approach to technology working to develop test and launch autonomous trucks in a smart methodical way. All this new exciting technology and change is happening faster than ever and it's creating a lot of speculation. We know that when it comes to the future we all have a lot of questions. As industry leaders we had international truck together with you want to address these advancements from electrification to connectivity to digital supply chain and more so we can all move towards a successful tomorrow. Join international truck and driving transportation forward. Upnexttrucking.com as we wind down. I'd like to close with some final thoughts on our original question of who will be behind the wheel of tomorrow's trucks as we've heard from our guests truck drivers will be indispensable to this industry for the foreseeable future. Automation is coming to trucking. But change of this magnitude in an industry this large and diverse simply doesn't happen overnight and it doesn't happen everywhere once we'll see automation emerge in stages with driverless technology. And yes we may also see some fully autonomous trucks start to hit the road in the not so distant future but only in specific applications like driving on freeways in areas where conditions are optimal. It's clear that autonomous trucks will not be ready to handle all driving tasks and conditions anytime soon. That means it's really best to view this as an evolution rather than a revolution. We're in the early stages of what will likely become one of the defining changes we'll see in the next half century of trucking.

The Future Looks Bright

So if you're a 21 year old who just earned a CDL I feel confident that you'll have a full career ahead of you in the trucking industry and your services will be in high demand for decades to come. Your job may change over time as some tasks become automated but you'll have a job and that job may turn out to be far more interesting than you ever expected. On that note we've run out of time but if you'd like to learn more about the road to automation and trucking I'd encourage you to check out the most recent issue of ITech included in the June 11th issue of transport topics or visit T.T. News.com where I can find that content online. Roadsigns will return in August when we'll explore the potential of electric vehicles in the trucking industry. Until then thanks for listening.