Liability, technology and secrecy were among the issues Shailen Bhatt had to consider before allowing Uber Technologies’ Otto subsidiary to use a self-driving truck company to haul a load of Budweiser 120 miles along Interstate 25 in Colorado last week.
Bhatt, the executive director of the state’s Department of Transportation, liked the idea of Colorado hosting the first commercial shipment by a self-driving truck. But he said coordinating the logistics of the two-hour delivery of 51,744 cans of Budweiser beer from Fort Collins to Colorado Springs was complex and demonstrated why it will still be years before self-driving trucks are hauling goods on a daily basis.
In an interview with Trucks.com, Bhatt explained how the Otto beer run came together and where he thinks autonomous vehicle technology is headed.
Did Otto approach you? When was that, and how did it happen?
Otto came to us about three months ago and said, “Hey, we have this truck that we’ve driven across the country in autonomous mode with a driver in the seat, and the technology works. We’re looking for a place to do the first autonomous commercial shipment, and we want to do it in Colorado.”
What was your first thought?
We looked at a few possible answers. We could have said no, we didn’t want to be involved because of liability issues. Or we could say yes, which we did, because we believed this was something that could save lives. We wanted to be involved so that we could ensure this [technology] was as safe as possible for public safety in the state.
What did your department have to do to prepare for the autonomous drive?
This set us on a course of three months of conversations with multiple agencies. The process took hundreds of hours of preparation, sometimes testing 16 hours a day with members of our team and the Colorado State Patrol, as well as our RoadX program, which is a national leader in connected and autonomous vehicle technology.
Our protocol was that the truck had to make this trip eight times with a driver [in the cab] without the driver taking over before we would allow the computer to drive the whole route [with the driver in the sleeper compartment].
Our involvement was multifaceted. We made sure the roadway was swept of debris, and we got our tow trucks involved to make sure there weren’t any abandoned vehicles on the side of the road that would hamper the test. We worked with our road construction contract partners to make sure there weren’t any unexpected road projects going on that would cause weird traffic conditions. The highway patrol had a lot of questions about how they would pull over a truck with no driver in the seat and how we communicate effectively with the autonomous vehicle.
Was this a one-time event or does the Colorado Department of Transportation anticipate more autonomous truck tests in the state?
This was a one-time special event, and this is how we handled it. If we are going to have autonomous trucks across the country, we don’t want them to have to work through a hodgepodge of different regulatory environments. We want to work with the feds and with our state Legislature to say “OK, what are the legal frameworks that are needed so this can be done?”
We will continue to work with Otto and partner with other companies who want to pursue this technology in a way that is done safely. This isn’t something we anticipate regularly running up and down I-25.
For the most part, was the Otto-Budweiser autonomous beer run kept secret during the three-month process?
It wasn’t top secret, but we weren’t announcing it except to the agencies that needed to know. Our state patrol partners were concerned: If word got out to the public that a driverless truck was going to be going down I-25 at a certain time, they didn’t want anyone to say, “Hey, how can we mess with that truck?” We didn’t want there to be a spectacle, so it was done without a lot of fanfare.
When does the Colorado Department of Transportation anticipate seeing this technology on the road?
It’s hard to make predictions about the future, but I think we are going to have autonomous trucks here in the short-to-medium term. I think in five to 10 years, there will be a lot more trucks driving autonomously, but I don’t think this will take drivers out of the truck.
The reason we are interested in autonomous vehicle technologies out there is because there are about 35,000 deaths on [U.S.] roadways each year. We somehow accept 35,000 deaths on our roadways as a fact of life. However, a majority of crashes are caused by driver error. Autonomous trucks and computers could do a better job than humans because they don’t get tired, they don’t take substances they shouldn’t, they don’t get distracted. But the technology isn’t perfect. There are still cybersecurity concerns that must be addressed.
Does Colorado want to be a test bed for autonomous driving?
If you look at our RoadX program, which is becoming a leader in autonomous vehicle technologies, our state is experiencing a high level of growth. We can’t build our way out of congestion because of this growth. So the reason we want to become a state where companies feel like they can deploy their technology is that we believe this technology is going to help us reduce the loss of life on our roadways.
How do you think drivers will react to seeing big rigs rolling down the road without a driver in the cab?
I would tell any truck driver now that I don’t think the public is going to accept driverless trucks driving down the roadway without a driver in the cab. I think it provides the public some assurance that a driver is still in the truck because the technology is still brand new.
The way I see this working in the future is that when truckers get tired, as all humans do, the self-driving technology would take over and allow the driver to rest in the sleeper berth. Then when the driver is rested, they would get back behind the wheel and drop off their load. I see a hybrid of this in the medium-term of where truck drivers will still be in the trucks.
What surprised you the most about the test?
The performance of the truck. It drove right down the center of the lane, it adjusted its speed to other trucks and cars around it. There was one instance where a car slowed down to about 35 mph for some reason and the truck slowed down to the appropriate distance, so to me that was striking how well the truck performed.