Autonomous driving: robotic truck trains in the rocky desert

Artificial intelligence should allow trucks to be on the road without drivers in the future. The manufacturer Daimler collects the necessary data in the US state of New Mexico. A visit to the site.

In Albuquerque, New Mexico, a city of half a million people, one color dominates: terracotta. Around a stone desert, the air is dry and dusty. “If you can do Albuquerque, then you can do the whole of America,” says Peter Vaughan Schmidt, head of autonomous trucks at Stuttgart-based automaker Daimler Trucks. “For us, data is really the source of optimization.” It’s data from “an extremely demanding environment here in Albuquerque.”

Three years ago, Daimler teamed up with US software maker Torc Robotics. The Stuttgart-based company is a leading truck builder. But when it comes to software and, in particular, self-driving trucks, all European manufacturers have a big deficit.

Up to 20 trucks on the road every day

The Cascadia is a powerful truck from the Freightliner brand, a 100 percent subsidiary of Daimler in the USA. From the test site in Albuquerque, Daimler sends up to 20 trucks onto city roads every day, day and night. Training is everything for AI systems.

At Torc there are different teams that evaluate and categorize each driving test. This is the only way for an artificial intelligence (AI) system to know how to react later. “Perception is really how you see the world. What’s around me? What’s moving right now? That’s the only way to really drive safely,” says Andrew Culhane, chief strategy officer at Torc. “But we also assess behavior. Does the truck change lanes, speed up, slow down, merge? This gives us an overview of how a truck should react.

AI systems are only as good as their training data. This is fed to the computers in the test truck. This is already happening today with Daimler trucks in the US, which have on-board assistance systems and are operated by trucking companies. These also collect data for future robotic trucks.

It was initially only used in the US.

Anyone sitting in the driver’s cab will notice a locker in the rear. A server is running there that evaluates dozens of sensors, including lidars, that is, special optical distance meters, laser sensors, and cameras. The high-performance computer is the brain of the vehicle. You must be able to react to all eventualities.

“A motorcyclist could fall on the road at night wearing black clothing,” says Vaughan Schmidt. “Can we handle something like this or not when the self-driving truck arrives? That’s one case we’ve solved. There are also many, many abstruse cases that you can’t even imagine.”

Trucks will probably be the first vehicles to drive on our roads without drivers. At Daimler, it is hoped that in a good seven years, that is, in the year 2030, the time may come. However, only in the US due to its narrower borders and roads, Europe is likely to be several years later.

Slower pace, wider streets

Advantage of the United States: Here all vehicles drive more slowly, trucks and cars move at the same speed. The roads are wider and there are fewer curves. The plan is for the trucks to only drive sections of the route autonomously. At the highway exit, a person gets on and takes over.

Daimler also provides a control room for freight forwarders. There should be people here, often thousands of miles away, who can remotely intervene with any truck. “In the autonomous truck, we need the ability to communicate. This is particularly important because real-time communication is necessary in logistics to make everything work,” says Daimler’s Klara Oberhollenzer. “In the future, our Mission Control will allow driver managers to communicate with this autonomous vehicle without being software engineers.”

It turns out that the world of robotic vehicles is a lot less spectacular than Elon Musk and Tesla would have you believe.