We are excited to bring Transform 2022 back in-person July 19 and virtually July 20 - 28. Join AI and data leaders for insightful talks and exciting networking opportunities. Register today!
This article was contributed by Andrey Bolshakov, founder and CEO of Evocargo.
Autonomous vehicles are subject to far more stringent requirements than long-haul truck drivers and their vehicles. After all, there is no margin of error for robots. That is why self-driving cars are tested more thoroughly, even on public roads, and why the media and the public react so strongly when they fail. Driverless trucks, on the other hand, are generally tested on closed courses because of their weight (over 40 tons) and speeds (with braking distance more than 300 feet).
All the same, forward-looking companies are testing driverless trucks outside their usual warehouse and terminal locations. On February 16, top U.S. logistics provider C. H. Robinson announced that it will be collaborating with Waymo Via to run autonomous Class 8 trucks on a logistics route between Dallas and Houston. If testing is successful, C. H. Robinson will be able to offer the technology to its network of 200,000 shippers and carriers, and the United States will be at the center of a logistics revolution.
Self-driving truck startup Gatik announced in late 2021 that it would partner with discount retailer Walmart. Several of its driverless trucks are already running along a 7-mile route between a Walmart distribution center and one of the company’s stores in Bentonville, Arkansas. The trucks can also be seen on routes in Ontario, Canada and Texas.
Event
Transform 2022
Join us at the leading event on applied AI for enterprise business and technology decision makers in-person July 19 and virtually from July 20-28.
In addition, autonomous truck developer Aurora recently partnered with Volvo to create driverless truck and automate processes for US Xpress. San Diego startup TuSimple also announced last year that one of its trucks delivered a shipment of watermelons 951 miles from Arizona to Oklahoma City in record time, 42% faster than a driver could have made the trip.
While these projects give cause for guarded optimism, it is important to remember that they all involve a set route with established infrastructure. The lack of clear rules for using public roads is another factor holding the sector back. All hub-to-hub driverless routes are still in test mode, and many more such tests will need to be completed to prove the safety of autonomous trucks. For now, California, Arizona, New Mexico and Texas are the largest states that allow testing of driverless vehicles on public roads. Until a uniform set of rules is developed, autonomous interstate logistics will remain out of reach.
Launch an autonomous system in five days
Using driverless trucks in a warehouse environment is significantly simpler than on public roads, so companies can focus on autonomous vehicles for enclosed spaces and the benefits it can provide.
The first step to launching an autonomous logistics system, with or without existing IT infrastructure, is to create a digital twin of the entire enclosed space. After that, the routes are marked and loaded into each vehicle so they can navigate the space even without a strong GPS signal or additional infrastructure like RFID and magnetic tapes typical AGV requires. Any logistics provider with driverless solutions will also train the people who will charge and start the vehicles and make sure everyone knows where the vehicles will run. Evocargo’s trucks, for example, are equipped with computer vision, so they stop or maneuver around objects and people that appear in their path. Supervisors work remotely, with up to ten trucks controlled by a single individual. They can reboot a truck’s onboard computer or return it to its base if it encounters problems, but for safety reasons the supervisors do not have full control over the trucks.
Companies must develop software that leverages machine learning and classic robotics to control and service the driverless trucks, even outside a controlled environment.
The Pod from Einride can travel up to 12.5 mph (for comparison, the EVO.1 has a top speed of 25 mph) and has 1 AET and 4 SAE autonomy ratings. This means that the truck can follow set routes in a controlled environment. Instead of a driver’s seat, the Pod has sensors installed behind a tall windshield for the best possible lidar access. Einride is currently running several pilot programs for the Pod, including one at the 920-acre Appliance Park campus in the United States and one with DB Schenker, SKF, and Coca-Cola Europe in Sweden.
Some of the world’s largest port terminals are also turning to automation. The Port of Rotterdam has 85 driverless trucks that have moved over 100,000 containers since 2011. In 2023, VDL Automated Vehicles will supply the port with 77 new trucks featuring an upgraded navigation system with a vehicle-to-vehicle communication protocol. The new trucks will reduce wait times and move more containers for improved efficiency in operations.
Autonomous vehicles are still expensive but offer benefits for drivers
Studies show that automated cargo hauling can save as much as 35% of logistics expenses, but the upfront cost of the infrastructure can be daunting, even for a major player like C. H. Robinson.
We estimate that companies can save up to 60% on logistics by implementing a driverless system, mainly through reduced human error, fewer accidents and improved scheduling opportunities. Electric motors are also simpler and more reliable than gas combustion engines and require little in the way of servicing. Adding hydrogen fuel cells to the package is a way to further reduce the industry’s carbon footprint.
Drivers aren’t disappearing any time soon, as they will continue to be needed for first- and last-mile trips and difficult driving in big cities. Automation is also creating new jobs for remote truck operators that can monitor multiple vehicles at a time. These changes are vital as fewer young people want to drive trucks, and those that do prefer shorter routes that allow them to spend more time at home.
Autonomous vehicles: Driverless trucks on the highway
A lot of legislative and infrastructure work has to happen before driverless trucks can become a regular feature on hub-to-hub routes.
The main technology driver will be the lidar devices that help autonomous vehicles measure the distance to objects in real time. As the market grows, the cost of lidar will drop and the devices will become more compact and easier to use. Eventually, lidar technology will improve autonomous vehicle productivity and reduce their energy use.
Self-driving trucks on hub-to-hub routes currently rely on a strong internet signal for their operation, but satellite internet projects like Starlink stand to solve that problem.
More electric vehicles on the road will require greater investments in the grid. This is a problem that can be solved through large-scale production of green hydrogen, which will both reduce the cost per mile and make driverless electric vehicles safer and more sustainable.
The transition to autonomous trucks will create new jobs and significantly improve logistics quality and safety. Right now, however, the industry is dependent on decisions being made by legislators and governments. One thing is certain: people need to see driverless trucks on highways before they can get used to them. And once people see how safe these autonomous vehicles are, they’ll be more interested in driverless passenger cars. As a result, the industry faces the triple challenge of carefully testing its vehicles, guaranteeing their safety, and working with legislators to create a legal framework in which autonomous logistics can succeed.
Andrey Bolshakov is the founder of Evocargo.
DataDecisionMakers
Welcome to the VentureBeat community!
DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own!