Delivering on Autonomous Transport
Einride ramps up industry partnerships and on-road demos of its all-electric self-driving truck that’s missing one thing: a cab for drivers.
“Data is the most efficient propellant.” This expression underpins the founding in 2016 of the Swedish startup Einride. Now, the fledgling tech company is partnering with well-established transport-intensive companies—at an increasingly accelerated rate—to run its cab-less autonomous truck, the T-pod, at their facilities and even on public roads. These activities are coordinated by intelligent routing software that integrates customer data, traffic data, vehicle data, etc. to optimize delivery time, battery life and energy consumption.
The first commercial deployment of the T-pod, which Einride claims is the first truck specifically designed for electric propulsion and autonomous driving, occurred this spring at a DB Schenker facility in Jönköping, Sweden. Ericsson and Telia made the installation of the autonomous electric transportation (AET) system possible via 5G-based connectivity at the site.
Daily transport between a warehouse and a terminal commenced in May. The route includes a short stretch of public road within an industrial area where traffic speeds are typically low. The Swedish Transport Agency permit to operate is valid until December 31, 2020.
“Autonomous trucks will become increasingly important for the logistics sector,” Jochen Thewes, CEO of logistics firm DB Schenker said in a release. “Together with Einride, we have now introduced autonomous, fully electric trucks to a continuous flow on a public road—a milestone in the transition to the transport system of tomorrow.”
Einride also is collaborating with Swedish logistics company Svenska Retursystem to initiate a pilot project with the goal of introducing multiple T-pods at a logistics center in Västerås, Sweden. And Michelin announced at the Movin’ On Summit in June that it would deploy Einride’s T-pod to transport goods at a production facility in Clermont-Ferrand, France, beginning in 2020. A second phase, requiring approvals from the French authorities, will involve transporting goods on public roads between different Michelin sites in the city.
Einride has its sights set on the United States as well. The company has several customers based in the U.S., a spokesman told Autonomous Vehicle Engineering, and it plans to roll out commercial installations there in the not-too-distant future.
Because the T-pod has no driver’s cab, the vehicle can be smaller—approximately 7 x 2.5 m (23 x 8.2 ft)—but with increased loading capacity compared to a conventional truck. It can transport 15 euro-pallets and weighs 26 tons fully loaded. The vehicle’s drivetrain concept is centered around an e-motor with direct drive coupling to traditional axles. With a battery capacity of 200 kWh, Einride estimates around a 200-km (124-mile) range on one charge. The electronically limited top speed is 85 km/h (53 mph).
The company has also developed an “up-powered adaptation” of the T-pod for the logging industry. Equipped with a 300-kWh battery, the T-log can carry a 16-tonne load and navigate forest roads. Capable of SAE Level 4 self-driving, the truck’s operation is supervised and can be operated by remote control when necessary.
“Connectivity is key for Einride,” chief technology officer Pär Degerman told Autonomous Vehicle Engineering. “We can use 4G for the installations we have today, but of course 5G will become critical over time when we start managing fleets of autonomous, electric vehicles.”
Einride has partnered with Ericsson to explore the opportunities for 5G mobile connectivity. The two companies demonstrated the technology earlier this year at the Mobile World Congress in Barcelona, allowing show-goers to remotely operate a T-pod.
“Our main concern now is that operators start rolling out 5G as soon as possible and that they rethink how they build their mobile networks,” Degerman said. “5G is all about industrial applications like transport and manufacturing. It is where value creation can happen first, which means we need 5G connectivity not just where people live, but also where businesses and organizations operate and in between those places, along transport-intensive roads.”
T-pod boasts 360-degree awareness of its surroundings, with no blind spots nor dead angles, thanks to strategically placed cameras, radar sensors and lidar, or “3D laser scanners.” The NVIDIA Drive platform provides a robust GPU architecture that can process the high data rates coming from the sensors as well as a safety-certified base platform for high integrity control loops, according to Degerman.
“Apart from NVIDIA, we also have a number of other computing systems that power different parts of our solution,” he explained. “They range from safety-critical systems, real-time embedded platforms to powerful processing units for resource intensive algorithms.” Another important partner in Einride’s quest to bring autonomous vehicles to market is DeepMap, a Silicon Valley-based provider of HD mapping, real-time localization and simulation data.
Wanted: smart, creative types
Einride is continuously growing its development team, Degerman said, and is actively recruiting “smart and creative” people with a range of aptitudes and skillsets. Alongside job openings for systems engineers, test engineers and safety engineers on the company’s online career page are open positions for software engineers, visual and UX designers, and perception/localization developers.
He noted that Marc Llistosella, former president and CEO of Mitsubishi Fuso Truck & Bus and head of Daimler Trucks Asia, recently joined Einride’s board of directors. “His track record of building commercially successful businesses around electric vehicles will be a great contribution to Einride, as we continue to scale the company. Mr. Llistosella’s prime focus will be to secure the scaling of supply of hardware.”
Einride engineers are currently developing the second-generation T-pod, according to Degerman. “We are working on the industrialization process, to scale production, as well as sizing our operational capabilities in order to deploy commercial customer installations,” he shared. “Another challenge is standardization of different connectivity types and business models on all markets we operate or plan to operate.”