TuSimple, a U.S.-Chinese tech startup racing to commercialize self-driving capability for long-haul semi-trucks, just raised a further $95 million in a funding round that the company estimates moves it to unicorn status with a valuation now reaching $1 billion.
The funds will help the company – with technology designed by Caltech-trained neuroscientist Xiaodi Hou, who’s also TuSimple’s founder, president and CTO – expand its fleet of robot semi-trucks to 50 by June from 12 currently, as well as its team of developers, the company told Forbes.
The Series D round closed at the end of 2018, led by Chinese tech firm Sina Corp. and included investment from Hong Kong-based Composite Capital, boosting TuSimple’s total funding to date to about $180 million. Sina also invested in earlier rounds, as did Nvidia and U.K.-based ZP Capital.
Based in San Diego and with an engineering facility and truck depot in Tucson, Arizona, TuSimple has been hauling revenue-generating loads since last year. And while self-driving passenger vehicle programs, such as the service Alphabet Inc.’s Waymo launched in Phoenix in December 2018, have drawn the most public attention, use of the technology for autonomous trucking and delivery could be on a faster path to commercialization.
“The real economic drivers for autonomy are in truck automation,” Chuck Price, TuSimple’s chief product officer, said. And at the moment, “the economic drivers for passenger cars is somewhat different and diffuse.”
Long-haul trucks operating on highways don’t have to deal with pedestrians, bicycles or as much environmental complexity as do vehicles making their way through dense urban settings. There’s also a shortage of long-haul truck drivers, topping more than 50,000 in 2017 according to the American Trucking Association, which means there’s near-term demand for technological solutions. TuSimple also has lots of competition, with rivals including San Francisco-based Embark and Starsky Robotics, startups Kodiak and Ike, which both have former members of Google’s Self-Driving Car Project, and Waymo, the Google program’s commercial successor.
Along with aggressive fundraising, TuSimple has lined up production partnerships with truckmakers Peterbilt and Navistar, as well as powertrain specialist Cummins. And although it’s building the self-driving tech and not the vehicles themselves, it’s going to keep raising funds.
“We will do another raise this year,” Price said. “It is an expensive proposition even though we’re not building a factory.”
TuSimple splits its R&D focus and commercial applications between the two markets in which it operates. In the U.S., the focus is solely on long-haul trucking whereas in China, the technology is used for drayage operations, hauling cargo at ports, founder Hou has told Forbes. Its Arizona truck fleet is loaded up with many of the same types of advanced sensors used for self-driving cars, including laser lidar units, radar and high-definition digital cameras.
TuSimple’s vision system, however, relies most heavily on long-range cameras that it says allow its robot drivers to track conditions up to 1 kilometer ahead, or about three times farther than Waymo’s long-range lidar units can see. The company says that's a potentially big benefit for trucking companies, as it helps vehicles maintain the best possible speeds to maximize fuel economy and reduce wear and tear on brakes and tires. If it's right, that would translate into lower fuel and maintenance costs.
“Autonomous driving is one of the most complex AI systems humans have ever built," Hou said in a statement. "After three years of intense focus to reach our technical goals, we have moved beyond research into the serious work of building a commercial solution.”
