Pitfalls of autonomous car startups could slow the adoption of self-driving trucks

Under pressure from deteriorating economic conditions, some investors and companies are slowing their rush to build fully self-driving cars and instead focusing on semi-autonomous vehicles with “driver-assistance” features to promote safety for human drivers, an industry expert says.

The latest example came last week when Ford and Volkswagen shut down Argo AI, the Pittsburgh-based startup they co-founded — and heavily funded — to develop fully autonomous commercial vehicles. This news was closely followed by the unveiling of lawsuits against emerging autonomous vehicle (AV) automakers Tesla (accused of overdoing its “Autopilot” system in passenger cars) and TuSimple (over alleged technology transfer to a Chinese company). TuSimple promptly fired its CEO, Xiaodi Hou, but said it would continue work related to reservations for autonomous trucks from major cargo transportation providers like Penske, Schneider, and US Xpress.

Those stumbling blocks come as both small startups and big automakers have been pouring resources into AV technology since around 2014, says Tony Wayda, director at JBF Consulting, a Connecticut-based logistics consulting and systems integration firm. This race to bring new products to market has recently slowed to a more strategic rollout of the components that have proven successful so far, as investors look to generate reliable revenue rather than fund speculative R&D.

According to Wayda, the approach is measured against the six levels of autonomous driving defined by the Pennsylvania-based technical standards group SAE International. These ranks range from level 0 (no driving automation) through level 1 (driver assistance), level 2 (partial driving automation), level 3 (conditional driving automation), level 4 (high driving automation) and level 5 (full driving automation). ).

“Given the current economic downturn, recession fears and tightening [private equity] and [venture capital] Dollars, companies reevaluate their position at Level 4 and Level 5. While we believe Level 4 is achievable, it’s still a few years away from going mainstream,” Wayda said in an email. “Because of this, we are seeing automakers starting to focus less on Level 4 & 5 AD and focus on automation, which can increase safety, which in turn can increase sales. Level 2 and Level 3 are achievable in the near future and in some cases already implemented, like lane assist and automatic braking.”

These events appear to have halted the development of full AV technology for commercial passenger vehicles, but related efforts in road haulage continue. Just this week, San Francisco-based Embark Trucks announced that it has expanded its coverage map of pickup points for trucks equipped with its autonomous technology. The company continues to pilot trials where its vehicles will haul loads across freeway stretches before handing them over to human drivers for last-mile delivery. Other AV technology providers such as Torc Robotics, Volvo Autonomous Solutions, Kodiak Robotics, Waymo Via and Gatik are following similar patterns.

Despite these advances, the many emerging AV truck vendors may also slow their progress and begin to take a more conscious path toward the twin goals of full commercialization and full autonomy, Wayda said.

“The more fundamental problems with [AVs] are the safety concerns and the public perception,” Wayda said. “While we anticipate automakers will continue to integrate more Level 2 and Level 3 capabilities into their systems, moving to Level 4 will require much greater effort. Legislators and government agencies must work closely with the [AV] Technology companies and automakers to gain public trust. We believe this will likely happen in phases.”

For example, with enough planning, autonomous trucks could begin doing their highway miles on dedicated lanes on certain U.S. interstates, where they could travel along lanes cordoned off with barriers, similar to expressways in high-traffic areas, he said.

Leave a Comment