Autonomous vehicles are often touted as the transportation method of the future, and GM has invested heavily into its Cruise division, which aims to bring robotaxi to the masses. So far, Cruise robotaxi prototypes, which are based on the Chevy Bolt EV, have clocked 250,000 miles without human input, but one unit was involved in an accident with injuries.
Perhaps not unexpectedly, many are still wary of the technology, unsure if computers are actually capable of safer driving than humans. It turns out that there may be some credence to those concerns, as a recent report suggest that fully autonomous cars may need some human input in order to operate safely.
According to Reuters, industry executives and experts agree that autonomous vehicles may require remote human supervision in order to function safely. The outlet states that making AV technology safer than human-driven vehicles is “immensely tough because self-driving software systems simply lack humans’ ability to predict and assess risk quickly.”
This could especially spell danger when unexpected incidents arise. In other words, it’s nearly impossible to predict every situation on the road and program software to handle it. Instead having remote human supervision can help AVs react to these “edge situations” and potentially get them out of incidents they wouldn’t otherwise be able to handle on their own. More specifically, human supervisors could be able to remotely control AVs that have stopped because they don’t know how to proceed.
Kyle Vogt, CEO of Cruise, weighed in on the situation. When asked if humans could be completely removed from the robotaxi equation, he said, “Well, my question would be, ‘Why?’ I can provide my customers peace of mind knowing there is always a human there to help if needed. I don’t know why I’d ever want to get rid of that.”
Cruise’s autonomous vehicles currently rely on humans for less than one percent of driving time. While that may seem like an insignificant statistic, it can certainly stack up over time.