Federal safety regulators are opening an investigation into Tesla's self-driving system, responding after years of complaints and accidents involving the software.
The announcement by the National Highway Traffic Safety Administration (NHTSA) comes just a few days after Tesla's Elon Musk held an event in Los Angeles to introduce plans for a self-driving taxi -- called a cybercab -- to transport passengers without a driver.
Specifically, NHTSA said its investigation would seek to determine whether Tesla's "supervised full self-driving" system was responsible for four collisions, including one that killed a pedestrian.
The probe is intended to find out whether the system has enough safeguards that require drivers to take control when the self-driving software can't handle a situation on its own.
NHTSA said it would “examine the system’s potential failure to detect and disengage in specific situations where it cannot adequately operate.”
How well can it see?
At issue in the accidents being studied is whether Tesla's system operates well in low-visibility situations -- rain, fog, dust and darkness. Other manufacturers use radar and laser technology to detect objects that are hard to see, whereas the Tesla system primarily depends on cameras.
The investigation covers 2.4 million Tesla vehicles, including cars manufactured as far back as 2016. All of Tesla’s passenger models are involved, the agency said.