Tesla’s super-sensory autopilot system might be able to predict a collision and react faster than a human driver, a new dash-cam video suggests, using radar technology that allows it to see what we cannot.
The 40-second clip, first reported by tech site Electrek, shows a highway in the Netherlands stretching before a moving Tesla (TSLA). Eight seconds in, a rapid beeping sound is followed by a car rear-ending an SUV, which then flips over on impact.
Original video, authorisation from the owner. Essential, no one could predict the accident but the radar did and acted by emergency braking. pic.twitter.com/70MySRiHGR
— Hans Noordsij (@HansNoordsij) December 27, 2016
CNBC reports that the company has confirmed that that the sound heard in the video, which was shared on Twitter more than a thousand times, was in fact the vehicle’s Forward Collision Warning.
As reported by CNBC, Tesla said in a September blog post that the vehicle’s Autopilot 8.0 software uses radar signals to “bounce” around objects, detecting not only what lies directly ahead but also other movements that a driver may be blind to.
“Taking this one step further, a Tesla will also be able to bounce the radar signal under a vehicle in front —using the radar pulse signature and photon time of flight to distinguish the signal — and still brake even when trailing a car that is opaque to both vision and radar,” CNBC quoted from Tesla’s post.
The car’s driver reportedly said the warning system, sensing that a car up ahead was stopping suddenly, activated its brakes before he could react, stopping well behind the other vehicles and avoiding collision, according to Electrek.
Despite what looked like a major crash—sending an SUV rolling over along the highway—the driver reportedly said that everyone involved “turned out to be OK.”
everbody turned out to be ok as I read in the Whatsapp message that brought me the video.
— Hans Noordsij (@HansNoordsij) December 27, 2016
For more on Tesla’s autopilot development, watch Fortune’s video:
Tesla’s CEO, Elon Musk, has said he aims to test fully automated vehicles by the end of 2017, according to TechCrunch, a development the company believes will improve its safety capabilities.