After a series of accidents, some of which may have involved Autopilot, Tesla switched from primarily using cameras to “see” the road much like human eyes to adding radar capable of detecting objects as far as several cars ahead. Tesla’s database uses a location-tagged whitelist that lets cars drive safely past “approved” radar objects, while braking slowly if anomalous objects appear. Once the system determines there is a 99.99% chance of a collision, it fully brakes the car.

Vehicle sensing systems by Google (here) are similar to Tesla’s use of cameras, radar, and other instruments to map the surrounding environment in real-time.
Vehicle sensing systems by Google (here) are similar to Tesla’s use of cameras, radar, and other instruments to map the surrounding environment in real-time.
Image: Google via IEEE

When announcing the system on Sept. 11, Musk said the new system would “very likely” prevent accidents like the Florida collision that killed Joshua Brown after his Tesla Model S slammed into a tractor trailer in May.

Tesla has yet to confirm the details of the accident from the video posted by Noordsij.

📬 Sign up for the Daily Brief

Our free, fast, and fun briefing on the global economy, delivered every weekday morning.