Tesla Turns to Radar to Make Autopilot Safer

Tesla has upgraded its software to version 8.0 and with it comes some major changes to Autopilot, the biggest of which is the prominent use of radar. 

When radar capability was added to all Tesla models in October 2014, it was meant to only supplement the existing camera imaging function. Now, Tesla has modified it’s Autopilot to be primarily dependent on radar imaging while driving the car.

“After careful consideration, we now believe it can be used as a primary control sensor without requiring the camera to confirm visual image recognition,” Tesla said in a statement posted to its website on Sunday.

SEE ALSO: Autopilot Not to Blame in Fatal Accident Involving Model S, Tesla Says

That means having to greatly improve the radar. The 8.0 software does that, by adding access to six times as many radar objects, with more information per object. Also, the new software compiles radar snapshots to form a 3D “picture” of the world. That allows Autopilot to better map the road and make the right decisions in steering and braking the car.

Another system used with the new Autopilot is called fleet learning. This feature will see each car uploading mapping information to the Tesla database so that every car in the fleet can access it. Road signs and cross bridges are just some of the obstacles that will be scanned and uploaded to the database, allowing every car to understand that these do not pose a collision danger.

Finally, Tesla has given the radar the ability to bounce a signal under a vehicle ahead, allowing the car to brake in the event of an obstacle presenting itself suddenly on the road in front of the vehicle ahead, even in zero visibility situations.

Tesla’s Autopilot has come under scrutiny lately, with several crashes suspected of having occurred with Autopilot engaged. So far, though, there has been no proof that the Autopilot has been the cause of a crash.

Discuss this story at our Luxury Lifestyle Forum

Leave a Reply