Tesla is improving driving visualizations to differentiate cars in its own lineup - Electrek.co

Tesla is working on improving its Autopilot driving visualizations to differentiate cars in its own lineup.

Over the last year, Tesla has been putting a lot of effort into improving its driving visualizations.

Powered by Autopilot’s suite of sensors and the automaker’s computer vision system, Tesla renders the vehicle’s surroundings on the screen.

It’s in no way a substitute for watching around the car, but it has been regarded as a confidence builder for Autopilot.

The visualizations have improved over the years, and they are increasingly showing more of the environment accurately, with Tesla adding trucks, SUVs, motorcycles, and pedestrians over the years.

More recently, Tesla has implemented significant upgrades to render traffic lights and stop signs as it prepares to release more driver-assist features for city driving.

Now Tesla is also adding several new renders to differentiate vehicle models and types on the road, as reported by Tesla hacker Green:

It looks like Tesla is planning (already doing?) display of (recognized) Tesla cars on autopilot cars visualization as such. Keep your eye on it?

Also no models yet but coming later:
Ambulance, Fire truck, Police car and "construction" pic.twitter.com/pdnJVV8Dwz

— green (@greentheonly) August 9, 2020

As shown in the image, Tesla is going to feature renders of its current vehicle lineup, Model X, Model 3, Model S, and Model Y.

It is also apparently preparing to render service vehicles and construction vehicles.

As we recently reported, Tesla is going through “a significant foundational rewrite in the Tesla Autopilot.” As part of the rewrite, CEO Elon Musk says that the “neural net is absorbing more and more of the problem.”

It will also include a more in-depth labeling system.

The CEO added that it will “dramatically improve” Autopilot visualization and Tesla will be able to quickly roll out better iterations.

Electrek’s Take

Driving visualizations are great to build confidence into Tesla’s driver-assist system and eventually, Tesla’s self-driving system.

If it accurately represents its environment, it means that it accurately detects it – now whether or not it acts appropriately on it is another thing, but it’s a step in the right direction.

Detecting a specific car model doesn’t change much, but it does show the capability of Tesla’s computer vision system.

FTC: We use income earning auto affiliate links. More.

Subscribe to Electrek on YouTube for exclusive videos and subscribe to the podcast.