Tesla Autopilot now enables the car to perceive space around it

Date:

Tesla Autopilot now allows the car to sense the space around it thanks to the development of its Occupancy Networks. Tesla’s Autopilot Software Director, Ashok Elluswamy, shared a detailed thread on Twitter about a recent workshop the Autopilot team held. He also shared the workshop on Twitter.

In the video and Twitter Thread, Ashok explained how Tesla developed Occupancy Networks to literally give the car a sense of its surroundings. Humans have the ability to understand the objects around them at any time. Is that car driving on the road at a low speed or a high speed? As a pedestrian, do I have enough time to cross the street before I get hit? What’s that in the middle of the road? What’s falling from the sky? I should get out of the way.

These reactions to split-second scenarios and decisions come naturally to humans. Tesla’s Autopilot team is working to program the vehicles to do the same and this will save lives. Imagine if the car can correctly detect its surroundings while the driver is not even paying attention. An example is sudden unintended acceleration (SUA). Ashok pointed out that Autopilot prevents about 40 such accidents every day.

The workshop took place in June at this year’s Conference on Computer Vision and Pattern Recognition (CVPR.) in New Orleans. Ashok explained that the team developed Occupancy Networks that allow the car to predict the volumetric occupancy of everything around it.

Ashok explained that the typical approaches, such as image space segmentation of free space or pixel-wise depth, have many problems. The solution to those problems is Occupancy Networks.

In other words, Occupancy Networks allow the car to sense the space around it and determine whether it can drive in that space. For example, if a UFO were to suddenly crash in front of you while you were driving, you would react quickly and in the safest way possible. This is what the Autopilot Team trains the software for.

Ashok shared details on how Occupancy Networks used Neural Radience Fields (NeRFs). “The occupancy representation of these networks allows for differentiable representation of images (based on the work of Neural Radiance Fields). However, unlike typical NeRFs, which are scene-by-scene, these occupancy nets generalize across scenes.

You can read Ashok’s full twitter thread here and you can watch his presentation here. We’re just over a month away from Tesla’s AI Day and I’m sure Tesla will share more about the life-saving technology it’s working on, as well as the Optimus Bot.

dr. Know It All recently posted a video about the new 10.69 update and shared its thoughts on Occupancy Network.

In a post on Twitter, he told me: “The great thing about Occupancy Networks is that the car doesn’t need to know what the objects it sees are, it just needs to know that they to be there to avoid them!”

Note: Johnna is a Tesla shareholder and supports its mission.

Your feedback is important. If you have any comments, concerns, or a typo, please email me at [email protected]. You can also reach me on Twitter @JohnnaCrider1

Tesla Autopilot now allows the car to sense the space around it







The Valley Voice
The Valley Voicehttp://thevalleyvoice.org
Christopher Brito is a social media producer and trending writer for The Valley Voice, with a focus on sports and stories related to race and culture.

LEAVE A REPLY

Please enter your comment!
Please enter your name here

Share post:

Popular

More like this
Related

Hurricane Ian slams southwest Florida with “catastrophic” storm surge

Hurricane Ian is expected to inundate some parts of...

Khloe Kardashian reveals on The Kardashians that Tristan Thompson PROPOSED

After finally opening up about her ex Tristan Thompson's...

Biden Defends Social Security Against GOP Proposals

Democrats have spent much of the mid-term campaign arguing...

Brett Favre’s Charity for Needy Kids Gave $60K to His Daughter’s High School for Volleyball Facility

Before Brett Favre reportedly transferred $5 million in Mississippi...