Image: Dirty Tesla/YouTube
Tesla is gearing up for the imminent release of the FSD Beta V11 update, which contains the highly anticipated single stack. The wide release to testers will happen gradually over the next weeks and will probably be available to all participants in December.
Tesla is steadily moving closer to achieving full autonomy by improving its Full Self-Driving (FSD) Beta. While FSD Beta is already amazing and often able to move its passengers without a single intervention, up to this point it has not been as good as Elon Musk and Tesla Autopilot team want. This was because it still used a separate stack for city streets and a separate stack for highway traffic. Switching to a single stack required a huge amount of neural network training, which Tesla's team has been busy with in recent months.
On November 11, Musk announced that at 11:11 pm, the company will start rolling out FSD Beta V11. Initially, it will be available only to some employees of the company, who must test it. The head of the company explained that given all the updates that are in the new version, it will take several weeks to expand it. After that, it will take a few more weeks to get a wide release in the US and Canada.
Given all that is in V11, it will take a few weeks to expand the beta, then another few weeks to go wide release to US & Canada
— Elon Musk (@elonmusk) November 12, 2022
The leak of a small part of the release note for the new FSD Beta V11 update confirms that the team has achieved a single stack.
“Enabled FSD Beta on highway. This unifies the vision and planning stack on and off-highway and replaces the legacy highway stack, which is over four years old. The legacy highway stack still relies on several single-camera and single-frame networks, FSD Beta's multi-camera video networks and next-gen planner, that allows for more complex agent interactions with less reliance on lanes, make way for adding more intelligent behaviors, smoother control and better decision making.
Improved Occupancy Network's recall for close by obstacles and precision in severe weather conditions with 4x increase in transformer spatial resolution, 20% increase in image featurizer capacity, improved side camera calibration, and 260k more video training clips (real-world and simulation).”
© 2022, Eva Fox | Tesmanian. All rights reserved.
_____________________________
We appreciate your readership! Please share your thoughts in the comment section below.
Article edited by @SmokeyShorts; follow him on Twitter