Elon Musk set an aggressive deadline for Tesla to achieve full self-driving capability, but the electric automaker might have an ace up its sleeve that mostly went under the radar: project ‘Dojo’.
This weekend, Musk hinted that it could make the difference.
During Tesla’s Autonomy Day earlier this year, Musk and other Tesla executives gave presentations about what the company is doing to try to achieve full self-driving capability by the end of next year.
While most people were focused on the unveiling of Tesla’s new HW3 ‘Full Self-Driving Computer’, which was being explained for the first time and is now installed in all new Tesla vehicles, there was a brief mention of another computer, the Dojo computer, that Tesla is working on and it could be a game-changer
Last weekend, Musk was asked about the secret project and while the CEO didn’t reveal anything new, he did hint that it could make the difference:
Will Dojo to be the difference
— Elon Musk (@elonmusk) August 4, 2019
During Autonomy Day, Musk briefly mentioned the project ‘Dojo’:
“We do have a major program at Tesla which we don’t have enough time to talk about today called “Dojo”. That’s a super powerful training computer. The goal of Dojo will be to be able to take in vast amounts of data and train at a video level and do unsupervised massive training of vast amounts of video with the Dojo program – or Dojo computer.”
Currently, labeling visual data is a difficult task and most neural net systems are using image frame processing to feed its network, but it sounds like Dojo would instead be able to handle videos on its own, which would give a lot more context to the data and could train the network in different and more complexed contexts faster.
More information about Tesla’s ‘project Dojo’ is expected to be released at Tesla’s next update about its road to autonomy early next year.
I can see project Dojo making the difference for Tesla to achieve full self-driving on a more aggressive timeline than competitors.
If you look at the progress of Autopilot in recent years, it’s constant enough that I can see it reaching a useful level on surface streets within the next year or so, but that doesn’t mean that it’s ready to be autonomous and unsupervised.
While it could be useful as a driver assist system, it wouldn’t be able to handle all the weird situations you can encounter on city streets.
However, if you combine the incredible number of videos and data that Tesla’s fleet of hundreds of thousands of vehicles can generate with a computer and software capable of interpreting the videos and using the data to train its self-driving neural network, you can accelerate the pace of progress greatly.
It will be able to go through the almost infinite number of corner cases that we encounter in vehicles much quicker and also adapt faster as more data is coming in.
What do you think? Let us know in the comment section below.
Subscribe to Electrek on YouTube for exclusive videos and subscribe to the podcast.