Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Optimus already sorting objects autonomously | TMC Podcast #51

This site may earn commission on affiliate links.

Apparently Optimus' neural network is trained fully end-to-end and it can now sort objects autonomously. We discuss this and more!

Topics-
0:00 Stream begins
0:15 Intro
1:06 New Optimus video
24:44 Doug's road trip - FSD and Starship
54:54 Rove's 40-station EV charging center
1:06:14 FSD price reduced
1:10:12 Tesla FSD - A Five Year Perspective
1:18:00 Referral Program updates

Link to the article "Tesla FSD - A Five Year Perspective" - Tesla FSD - A Five Year Perspective

Co-hosts-
Louis: @nebusoft
Mike: @SteelClouds
Doug: @doug
Seb: @Seb P85D

Producers-
Adam: @ElectricAve84
James: @scrapps
Daniel: @danny

Social-
Twitter: https://twitter.com/TeslaMotorsClub
Facebook: Log into Facebook
Instagram: Tesla Motors Club (@teslamotorsclubofficial) • Instagram photos and videos
YouTube: https://www.youtube.com/teslamotorsclub

Audio versions-
Apple Podcasts: ‎‎Tesla Motors Club Podcast on Apple Podcasts
Spotify: Tesla Motors Club Podcast
Amazon Music: https://music.amazon.com/podcasts/38bacc87-f8b7-4f5c-aa64-2db865214942
TuneIn: Listen to Tesla Motors Club Podcast on TuneIn
RSS: https://feeds.buzzsprout.com/1950101.rss

Support us-
PayPal: PayPal.Me
TeslaMotorsClub website: Account upgrades
20230924_132631_0000.png
 
Last edited:
Comparing the relatively low-key skills that Optimus has shown off so far to other "research" isn't really a valid comparison. Everyone else is really just doing it as a research project. If they're lucky, they'll finally get one working example of a better-than-Optimus robot engineered, and it will be not be reproducible at scale for reasonable cost. Even if they get there, they then have to re-engineer everything to scale, and it will probably turn out that a lot of the decisions made in research don't pan out at scale.

What Telsa's doing with Optimus is going in from the start with a goal of being able to cheaply manufacture this at scale. It's a whole different ballgame. Every time they succeed at some small goal like sorting objects, they're not competing with some research project that did it in a lab once a decade ago. They're in an empty playing field, where they're arguably the only company that will be able to mass-produce what they're showing off at a reasonable price.
I was seeing Spot robots in the wild many years ago, bought by companies in my industry who use them for inspections in dangerous industrial environments.

Boston Dynamics is now owned by Hyundai btw
 
Yeah, Spot actually is pretty cool, but they're not humanoid, and based on some quick googling: they've sold on the order of 1K of them, and they cost ~$75K. I don't think it will ever be comparable to Optimus potential, but I guess you never know!
 
Yeah, Spot actually is pretty cool, but they're not humanoid, and based on some quick googling: they've sold on the order of 1K of them, and they cost ~$75K. I don't think it will ever be comparable to Optimus potential, but I guess you never know!
Makes you wonder how a 5'8 humanoid robot could ever be cheaper than a little dogbot, but I guess that's volume for you eh
 
I just realized that if/when Optimus starts assembling the new Model 3 or unboxed, they may want to have more than a 1:1 replacement of humans. Initially as a way to compensate for any initial slower movements, but let's go with a couple for a moment. Or what about 3:1 etc... What's the limit besides energy and initial cost?

It's just a crazy what-if... don't freak out. We've seen whole cars get picked up and moved around up and down right?

Recalling the diagram with 4 (humans/robots) around an unboxed section, I'm thinking it could be that plus all the one's behind them in line, as fast as the materials could be delivered and placed correctly. Could the robots be combined to form a chain or just loop like ants? Or placed on a track that moves quickly yet unsafe for humans - like up above? Take this all the way back to unloading the truck or the last process end to end? Pretty much everything could change with material flow through a factory - clearly depending upon the type of processes and existing equipment at first. This may have a better fit in other Assy type factories that I have in mind.

Why couldn't Optimus strap on (snap into) roller blades with 90% energy efficiency, and work between 2 adjacent operations to reduce load/unload operations along with storage? Imagine a Hyper-balanced Line where the robots fill in the gaps and stores are eliminated. Could 2 carry a casting or some ergo type carrier and eliminate 2 other XYZ robots? Or just grab the thing right out of the mold! This approach would split operations differently. And yes, robots might be flowing through the factory.

What's this have to do with investments? It's the Margins. Shop Floor Control could be 100% integrated with the Robots, not a separate database or procedure. Basically, Throughput Time (TPT) drops when the river of materials start to flow more smoothly. Tesla could literally flood the market with EVs and more. And hey, I wonder how long before they could throw and catch parts? 5 yrs? 10 yrs?