Search results

Do we know what Tesla has in their lab? I would expect Tesla to continue to improve their "FSD" too, so they may not fall so far behind the competition by the time those are deployed in the US.

Not that I expect this to happen, but I thought the initial idea of the "Tesla Network" was to have owners run their private cars as robotaxies. If that's the case, I would presume the owners would do the maintenance.

It is an interesting take on Tesla's "business model" - in that they do state a goal of robo-taxies, but unlike every robo-taxi company, are already getting revenue for the unfinished product. If they do get to robo-taxies, then that's even more revenue. My uninformed prediction is Tesla won't...

There are so many use cases for a humanoid robot, basically any physical job a human can do. The issue is making one actually capable and smart enough to do anything useful at a reasonable price. I think that's what was missing from Asimo. I doubt Optimus will be anywhere near that. In 5...

For what it's worth, Dr. Know-It-All speculated Optimus is running on a HW4 chip in his latest video. He noted the board only had one chip, instead of the usual two, and when he asked a Tesla engineer if they were using a new chip for it, they gave a somewhat surprised and guilty look, but no...

I sold my 2018 Model 3, and was a bit disappointed my new Model Y didn't come with the nicely flocked rear center console tray. I already ordered a front console tray I liked on Amazon, so I didn't want to buy the set from the Tesla shop. So I designed one, that has the option to fit card...

I suppose in a case like that, you could have excess cars from outside of the area drive into the evacuation zone to alleviate the shortage. Imagine a swarm of a million robotaxies driving in. That said, what will it be like if you want to go camping and don't want to unload the car...

Any chance they can just stop selling Teslas in Florida?

Why do you think they're making the Teslabot? Coupled with Neuralink, he's going to upload his consciousness into Dojo with the robots as his bodies, like Ultron.

If we're talking about Atlas, we should remember Atlas is hydraulically powered. That means a hydrolic pump, and a motor big enough to power it along with the batteries. Atlas is also incredibly strong. Strong enough to do somersaults. As long as TeslaBot doesn't need to do parkour, I think...

I've seen Honda's Asimo, and Atlas (along with other bi-pedal robots) as the DARPA Robotics Challenge a few years ago. I also saw very human hand like prosthetics while there. I don't see why building a humanoid robot would be impossible. My Neato robotic vacuum has a rudimentary LIDAR and can...

Looking at Boston Dynamics, I think their problem is they're essentially an R&D company. They're not set up for mass manufacturing, nor do they seem all that interested in it. You can already purchase Chinese made versions of "Spot" for far less than they're selling them for.

Anybody have any thoughts on how persistent those spacial RNN maps are? I'd be surprised if they have a long term persistence, but that would be really useful for oft travelled paths.

I look forward to being entertained by headlines about "Florida Car."

Might depend on the state. Looking at Nevada's DMV form for public use autonomous vehicles, it looks to be "self-certification." Just check off the box that says "Is capable of operating in compliance with all applicable motor vehicle laws and traffic laws of this State" along with a bunch of...

My guess is whatever gets implemented will make stupid mistakes any experienced human driver wouldn't make leading to stupid accidents (like running into parked/overturned trucks/uncommon objects), but also avoid a lot of common accidents humans make due to the 360° + 100% paying attention...

I suppose the question I have is, will the car be able to get to better than average human safety with the current sensor suite? Clearly, we're not getting Lidar in our Teslas.

Isn't there also an additional distance estimation based on visual segmentation? I wonder how well that does past 62 meters.

You're thinking of one of her identical twin sisters, "Zoot" or "Dingo."

Because of the naps! ;)

Yeah, not mentioning Waymo seems to be the minority. Instead of searching "Waymo" which will invariably bring up "Waymo" I searched for "self driving car cones" and 5 of 6 mentioned "Waymo" in the headline, with just the La Crosse Tribune not.

Back to the original topic: Do we know if this new vision only / non-Radar software will apply to HW 2.x, or just HW3? I'd guess if they're using Pseudo Lidar to gauge distance, it'll have to be HW3 only.

Klaatu barada nikto! You're assuming that if such a patent existed, VAG offered to license it out. Very possible they would hold on to it for competitive reasons.

This is what makes me pessimistic about Tesla getting to Level 5 anytime in the next few years. They're still figuring out perception. I can't imagine they're very far along on driving policy and planning.

Possible VAG has a better solution they patented, but haven't licensed out to Tesla.

I figured this would be a benefit of removing Radar. My question is, are they removing Radar input from AutoPilot, or just Streets / FSD Beta? Autopilot on the highway is where I get the (presumably) Radar based phantom braking. This won't help with the phantom braking events due to incorrect...

To be fair, Musk is the Technoking of Tesla, though I don't believe he's the Chief Engineer, like at SpaceX.

Might be getting ahead of ourselves.

Same behavior at the same stretch of the 405 back from April 2019, as per this video. It will continue to signal, but won't change lanes with the solid line. Yep, confusing, and not good.

Automation / astrogation aren't the only challenges of getting people to Mars. You need to invent more efficient radiation shielding, possibly develop nuclear propulsion, develop on orbit refueling, develop in situ resource utilization capable of refueling a return mission, send said refueling...

Let's just say even money if we're racing FSD with boots on the Moon, instead of Mars. I reserve the right to change my mind when I see the next FSD beta update in "two weeks."

For some context, the current US timeline is 2024 for boots on the Moon. Everyone but SpaceX and NASA (officially) believes is overly optimistic. Mars is in the mid 2030s at the earliest according to NASA. Not sure SpaceX's timeline.

I disagree. It seems clear there was a re-write to the perception system, as it's now doing 360 video, which was not the case before. This isn't being used in the current AP, as far as I know.

I'd expect to be blown away and also underwhelmed. Blown away that a car can do any sort of self-driving, but underwhelmed at how well it does compared to a human driver. Same thing with Advanced Summon. If you told me 10 years ago I'd have a car that can drive around in a parking lot, I'd be...

Knightshade is talking about what Tesla is currently producing, and you are talking about what is currently in the existing Tesla fleet, which must also be capable of FSD at some point according to Tesla.

To me, "co-pilot" is worse, as a co-pilot is capable of fully operating a vehicle when the pilot goes to eat, sleep or go to the bathroom. Autopilot (both in a plane and Teslas) require the vehicle operator to be at the controls.

I predict FSD wide release will be relatively safe, but overly cautious as a driver. It'll probably miss a lot of turns because it can't get into the proper lane in time, requiring re-routing. It'll probably drive slower than advisable, and be fairly annoying.

Cost and precision. When Tesla originally launched Autopilot 2 / Hardware 2, Lidar systems cost about the same as a car, so would be prohibitively expensive. Radar is currently used for distance, but what they fielded / available for the price at the time doesn't have nearly the resolution...

Pretty cool. I wonder in the context of Tesla, if the FSD computer is capable of handling both. We've seen Pseudo-Lidar from the Karpathy talk, which must be monocular for the side cameras at least. I'm also unclear if the standard object/image recognition used is considered panoptic...

I'd have to agree with this. Of course if they find Pseudo-Lidar / cameras still can't cut it, then perhaps. While they're at it, might as well put in cross traffic cameras. Link still doesn't work. Why not give the authors and article title so we can Google it?

In the case of having someone in the driver's seat, I wouldn't be surprised if they swerved to avoid a deer.

The problem is fast you want the wipers going is subjective. I'm guessing it's what may be optimal for the Autopilot cameras to see. Sort of a problem for us manual-pilots.

I don't think the PDAF part of the camera can do per pixel depth. It sounds to me like PDAF (phase detection auto-focus) is just what my SLRs have been doing for years. It can determine which direction and approximately how much focus needs to be adjust to get whatever is under the auto-focus...

Pseudo-Lidar may do it, but I'm not sure how accurate it is at distance, though having cm level precision probably isn't important at greater than 10 meters. Skip to 22:13 in Karpathy's talk to see it. I, too, worry about losing that functionality. I'm guessing Tesla couldn't get sensor...

I'm pretty sure I've seen roadrunners enter those tunnels without a problem. The trick is to meep-meep first. In all seriously, I would think pseudo-Lidar would be able to tell via parallax it's a solid wall...maybe.

I've been under the impression TACC hasn't been confirmed to be on, but a couple of articles interpreted the call to say that. Also, for what it's worth, headline writers are usually the editors, so it's not always the article writer's fault.

Probably read the inaccurate Electrek article. If only Tesla had some kind of department that could respond to media inquiries to correct this sort of thing.

Pretty sure everyone agrees cars bursting into flames is a bad thing. It's just hard to sustain 1000+ posts of everyone agreeing.

For what it's worth, all modern cars have collapsible steering columns, which act as "crumple zones" in a crash. So someone hitting that in addition to an airbag might still be capable of moving around the cabin under adrenaline. For all we know, he could have been a big yoga fanatic and had...

On the investment call, they mentioned the steering wheel was deformed, so there was likely someone in the driver's seat upon impact. Also, all seatbelts were unbuckled. They weren't able to recover the SD card data.