It's almost as if removing LiDAR cameras in favor of traditional cameras "because they're stupid" was the most idiotic ketamine-fueled decision a company could ever make. Why anyone believes Teslas can drive themselves in uncertain situations is beyond me.
this post was submitted on 30 Jun 2025
156 points (98.8% liked)
Autonomous Vehicles
119 readers
1 users here now
Autonomous Vehicles is a community dedicated to the news, discussion and exploration of autonomous vehicles and how we as a society, will embrace this futurology today!
founded 1 year ago
MODERATORS
The even crazier part is that with all of their advancements in software, they'd probably legitimately have fsd launched and running well by now, given how much driving data they can ingest at will, if they'd just included a few hundred dollars worth of low resolution lidar and ultrasonics. Irrc they stated they were having issues with chain of trust among the sensors, but I'm not sure I believe that.
Well, do we know what the blockers are for Tesla?
I feel like when I watch videos of FSD on cars, the representation of the world on the screen is rather good.
Now given this datapoint of me watching maybe 30minutes of video in total, is the issue in:
a) creating the distance to obstacles in the surroundings from cameras or in:
b) reading street signs, road markings, stop lights etc, or in:
c) doing the right thing, given a correct set of data about the surroundings?
Lidar / Radar / Sonar would only help for a).
Or is it combination of all of them, and the (relatively) cheap sensor would at least eliminate a), so one could focus on b and c?
The blockers for Tesla are that it's processing a 2D input in order to navigate 3D space. They use some ai trickery to make virtual anchor points using image stills and points of time to work around this and get back to a 3D space but the auto industry at large (not me) has collectively agreed this cannot overcome numerous serious challenges in realistic applications (the one people may be most familiar with is Mark Rober's test where the tesla just drives right into a wall painted to look like the road Wiley Coyote style, but this has real world analogs such as complex weather). Lidar and ultrasonics integrated into the chain of trust can (and already do for most adas systems) mitigate a significant portion of the risk this issue causes (Volvo has shown even low resolution "cheap" Lidar sensors without 360 degree coverage can offer most of these benefits). To be honest I'm not certain that the addition would fix everything, perhaps the engineering obstacles really were insurmountable... but from what I hear from the industry at large, my friends in the space, and my own common sense; I don't see how a wholly 2D implementation relying on only camera input can be anything but an insurmountable engineering challenge to overcome in order to produce the final minimal viable product. So from my understanding it'd be like being told you have to use water and only water as your hydraulic fluid, or that you can only use a heat lamp to cook for your restaurant. It's just legitimately unsuitable for the purpose despite giving off the guise of doing the same work.
Also and I'd forgotten to mention, but what you see in the on-screen representation is entirely divorced from the actual stack doing your driving. They're basically running a small video game using the virtual world map they build and rendering in assets and such from there. It's meant to give you a reasonable look into what the car sees and might do, but they've confirmed that it is in no way tied to the underlying neural decision network.
But that's exactly the point.
if the virtual map they're building from cameras is complete, correct and stable (and presumably some other criteria that I didn't think of from the top of my head), then the cameras would be sufficient.
The underlying neural decision network can still fuck things up from a correct virtual world map.
Now, how good is the virtual world map in real world conditions?
You're missing the point though maybe? You can't take data, run it through what is essentially lossy compression, and then get the same data back out. Best you can do is a facsimile of it that suffers in some regard.
To be fair, Elon probably wasn’t taking ketamine back when he made that decision. He made that decision all by himself :)
I hope someone posts Elmos butthurt tweets about this, so I don't have to visit his cancer site.
Reacting to the news:
"They cost Waymo money"
The typical high-IQ reaction to being abandoned as de-facto standard in a new realm of tech competition.
Just an fyi, you can replace any garbage x url with xcancel.com to prevent having to go to his site. In a perfect world we wouldn’t need it, but I wanted to make you aware of it
All because Elon was too cheap to add lidar
The CEO of Waymo has also not gone around throwing out Nazi salutes like Elon Musk.
Everyone here should check out comma.ai’s open pilot. It is similar to Teslas self driving system, but it’s open source. Perfect Elon replacement. And since they tap into the sensors on your existing car, I imagine they could leverage lidar if more cars had those sensors.
They sell hardware that is designed to run open pilot. Last I checked it was $1k, which feels pretty reasonable compared to what Elon is charging.
Similar how? Tesla has over 2 billion self driving miles, comma's own website says it's 100 million. Where's the comparison?
I know they both have semi autonomous systems that don’t rely on lidar. The navigate on autopilot presentation comma gave years ago seemed very Tesla. Comma’s self driving capabilities seem to lag a couple years behind Tesla.
I like George Hotz and I know the Lemmy community prefers open source tech, so I wanted to bring some awareness to an alternative you can add to your current car. But I know very little about self driving tech. Maybe I’m wrong in calling them similar.
It's something worth a post of its own, but I wouldn't put it down as a competitor/replacement.
Yeah but they're not wrong. I prefer Waymo to Uber/Lyft. Slightly pricier but there's no tipping so it balances out.
They're also more consistent. I never have to worry about Waymo parking in front of the wrong building, then fucking off with 2 minutes left on the timer. They park in the right spot every time. They sometimes make stupid mistakes like not using the middle lane to turn left, but overall they drive more safely than the average human driver. The radar sees everything; one time a Waymo avoided an accident that I would have never seen coming. Speaking of which, I like how you can see what it sees on the touchscreen for peace of mind.
Overall, I'm okay with using a Google service every once in a blue moon to have a much better taxi cab experience.
Womp womp.
As much as I don't like self-driving cars, I must admit that Waymo indeed seems better.
However it would be interesting to see how this self-driving tech could be implemented in personal vehicles.
Tesla's "full self driving" is not a possibility for people without a driving license. AFAIK you already need to drive in order to have "full self driving", while Waymo can just act as a regular taxi but without a full time driver. And they don't just rely on cameras.
As someone without a driver's license the difference between both is enormous, but for a car company like Ford looking for real self driving, you also want something proved to be actually working without the need to be actively monitored by a human.
As much as I don't like self-driving cars
Why?
Because they still have all the disadvantages of cars and they don't really bring anything of value. A car is dangerous for the people around not in a car, inefficient at transporting people, takes valuable space, and pollutes the air with tire and brake dust particles. They are also noisy. Electric cars won't change that.
In the case of Tesla, FSD is just a gadget for car owners that somehow don't want to drive cars.
In the case of Waymo, they are not a solution to relieve transit in busy cities. There's even more cars now, but without a driver and sometimes they get stuck and still need to call for help.
If you want me to get mildly excited for an autonomous vehicle, show me a train or a metro. Otherwise, if people want to get to places without driving, we already have solutions for that, but apparently it's better to ask everyone to buy a private autonomous vehicle that requires a driving license and won't work in the country side, rather than speak of the horrors of the T word.
EDIT: I know which community I'm in and didn't want to get into this but I was asked. Self driving cars are also wasting energy because of the computing power needed to process all the stuff that a human brain was already processing "for free". Now, instead of having billions of cars that require significantly more energy than transit to move around one or maybe two humans, we will have billions of cars that requires even more energy to plot courses and drive the individual cars around.
pollutes the air with tire and brake dust particles. They are also noisy. Electric cars won't change that
Electric vehicles release fewer brake dust particles, as they can use regenerative braking. Brake pads are really only needed when an unexpectedly sudden stop is needed (emergencies).
EVs are also quieter, as others have already pointed out.
I don't want to mention the other major output of regular cars versus electric cars, either.
(Arguments about just shifting the location of burning aside, we are specifically talking about the effect on nearby people)
Indeed, electric cars may produce less brake dust particles than an ICE. However they are also heavier than ICE and the tires will shed more particles over time.
And electric cars are also somewhat less noisy but they cannot beat the laws of physics. You can still hear an electric car coming because of the noise tires make on the roads over certain speeds. The higher the speed, the noisier it is. Even if all the cars on a multi lane highway were all electric, it would still be noisy and difficult to have a conversation nearby. A six lane highway filled with electric cars will still need a sound wall.
There are also regulations to have electric cars make artificial noise to be "safe" at low speed, and some sound like they are constantly honking at low volume.
Like, yes, those two points are marginally better with electric cars, but it's not much. And it doesn't cancel all the other negative issues with cars.
Playing devil’s advocate here: trains and busses are loud af.
They are also noisy. Electric cars won't change that.
But they do. Electric vehicles are much much quieter than the average ICE car. The gap is smaller for newer ICE vehicles, but it's still there.
In the case of Waymo, they are not a solution to relieve transit in busy cities.
Agreed, but it doesn't mean they don't have their place. In an ideal world, if would result in more car sharing. In reality, they provide the ability to mobilise the disabled as well as people that can't afford cars in places with poor transport links.
I hope we eventually get full self driving cars, I'm so sick of driving
I guess I can understand if you live in a place without public transit, but if this is the case, it's also remote enough so that "self driving cars" may not get there for some time. Waymo is using mapping tech and is not going out of a city any time soon.
And if your self driving vehicle is restricted to a single city or area, it becomes a matter of that city asking everyone to buy or use a self driving car because they can't finance an adequate and efficient public transit system.
It's just a gadget.
EDIT: I don't have car and already don't drive anywhere. Just like millions of other people in the world. But apparently, if you don't want to drive, you need a car that drives for you. There is simply no other solution.