In the piece — titled “Can You Fool a Self Driving Car?” — Rober found that a Tesla car on Autopilot was fooled by a Wile E. Coyote-style wall painted to look like the road ahead of it, with the electric vehicle plowing right through it instead of stopping.
The footage was damning enough, with slow-motion clips showing the car not only crashing through the styrofoam wall but also a mannequin of a child. The Tesla was also fooled by simulated rain and fog.
Why would a car that expensive not have a LiDAR sensor?
Because commonly they use radar instead, the modern sensors that are also used for adaptive cruise control even have heaters to defrost the sensor housing in winter
Because Musk insists that cameras are better and that LiDAR is flawed
That’s not really true.
He use lidar in SpaceX because he knows it’s the right tool for their specific job.
His stance is it’s not that cameras are better, but that cameras have to be so good for a truly AV that putting effort into both means you’re not going to make your cameras good enough to do it and rely on lidar instead. That and cost.
If the car can’t process and understand the world via cameras, it’s doomed to fail at a mass scale anyway.
It might be a wrong stance, but it’s not that lidar is flawed.
Tesla even uses lidar to ground truth their cameras
Edit: just adding a late example - Waymo, Cruise, and probably everyone out there still use humans to tell the car what to do if it gets stuck. I even bet Tesla will if they ever launch a robotaxi as they need a way to somehow help the car if it gets stuck. When we see these failures with Waymo and Cruise, it’s less “is something there” and more “I don’t understand this situation”. The understanding comes from vision. Lidar just gives the something is there, but it isn’t solving their problem.
I think the bigger issue is that he is saying redundancy is not important. He thinks cameras could be good enough, well fine, but the failure results in loss of life so build in redundancy: lidar, radar, anything to failover. The fact that cutting costs OR having a belief that one system is good enough is despicable.
Ya, no redundancy is a problem for sure.
Light aren’t radar systems don’t work internationally because they’re functionally band in many asian and european countries. Instead of making one system that was almost complete finished, they went all camera and now none of it works right.
You can get a Tesla for $42,000… They aren’t that expensive.
With that said, they’ve really cheaped out and even removed the cheaper radar sensors they used to have because Elon wanted to save a buck and really thinks all you need is cameras because he’s an idiot.
Hell, they don’t even have radar anymore, despite even a lot of low end cars having that.
Technically cost savings, but it seems mostly about stubborn insistence on cameras being enough.
They cost hundreds of dollars!!
/$
Cameras are cheaper…that’s it
It’s a car that’s at least £10k
Cost cutting. Lidar is cheaper now but was relative expensive and increased tech debt and maintenance. Also he legit thought that “human see good - then car see good too”. Tesla is being led by a literal idiot.
Read about this somewhere. Iirc, Elon felt cameras were better than LiDAR at a time when that was kinda true, but the technology improved considerably in the interim and he pridefully refuses to admit he needs to adapt. [Edit: I had hastily read the referenced article and am incorrect here; link to accurate statements is linked in a reply below.]
that was never true
Found the article! I had breezed through the thing. I was incorrect about the LiDAR/camera thing. Instead it was: ‘Elon even admitted that “very high-resolution radars would be better than pure vision”, but he claimed that “such a radar does not exist”’
He, of course was incorrect and proven incorrect, but ‘the problem is that Musk has taken such a strong stance against [LiDARs] for so long that now that they have improved immensely and reduced in prices, he still can’t admit that he was wrong and use them.’
Lol just like his Nazi forefathers in WWII who refused to believe (more than once!) the British had the advanced radar that they actually did have.
That was the story but it was supply chain issues that lead him to that conclusion. Same reason why lumbar controls were removed from passenger seats.
I don’t even understand that logic. Use both. Even if one is significantly better than the other, they each have different weaknesses and can mitigate for each other.
It was always just to save money and pad the profit margins
And to make him think he’s a smart boy.
A LiDAR sensor couldn’t add more than a few hundred to a car, surely
They ditched radar at a time when radar only added probably about $50 a car according to some estimates.
It may technically get a smidge more profitable, but it almost seems like it’s more about hubris around tech shouldn’t need more than a human to do as well. Which even if it were true, is a stupid stance to take when in that scenario you could have better than human senses.
@jj4211 @Flax_vert It sounds to me, from various sources with bits of the picture, that active ranging was dropped during the pandemic because supply was harder and it would have slowed production, with potential financial consequences for the brand (due to Elmo’s customary over-leveraging backed by his boastful BS). And then tried to claim that that was intentional, even planned, and tried to make up the loss with some kind of software magic that still can’t violate the laws of physics.
And thats a few hundred less profit, so we cant have that.
He didn’t think they were better. He thought Tesla could get away without the more expensive lidar. Basically “humans can drive with just vision, that should be enough for an autonomous vehicle also.” Basically he did it because lidar is more expensive.
Even if humans can drive with just vision:
Human vision is great, human attention is not and neither is their reaction time. Computers are 100x better at both of those. If you throw lidar into the mix, then a car’s vision is now much better than a humans.
IMHO self driving cars have to be statistically 10x better than humans to be widely implemented. If it passes that threshold them I’m fine with them.
I didn’t think it was about the cost. I think he just likes to be contrarian because he thinks it makes him seem smart. He then needs to stick by his stupid decisions.
I’m assuming it’s a cost because it makes sense to me. His goal was to build full-self-driving (FSD) into ever car and sell the service as a subscription.
If you add another $500 in components then that’s a lot of cost (probably a lot cheaper today but this was 10 years ago). Cameras are cheap and can be spread around the car with additional non-FSD benefits where as lidar has much fewer uses when the cost is not covered. I think he used his “first-principles” argument as a justification to the engineers as another way for him to say “I don’t want to pay for lidar, make it work with the cheap cameras.”
Why else would management take off the table an obviously extremely useful safety tool?
Couldn’t he just use both… Like LiDAR as a contingency
Every LiDAR system must use at least both. LiDAR can’t tell you about lane markings, what’s on signs, and state of traffic lights.
But absolutely, you could have multiple sensing technologies and have access to the best of all worlds.
I added a correction in another reply. Basically he stubbornly refuses to believe a powerful enough LiDAR exists. So I suppose he is all-in on “LieDAR” technology instead (yes, I kinda feel bad about this pun too)
He could. In fact Waymos, for instance, do and are fully autonomous commercial taxis while Tesla are still 2 years out from full self driving for the tenth year in a row
Because Tesla makes money, with the byproduct of cars.
There was a comedy channel on Youtube aeons ago that would do “if x were honest” videos. Their slogan for Valve was “We used to make games. Now we make money.”
Honest Ads is still around, they’ve just moved off the Cracked channel like how PitchMeetings moved off the ScreenRant channel.
The supplier he was using couldn’t supply lidar fast enough, and it was at risk of slowing his manufacturing.
So he worked in a way to not need it, and tell everyone this solution was superior.