It depends. If they're not recording the right things, they have the same 0 bytes of data as everyone else.
There's a not insignificant chance that their sensor data is insufficient to perform the tasks, and the recorded data won't be usable for training a system with whatever sensor is needed; in which case, no advantage for logging customer movements for N years.
Why isn't every car body made from carbon fiber? It's lighter and sturdier that alternatives. It's also more expensive so you can put it in $60k car but not in $25k car.
Radar and ultrasonic sensors are cheap, low power and plentiful.
Putting several lidars on a car would be too expensive, would use too much power (less range) and there isn't a supply chain to make 2 million a year just for Tesla (500k cars x 4 lidars per car).
It's just isn't possible to put a lidar today on a mass market car and Tesla wants to solve self-driving before lidar becomes cheap enough.
On the other hand, Waymo et. al. are not risking anything by using lidars. If they turn out to be redundant, it should be easy to remove them cheaply.
BTW, all companies that can, use lidars. Even MobileEye says that their L4 system will use lidars and they've been working on a camera based system for 2 decades. Tesla doesn't use lidars because they can't.
1. Lidar is expensive for everyone. It’s not like Tesla will pay more than other manufacturers. They’d prefer not to use Lidar, but if cameras just don’t cut it, I don't think they would die on that hill.
2. I’m sure there is some fine print around the current FSD being for camera based sensors only. They can easily say the car tops off at L4 for cameras, and L5 needs Lidar, if we come to that bridge
In self-driving, you basically have 3 parts - perception, prediction and planning. Perception is the easy part (if you have lidars). Tesla and AFAIK others rely heavily on NNs for perception. The real challenge is prediction and planning. To what extent are NNs involved here in Tesla's system is unknown, probably not that much. Overall, I don't think that Tesla has a fundamentally different approach than others.
> all it takes for Tesla’s NN approach to pay off is one leap in explainability.
It's not vision and disengagement-led training that's conferring the scale advantages you're talking about, because other AV companies actually do train on and attempt to drive in various kinds of weather, and safety drivers do the same thing normal drivers do except that they're more trained. The difference is the approach Tesla has to safety (vs the others). Which is to say: Tesla doesn't worry about training their supervising drivers, at the risk of those folks' lives.
Tesla also doesn't seem to worry about _proving_ the safety of their systems: explainability isn't simply going to be a matter of a new algorithmic breakthrough. How do you prove that your software is safe if you don't sample heavily from different operating domains you want your car to work in, and then either simulate on those miles or actually test your car? Before letting a driver behind the wheel of an autopilot, companies like Cruise/Waymo/Aurora all do manual driving first in a new domain, and then move to safety drivers (even 2 to start, for extra safety). This is not because their software is incapable of doing the same things as Tesla, but because it's very worthwhile to sample disengage rates and assess safety criticality to various issues before allowing the car to act autonomously. And even then there are safety nets in place to further mitigate risk. (You're probably right that L5 is a long ways off, but this is how we'll get to L4.)
Of course it's more "scalable" to avoid all this and put the responsibility of risk on users who, as your system improves, become less and less attentive. And so this is what Tesla does. I suspect solving the long tail of issues is going to be more of a challenge for Tesla because at least in my experience it's easier to {convert a naive rules/simple model --> complex model} than to {debug the complex model --> handle failure modes with logic}.
Tesla just decided to jump into the deep end from the beginning.
I don’t have much experience in SDCs, but did some computer vision/ robotics work previously.
I truly think that Tesla’s approach, if it works, will win in the long run (ie - next 15-20 years, I think the breakthrough for true L5 isn’t here yet - but will be in improvements in one-shot learning over the next decade).
Winter driving is a completely different beast than driving out in Arizona. You have to account for snow banks, slipperiness, lidar obstruction, and control itself - which is much harder.
How many thousands of miles of snow driving data does Tesla possess, compared to the #2 player?
Tesla’s approach of vision based, AP disengagement-led training will scale better, precisely due to what you’re looking at in this article - yeah, this is an unimpressive demo - but winter driving, rainy driving, dirt road driving are all general cases for Tesla, while they’re special cases for other companies (from what’s been revealed; Comma might be an exception).
Most people drive in the place they live, and drive on the highway to visit their friends and parents. It’s easy to be deceived into thinking driving is a small problem space of stop signs and lane changes. But even within just North America, there is enormous variety.
There is massive risk involved - these folks are pushing NNs further than any application I’ve seen, but with the last few releases, it suddenly doesn’t look all that crazy.
Secondly, combining rules based systems and neural networks is difficult. You end up with the N * M problem, and inevitably end up missing cases. It makes sense to do the absolute basic safety features as rules, and Tesla might be doing this, but all it takes for Tesla’s NN approach to pay off is one leap in explainability. There is an entire industry working on this - and Tesla will be the single biggest benefactor.
I don’t know who will win L5 - but I’m putting my bets on Tesla for L4.