All of the Tesla fanboys saying you don't need LIDAR, RADAR, or Maps as a kind of "ground truth" and you can get by on video only should take this as a huge repudiation of brag by Musk.
Look, Radar, Lidar, and Maps aren't perfect. A map can be wrong, Radar can have false echos, Lidar can have issues in bad weather, etc. But if you're depending on visual recognition of one-way signs, having a potentially out-of-date backup map to check your prediction again, and err on the side of caution, is better than having no Map at all.
Ditto for radar and lidar. Radar enabled cars have collided with semis, remember, because they're aimed low (underneath the semi). But in the case of these planters, or road cones, or monorail poles, the radar is going to be a strong signal to the rest of the system that it isn't seeing something.
There's no way FSD should be allowed on public roads, it's not even CLOSE to ready. The other car companies are behaving far more responsibility: full sensor suite, multi-sensor fusion, comparison with detailed road maps, and limiting operation to safe areas, and expanding as safety is proven.
Who the hell cares if it's not FSD driving if its so dangerous? I'll take 75% FSD that only operates on highways and long commutes or road trips, over 100% FSD that has so many disengagements and "woah" moments as to be unusable.
> The thing about automotive radar, and this isn't just Tesla, is that stationary objects are normally excluded. Otherwise the system brakes for things like overhead signs that have high radar reflectivity. Stopped cars, concrete barriers, etc are all basically invisible to most automotive radar.
I believe most automotive radars in the last few years are FMCW exactly to accommodate adaptive cruise control.
TLDR: a Tesla can't identify a box in the road. IO can finally identify people, but it still doesn't do a good job of avoiding them.
Tesla Vision has no issues detecting white semis crossing your path. Vehicles with radar, on the other hand, struggle with discerning those from overhead bridges, so if one appears close to a bridge, you're SOL due to whitelisting.
Both of these statements are false. Tesla Vision still has trouble detecting white semis as of October 2022. There are no self-driving vehicles that use radar for navigation(you appear to be mixing up radar with LIDAR, which has range-sensing built in, and all of Tesla's competitors are able to tell trucks apart from bridges; truck identification failure is unique to Tesla), though many regular modern cars do use it for autobraking systems. As these systems are only intended for use at extremely short ranges directly in front of the vehicle, it's irrelevant whether the object detected is a bridge or a semi.
Tesla Vision in FSD is a much, much more developed version which has been excellent about detecting its environment, especially now with the new occupancy network. Its decision making needs work but you will notice, when watching all those videos, that detection of vehicles - even occluded ones - is not a problem at all.
This does not match reality. At all. Teslas still regularly swerve themselves across lanes of traffic and into oncoming traffic. In a brand new Tesla acquired by a co-worker several weeks ago, Tesla FSD could not identify cyclists on the road, failed to identify a number of pedestrians crossing at a crosswalk, did not successfully distinguish between semi trucks and the open sky, and only successfully identified about 1/2 of the other cars on the road with it. Maybe the super-duper secret version of Tesla Vision performs well, but the one actually available on Tesla vehicles right now performs worse than a drunk teenager.
Both Tesla and Karpathy himself have shown on multiple presentations that they focus on training unique/difficult situations because more data from perfect conditions is not useful to them anymore. They have shown exactly how they do it, and even showed the great infrastructure they've built for autolabeling.
This is demonstrably false; admission into the FSD program requires a safety score which cannot be achieved in areas with rough or steep roads, and is almost impossible to achieve in urban traffic, ergo, they are by definition not focusing on training unique/difficult situations. Moreover, as they still can't identify semi trucks, other cars, cyclists, or pedestrians with any reliability, the "great infrastructure" for "autolabeling" is basically just fraud.
But Tesla had RADAR until the chip shortage took it away. An now Tesla is removing ultrasonic sensors (USS).
--------
If the problem couldn't be solved with Vision + RADAR + USS, why do they think they can solve the problem with Vision only?
Its clear that they're just cost-cutting, IMO. If the problem isn't solved either way, might as well deliver a cheaper car to make, with the FSD at $12,000 instead of $5000. Just a simple money-making scheme at this point.
"A forward-facing radar with enhanced processing provides additional data about the world on a redundant wavelength that is able to see through heavy rain, fog, dust and even the car ahead."
So it's rather bizarre that you would speculate about how they're not using radar when they explicitly say they do.
You remember incorrectly. Tesla have never used LIDAR except for training on specific developer cars. Elon has said from day one LIDAR is an expensive waste because of the compute requirements and latency. They did remove the parking ultrasonic sensors, which at one point were used as a close proximity, low speed auxiliary data source in FSD.
The mandate of FSD has always been that humans get away with driving using only their vision so it must be possible for a computer to do the same. If you have used FSD lately it's largely there.
I'm sure that Tesla will have to eventually bring back radar into the stack. The problem with a vision system only is that it relies on visual clues that can be easily blinded by the conditions of the road. For instance a sun glare, high beams from traffic, water on the windshield, or simply a small depression on the road terrain.
These are the same things that can cause a traffic accident for a human with human eyes, so there's really no logical reason to assume that you can create a more performing and safer computer driver by emulating human vision only.
Radar solves all these problems by enhancing the perception of the car and creating redundancy so the car can cross-check its assumptions about the road, the terrain, and the obstacles with two different and independent perception systems.
I can't understand why they removed radar. You don't need to be an expert to understand why vision alone won't work.
This is anachronic. Musk was making the vision argument to claim that Teslas sold at the time (which lacked any kind of LIDAR or radar) had all of the hardware they needed to do FSD, which was sold as "just a few months away via software update". I believe that they still make this claim officially, since it's important to deny that they dif false advertising on this front.
As a Tesla owner with a model 3 that used to have radar until Tesla disabled it I find this argument funny.
Radar failed when the front got covered in ice, that's about it. With Tesla Vision (camera only), it no longer has any active sensors so it means it doesn't really work and is generally unavailable (as in, not even cruise control is available) or highly unpredictable in conditions such as:
- Rain
- Fog
- Snow
- Low sun
- When using windshield fluid
- Sometimes when driving next to a truck (not sure if tunnels are still an issue as well?)
- Probably more scenarios I forgot about...
Having tried both setup, it's pretty clear to me that the best real-life results are with multiple sensor technologies. I have very little faith Tesla succeeds as long as they are camera-only.
All Teslas are "vision-based" now. They no longer use the RADAR even in cars that have it.
This is probably so they can have only one software build rather than supporting one with and one without RADAR. And also so that they can say the RADAR isn't needed anyway.
I thought Tesla's claim was that the radar was not necessary for FSD on roads. But it's necessary for driving around a parking lot?
reply