Imagine naming a feature “Full Self-Driving,” and yet you can’t take your attention away from the road and must be ready to take over at a moment’s notice.
This is on par for Elon’s entire career. He loves claiming success and taking credit for things he either didn’t accomplish himself, or things he hasn’t accomplished yet.
I remember reading a post that claimed that Tesla’s safety rating was given to them because a bunch of their crashes were determined to be human error - because the self-driving feature would automatically disconnect if it faced a crash it couldn’t avoid.
Fairly certain the statistic requires fsd to have been disabled for 10s before or is counted as human-caused
The issue is a bit muddied by the fact that hitting the brake or the accelerator will deactivate it, and people will usually hit one of those if they believe that they are going to crash.
I feel like even with fully autonomous cars, there’s going to be laws about how the main driver should always be alerted. This would be the case unless our cars are their own independent drivers like a cab.
Are there any truly autonomous machines which don’t require a human to monitor?
Lots. Toasters, refrigerators, robot vacuums, thermostats, smart home lights, etc.
The reason why self-driving cars are extra tricky is both because they have a much more complex task and the negative consequences are sky high. If a robot vacuum screws up, it’s not a big deal. This is why it’s totally irresponsible to advertise something as having “full” autonomy when the stakes are so high.
Those are mostly automatic, not autonomous.
It’s just a driving assistant, like in any other car. As far as I know, currently Mercedes is the only one who implemented autonomous driving, and even that one is limited to some specific areas. But at least that one is real. So much, that legally Mercedes (the company) is considered to be the driver of such cars, in case anything happens on the roads.
Depends on your definition for autonomous driving which mainly depends on your ODD but they’re not the only ones. Honda ,Volvo and GM have something. Others (i.e. BMW) have stuff next year but they’re all going with more accurate names. CoPilot, PilotAssist, Super cruise, Traffic Jam Pilot. Makes it clear that these are drive assists, not drive replace.
Mercedes has Level 3 autonomy in certain highway situations, so you are legally allowed to watch a video or read a newspaper. You just need to be able to take over again within 20 seconds or so.
Others are following close up, I think Audi had to postpone Level 3 a bit etc. BMW has something in the pipeline as well.
But these are really more than drive assists. I really find the “Level n” specifications more helpful than “drive assist” vs “autonomy”
None of the other brands oversell what they are offering.
Without LIDAR, this is a fool’s endeavor.
I wish this was talked about every single time the subject came up.
Responsible, technologically progressive companies have been developing excellent, safe, self-driving car technology for decades now.
Elon Musk is eviscerating the reputation of automated vehicles with his idiocy and arrogance. They don’t all suck, but Tesla sure sucks.
Even with LIDAR there are just too many edge cases for me to ever trust a self driving car that uses current-day computing technology. Just a few situations I’ve been in that I think a FSD system would have trouble with:
-
I pulled up at a red light where a construction crew was working on the side of the road. They had a police detail with them. As I was was watching the red light the cop walked up to my passenger side and yelled “Go!” at me. Since I was looking at the light I didn’t see him trying to wave me through the intersection. How would a car know to drive through a red light if a cop was there telling you to?
-
I’ve seen cars drive the wrong way down a one way street because the far end was blocked due to construction and backtracking was the only way out. (Residents were told to drive out the wrong way) Would a self driving car just drive down to the construction site and wait for hours for them to finish?
-
I’ve seen more than one GPS want to route cars improperly. In some cases it thinks a practically impassible dirt track is a paved road. In other cases I’ve seen chains and concrete barriers block intersections that cities/towns have determined traffic shouldn’t be going through.
-
Temporary detour or road closure signs?
-
We are having record amounts of rain where I live and we’ve seen roads covered by significant flooding that makes them unsafe to drive on. Often there aren’t any warning signs or barricades for a day or so after the rain stops. Would an FSD car recognize a flooded out road and turn around, or drive into the water at full speed?
-
Just like that cheaper non-lidar Roomba with room mapping technology, it will get lost.
I don’t know why people are so quick to defend the need of LIDAR when it’s clear the challenges in self driving are not with data acquisition.
Sure, there are a few corner cases that it would perform better than visual cameras, but a new array of sensors won’t solve self driving. Similarly, the lack of LIDAR does not forbid self driving, otherwise we wouldn’t be able to drive either.
Yes, self driving is not computationally solved at all. But the reason people defend LIDAR is that visible light cameras are very bad at depth estimation. Even with paralax, a lot of software has a very hard time accurately calculating distance and motion.
challenges in self driving are not with data acquisition.
What?!?! Of course it is.
We can already run all this shit through a simulator and it works great, but that’s because the computer knows the exact position, orientation, velocity of every object in a scene.
In the real world, the underlying problem is the computer doesn’t know what’s around it, and what those things around doing or going to do.
It’s 100% a data acquisition problem.
Source? I do autonomous vehicle control for a living. In environments much more complicated than a paved road with accepted set rules.
You’re confusing data acquisition with interpretation. A LIDAR won’t label the data for your AD system and won’t add much to an existing array of visible spectrum cameras.
You say the underlying problem is that the computer doesn’t know what’s around it. But its surroundings are reliably captured by functional sensors. Therefore it’s not a matter of acquisition, but processing of the data.
won’t add much to an existing array of visible spectrum cameras.
You do realize LIDAR is just a camera, but has an accurate distance per pixel right?
It absolutely adds everything.
But its surroundings are reliably captured by functional sensors
No it’s not. That’s the point. LIDAR is the functional sensor required.
You can not rely on stereoscopic camera’s.
The resolution of distance is not there.
It’s not there for humans.
It’s not there for the simple reason of physics.Unless you spread those camera’s out to a width that’s impractical, and even then it STILL wouldn’t be as accurate as LIDAR.
You are more then welcome to try it yourself.
You can be even as stupid as Elon and dump money and rep into thinking that it’s easier or cheaper without LIDAR.It doesn’t work, and it’ll never work as good as a LIDAR system.
Stereoscopic Camera’s will always be more expensive than LIDAR from a computational standpoint.AI will do a hell of a lot better recognizing things via a LIDAR Camera than a Stereoscopic Camera.
This assumes depth information is required for self driving, I think this is where we disagree. Tesla is able to reconstruct its surroundings from visual data only. In biology, most animals don’t have explicit depth information and are still able to navigate in their environments. Requiring LIDAR is a crutch.
I disagree with you, I don’t think visual camera’s alone are up to the task. There was an instance of a Tesla in auto pilot mode driving at night with the driver being drunk. This took place in Texas on the high way, the car’s camera footage was released and it showed the autopilot not identify the police car in the lane with it’s red/blue lights flashing as a stationary obstacle. Instead it didn’t realize there was a car in the way around 1 second before the 55 mph impact, and it turned of autopilot that 1 second before.
Having multiple layers of sensors, some being good at actually sensing a stationary obstacle, plus accurate range finding, plus visual analysis to pick out people and animal, thats the way to go.
Visual range only cameras were just reported to have a harder time recognizing people of color and children.
the car’s camera footage was released and it showed the autopilot not identify the police car in the lane with it’s red/blue lights flashing
If the obstacle was visible in the footage, the incident could have been avoided with visible spectrum cameras alone. Once again, a problem with the data processing, not acquisition.
The elephant in the room is that the NHTSA still doesn’t have a director, and hasn’t had a long-term director since 2017.
Steven Cliff was the director for 2 months in 2022. Aside from that, this important safety organization has been… erm… on autopilot (see what I did there?) and leaderless.
How are we supposed to keep tabs on car safety if the damn agency in charge of automobile safety doesn’t even have a leader?
Wow. Impressive collection.
Somehow reminds me of Jehova’s witnesses and the end of the world :-)
deleted by creator
Lol, ok. Your anecdotal experience can totally be believed over all the data gathered over years. Great. Thanks.
Counter-counterpoint: I’ve been using it since 2019. I think you’re exaggerating.
-
It aggressively tries to center itself, always. If you’re in a lane and it merges with a second lane, the car will swerve sharply to the right as it attempts to go back to the middle of the lane.
-
It doesn’t allow space for cars to merge until the cars are already merging. It doesn’t work with traffic; it does its own thing and is discourteous to other drivers. It doesn’t read turn signals; it only reacts to drivers getting over.
-
If a motorcycle is lane-splitting, it doesn’t move out of the way for the motorcycle. In fact, it assumes anything between lanes isn’t an issue. If something is partially blocking a lane but the system doesn’t recognize it as fully “your lane”, the default is to ignore it. The number of times I’ve had to disengage to dodge a wide load or a camper straddling two lanes is crazy.
-
With the removal of radar, phantom braking has become far, far worse. Any kind of weather condition causes issues. Even if you drive at sunset, the sun can dazzle the cameras and they don’t detect things that they should be able to - or worse, they detect problems which aren’t there.
-
It doesn’t understand road hazards. It will happily hit a pothole at 70 MPH. It will ignore road flares and traffic cones. When the lanes aren’t clearly marked (because the paint has worn away or because of construction), it can have dramatic behavior.
-
It waits so long to brake, and when it brakes it brakes hard. It accelerates just as suddenly, leading to a very jerky ride that makes my passengers carsick.
The only time I trust FSD is when it’s stop-and-go traffic. Beyond that I have to pay so much attention to the thing that I might as well just drive myself. The “worst thing it can do” isn’t just detour; it’s “smash into the thing that it thought wasn’t an issue”.
It doesn’t read turn signals
It does in the FSD beta (somewhat). It even brakes and allows them in if it detects that they have a signal on. It doesn’t understand merges as well, but it’s still better than regular autopilot. All your other points are pretty valid. I am constantly taking it out of AP Ana putting it back in during a city drive, even though I have “FSD”.
-
The worst it will do is pick the wrong lane and detour a bit to get back on track.
https://www.cbsnews.com/amp/news/tesla-car-crash-nhtsa-school-bus/
It is actually 10 times more likely that you will get killed by a human driver than by FSD, but obviously they can not release the technology saying, that it will cause less accidents, because some expect 0 as if any technology is ever perfect. Tesla’s FSD keeps improving, people are not.
