You didn't ask me that, but yes I believe the hype. Elon managed to recover a rocket and relaunch it. AP2 is near feature parity with AP1. All your comments about who is leading whom in FSD in moot as all you see are marketing demo videos and company blog posts. When there is an actual FSD car from GM, that we can buy, you can proclaim victory then. Till then there is actual L2 car that I drive daily.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
GM's cruise automation transcendent 90 mins self driving demo
- Thread starter Bladerskb
- Start date
-
- Tags
- Autonomous Vehicles
You didn't ask me that, but yes I believe the hype. Elon managed to recover a rocket and relaunch it. AP2 is near feature parity with AP1. All your comments about who is leading whom in FSD in moot as all you see are marketing demo videos and company blog posts. When there is an actual FSD car from GM, that we can buy, you can proclaim victory then. Till then there is actual L2 car that I drive daily.
The first real L2 car will arrive from GM in the form of supercruise.
What does launching and recovering rockets have to do with anything? my apologies, I forgot that elon was a rocket scientist, forgive me for thinking he was just a CEO that sets lofty goals and lies about timelines that eventually come to pass 2-3 years late.
Yes elon has been late on his predictions/promises/vows in both spacex and tesla.
Anyway I didn't know DMV reports are company blog posts. dmode I have sinned, cleanse me for thinking other companies can release fsd videos other than tesla, these videos are marketing demos and tesla videos of empty road with virtually no traffic is the real deal.
Tam
Well-Known Member
I think you pretty much sum it all up.
The trust might be questionable but I make my decision on what info have been available to me.
I placed a deposit for Model S when I saw what Tesla could do with its first model: Roadster.
I now bought AP2 because I saw what Tesla could do with AP1.
3 owners file a class action law suit against for Tesla for selling AP2 before it's perfected.
I on the other hand is glad that Tesla offers what's available without waiting for tomorrow.
alseTrick
Active Member
jschwefel
Tesla fan/TSLA, Model Y and Cybertruck owner.
Why all the animosity toward Elon and Tesla? No, they're not perfect, but they are doing some great work, and that's causing others to step up their game. Bring on the competition and may the best company win!
JohnSnowNW
Active Member
Why are we comparing an unavailable product to an available product...?
It's a great demo of what an impractical consumer system can do, I suppose.
It's a great demo of what an impractical consumer system can do, I suppose.
Would love to see how the lidar on the GM car can cope with fog, heavy rain, snowstorms, cyclists chasing racoons through puddles, dusty roads, etc.
As far as AP2 is concerned, I think we will see a step change in capability when mapping is added, even though I personally don't buy into the need for HD maps (or whitelists) for autonomy. IMO they are a short-term crutch to make up for lack of DNN capability (training & compute power). No matter how high the definition of the map, the car still has to cope with the unexpected - from parked cars and soda cans to pot holes and new fences. I would prefer the AP resources to be spent on object detection, path prediction and determination of intent (of other road users) rather than mechanically validating every object against a map / whitelist and having to deal with false-positives.
As far as AP2 is concerned, I think we will see a step change in capability when mapping is added, even though I personally don't buy into the need for HD maps (or whitelists) for autonomy. IMO they are a short-term crutch to make up for lack of DNN capability (training & compute power). No matter how high the definition of the map, the car still has to cope with the unexpected - from parked cars and soda cans to pot holes and new fences. I would prefer the AP resources to be spent on object detection, path prediction and determination of intent (of other road users) rather than mechanically validating every object against a map / whitelist and having to deal with false-positives.
What available product do you speak of? please don't say a model s/x with a L2 system that is 7 months late.
They also use cameras and radar all around their cars. You ppl still don't get it.
I don't think you understand the function of maps.
JohnSnowNW
Active Member
You're the one whining about AP2 here constantly, and started a thread about GMs automation system on a Tesla forum. My apologies if you were discussing the GM system outside of the context of your activity on this site.
If the Lidar is blinded, it could kick into human mode, and drive in the fog at a speed that makes it impossible stop inside your visual range.
Or it could kick into trained driver mode, where the car will not outrun it's sensors. But people would not stand for a safe car.
But more importantly, control navigation, just like the accelerator pedal and brake pedal, need to have redundancy. If a failure in the sole system kills, you run 2 or 3 systems, like we do with aircraft.
Geomapping is your life today. Your navigation you use today is based on a crude 2.5D version of it.
And you forget GM vehicles have V2V cars which will increase in numbers as time goes on. Hopefully they will mesh with other V2V systems, and enhance range finding.
Example? Fog. If some of the cars in your area can report their ground speed at various mile markers, your car will never be involved in a pile up. You will be slowing way before the crash scene.
I talk about ap2 because it is highly deficient.others don't see this because the y just lap up every thing Elon says like what Trump fans do. They haven't done independent thinking and independent analysis. If they did they would see that ap2 is a camera only system and that the radar is useless in almost all driving situations other than driving forward in a lane.
Intersection? Useless.
U-turn? Useless
Yellow lane turn? Useless.
T-section turns? Useless.
Roundabouts? Useless.
Regular left/right intersection turns? Useless
95% of urban driving requires side and back view coverage. Tesla's one forward radar is therefore useless in almost all urban driving scenario
Please enlighten us all: what IS the function of HD maps in autonomous driving?
Not really. Most of us can navigate from A to B using road signs, so long as we are aware of the general location of B relative to A. I don't need a map to tell me where traffic lights or overhead bridges are.
Bully for you! Very few people can get by without maps of any kind.
But it is safer to have heads-up and hands-free navigation when driving in traffic, especially when signs are missing or damaged.
Real hardcore folk go through the Mojave Desert 400 miles with no road signs, pavement, or maps. You use a ribbon of paper and your odometer (which is always wrong) to navigate through the mountains and deserts at 15-100mph while looking for ribbons and advancing your enduro sheet.
Reciprocity
Active Member
I think the big mistake that people keep making is that AP2 is not FSD V1.0. AP2 and FSD are different software and different code bases today. In a perfect world, Tesla would have been able to use AP1 in conjunction with HW2, but Mobileye wouldn't allow it knowing that Tesla was working on its own solution. This is public knowledge and an indisputable fact that Telsa and Mobileye had a falling out and would not allow Tesla to use it in parallel with HW2. This forced Tesla's hand to implement AP2 on HW2 and separately develop FSD. They obviously thought this would be easier then they had anticipated because they had already done it once with EyeQ3 and one camera.
FSD will not need to see lane markings, though it will still use them if it can. It will use high definition 3d maps that will use a combination of GPS and landmarks, sensors and telemetry from the car. This would include lane markings but doesnt require them as it can use other land marks to compensate for GPS's weaknesses. GPS gets you within 3 meters, adding high def maps with land marks can get you to 10 centimeters or so from watching nVidia's presentations. Adding the cars other sensors and telemetry can get you down to 2 centimeters. Once you have an identifiable landmark in a high precision 3d map, you dont even need GPS to continue to work. GPS gets you into the neighborhood, landmarks get you into the lane and the other sensors and the speed/direction the car is traveling gets you the rest of the way to 2cm. The high def maps could be created by Lidar and updated by radar in current cars as roads and landmarks change. AP1 and AP2 do not do any of that. They identify the lanes by seeing the lane markings and curb or edge of the road. It then uses the TACC radar not to run into the car in front of it. That is all AP1/2 does.
FSD is being developed separately and is nothing like AP2. At some point in the near future (EOY), FSD should replace AP2 and there will only be AP1 on EyeQ3 and FSD on HW2. They only had to do this clunky way because of the fallout with Mobileye after the accident where the guy died using AP1 and it didnt see a truck passing in front of it.
If you go back to the Oct. video that Tesla released when it started talking about FSD, you will the different camera, many more then 2 and you can see how the high def mapping works with landmarks and how it knows the surface of the road. It can see all the signs as noted by different color boxes around signs, people, cars and other objects. None of this stuff happens with AP2. It sees lanes and curbs and the radar keeps you from driving through the car directly in front of you. That is it.. that's all AP2 does. So unless you think the video from Oct. 2016 was just Hollywood product, there has to be two different solutions being developed independently by Tesla.
Tesla is also not building either system of the ground up. They are leveraging AP1 software that was created by Tesla to enhance it and they are piggybacking on nVidia's solution to jump start FSD. A major difference between FSD and nVidia's solution is FSD uses radar to create a point cloud and nVidia uses Lidar. Both have their strengths but I dont think Tesla will ever use a Lidar. Not because of the cost, not because of the fact that its ugly, but the fact that is not good at detecting objects in rain, fog or snow. You could say, well then why dont they use both. Because then you have to decide, which one do you trust? There will be times when they disagree on what they are seeing and you cant just slam on the brakes until it figures out which is right. So this problem is one that Tesla and as far as I can tell, only Tesla is attempting. They are attempting to use the radar to confirm what they are seeing using the Vision system and they believe that they can create a detailed enough picture of the world around them, that is more accurate then what you and I can see while be constantly distracted. When I say distracted I dont just mean talking on your cellphone, but even something as simple as looking at your current speed or Navigation or even admiring a passing Tesla or waiving at a nice looking lady in the next car over. Tesla is not going to sacrifice good enough for perfect, they have shown that they can save lives without a perfect solution.
FSD will not need to see lane markings, though it will still use them if it can. It will use high definition 3d maps that will use a combination of GPS and landmarks, sensors and telemetry from the car. This would include lane markings but doesnt require them as it can use other land marks to compensate for GPS's weaknesses. GPS gets you within 3 meters, adding high def maps with land marks can get you to 10 centimeters or so from watching nVidia's presentations. Adding the cars other sensors and telemetry can get you down to 2 centimeters. Once you have an identifiable landmark in a high precision 3d map, you dont even need GPS to continue to work. GPS gets you into the neighborhood, landmarks get you into the lane and the other sensors and the speed/direction the car is traveling gets you the rest of the way to 2cm. The high def maps could be created by Lidar and updated by radar in current cars as roads and landmarks change. AP1 and AP2 do not do any of that. They identify the lanes by seeing the lane markings and curb or edge of the road. It then uses the TACC radar not to run into the car in front of it. That is all AP1/2 does.
FSD is being developed separately and is nothing like AP2. At some point in the near future (EOY), FSD should replace AP2 and there will only be AP1 on EyeQ3 and FSD on HW2. They only had to do this clunky way because of the fallout with Mobileye after the accident where the guy died using AP1 and it didnt see a truck passing in front of it.
If you go back to the Oct. video that Tesla released when it started talking about FSD, you will the different camera, many more then 2 and you can see how the high def mapping works with landmarks and how it knows the surface of the road. It can see all the signs as noted by different color boxes around signs, people, cars and other objects. None of this stuff happens with AP2. It sees lanes and curbs and the radar keeps you from driving through the car directly in front of you. That is it.. that's all AP2 does. So unless you think the video from Oct. 2016 was just Hollywood product, there has to be two different solutions being developed independently by Tesla.
Tesla is also not building either system of the ground up. They are leveraging AP1 software that was created by Tesla to enhance it and they are piggybacking on nVidia's solution to jump start FSD. A major difference between FSD and nVidia's solution is FSD uses radar to create a point cloud and nVidia uses Lidar. Both have their strengths but I dont think Tesla will ever use a Lidar. Not because of the cost, not because of the fact that its ugly, but the fact that is not good at detecting objects in rain, fog or snow. You could say, well then why dont they use both. Because then you have to decide, which one do you trust? There will be times when they disagree on what they are seeing and you cant just slam on the brakes until it figures out which is right. So this problem is one that Tesla and as far as I can tell, only Tesla is attempting. They are attempting to use the radar to confirm what they are seeing using the Vision system and they believe that they can create a detailed enough picture of the world around them, that is more accurate then what you and I can see while be constantly distracted. When I say distracted I dont just mean talking on your cellphone, but even something as simple as looking at your current speed or Navigation or even admiring a passing Tesla or waiving at a nice looking lady in the next car over. Tesla is not going to sacrifice good enough for perfect, they have shown that they can save lives without a perfect solution.
Am I missing something here? I have an AP1 Model S. Is there another car with better driving assistance I can buy today ? Well obviously not a new Tesla with AP2 
but is there any other car which does those things better than Tesla's AP1? I would really be interested.
AP1 drove me from Switzerland back to Belgium last week, and on all highways (99% of the trip), I only had to intervene at (some) ramps between highways and (some) road works - in total three or four times over 750 km. If any car does better, I would really consider it (even an ICE car, to be honest).
Anyone know of such a car I could buy?
but is there any other car which does those things better than Tesla's AP1? I would really be interested. AP1 drove me from Switzerland back to Belgium last week, and on all highways (99% of the trip), I only had to intervene at (some) ramps between highways and (some) road works - in total three or four times over 750 km. If any car does better, I would really consider it (even an ICE car, to be honest).
Anyone know of such a car I could buy?
What does that have to do with fully self driving cars? Clearly you are missing something..Am I missing something here? I have an AP1 Model S. Is there another car with better driving assistance I can buy today ? Well obviously not a new Tesla with AP2
AP1 drove me from Switzerland back to Belgium last week, and on all highways (99% of the trip), I only had to intervene at (some) ramps between highways and (some) road works - in total three or four times over 750 km. If any car does better, I would really consider it (even an ICE car, to be honest).
Anyone know of such a car I could buy?
My car navigates from one place to another with only an organic computer coupled with two primitively vision sensors supplemented by two aural sensors. I tend to agree with Musk. We are overthinking the hardware requirements it's mainly a software/ computer problem. It does this in areas with few or no roads and no maps with astounding accuracy.
Similar threads
