scottf200
Well-Known Member
Processing power being one of the differences vs 5, 10, 50 ("half century") years ago.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Autonomous Car Progress
- Thread starter stopcrazypp
- Start date
-
- Tags
- Autonomous Vehicles
DanCar
Active Member
DanCar
Active Member
Last I heard they have 1,500 employees and are losing close to a billion dollars per quarter. They have the sincerest goals in their hearts, but that doesn't necessarily mean a winning recipe. My biggest issue, is they are risk adverse and with the long tail being infinite long, there will always be risk. So unless something changes drastically Waymo will will have a very difficult time to launch. Heaven forbid there is a crash with blood in the video and it will cost Google millions is market appeal.
If they sold to a venture capital firm, I can see them being instantly successful. The VCs can handle the bad press, by saying they are much safer than average and therefore saving lives, and improving the quality of life by not requiring driving.
diplomat33
Average guy who loves autonomous vehicles
mikes_fsd
Banned
And it almost killed them (if it wasn't for SpaceX loan money, it would have).
mikes_fsd
Banned
Let me try to make a different example.
The Starship prototype that SpaceX launched last night on a test flight... (if you have not seen the footage, it's here
)
As Eric Berger says in his article An early version of Starship takes its first tentative steps off Earth
There are 2 vital things (that fundamentally are required) they needed to be prior to getting there, the Raptor engine and the flight software.
Elon first started with carbon composite tanks and had expensive tooling made for making the large carbon fiber rings for the body of the Starship.
When early pressure tests failed and no way to increase production speed for their carbon fiber approach, he pivoted to stainless steel (30x variants) and literally scrapped the expensive carbon fiber tooling.
Now, they are cranking out prototypes every few weeks. We will start to see additional hops to smooth out the fueling, launch and landing process further and gather more data.
If you look at SN5 it looks like it has a bunch of appendages on the outside skin. The vast majority of those black canisters will be removed/deleted as the Starship will be using its own fuel to pressurize the tanks rather then using helium (aka Autogenous pressurization). And their RCS (reaction control system) thrusters will be replaced to use the Methane as fuel, further removing the need for additional storage containers and removing complexity of ground operations and maintenance.
Rockets are complex beasts, every system or sub-system that you add to your rocket must be able to withstand the forces during launch and landing... and handle the conditions in the vacuum of space. So, removing sub-systems and using the same fuel as the main propulsion allows them to eliminate a lot of complex testing and software to control the additional sub-systems.
I see the Tesla approach to FSD in a similar light.
They have the 2 vital things the custom designed FSD computer and the "Tesla Vision" software to run on that computer. (we are obv. still waiting on the rewrite)
The rest of the sensors need to provide enough detailed information for those 2 vital things to make useful & safe decisions/predictions while operating the vehicle.
Adding additional sensors that provide more information then is needed by the system is only going to have adverse effects on your entire fleet down the line.
If new sensor slow the the progress (just like carbon fiber did for Starship) it is time to scrap those sensors and rethink your approach.
Each sub-system has it's hardware and maintenance associated with it, then it has it's code and the maintenance that goes with that sub-system to get the useful data out of that hardware.
The "first principle" approach that Elon brings up often, in this application is passive cameras with active radar and ultrasonic sensors. Since they've revealed their vision in 2016, they've stuck to it, even when they announce that they are doing a rewrite of their software the underlying vision and principles of their solution remained the same.
As a throwback to "but Elon said" .... here is an interview with Wall Street Journal from September 2014.
Elon is quoted as saying
Also, it is clear that the current implementation of Autopilot is not there.
paywalled link Tesla CEO Musk Sees Fully Autonomous Car Ready in Five or Six Years
non-paywalled link through Verge Elon Musk says self-driving car technology still has another 5 to 6 years to go
The Starship prototype that SpaceX launched last night on a test flight... (if you have not seen the footage, it's here
As Eric Berger says in his article An early version of Starship takes its first tentative steps off Earth
There are 2 vital things (that fundamentally are required) they needed to be prior to getting there, the Raptor engine and the flight software.
Elon first started with carbon composite tanks and had expensive tooling made for making the large carbon fiber rings for the body of the Starship.
When early pressure tests failed and no way to increase production speed for their carbon fiber approach, he pivoted to stainless steel (30x variants) and literally scrapped the expensive carbon fiber tooling.
Now, they are cranking out prototypes every few weeks. We will start to see additional hops to smooth out the fueling, launch and landing process further and gather more data.
If you look at SN5 it looks like it has a bunch of appendages on the outside skin. The vast majority of those black canisters will be removed/deleted as the Starship will be using its own fuel to pressurize the tanks rather then using helium (aka Autogenous pressurization). And their RCS (reaction control system) thrusters will be replaced to use the Methane as fuel, further removing the need for additional storage containers and removing complexity of ground operations and maintenance.
Rockets are complex beasts, every system or sub-system that you add to your rocket must be able to withstand the forces during launch and landing... and handle the conditions in the vacuum of space. So, removing sub-systems and using the same fuel as the main propulsion allows them to eliminate a lot of complex testing and software to control the additional sub-systems.
I see the Tesla approach to FSD in a similar light.
They have the 2 vital things the custom designed FSD computer and the "Tesla Vision" software to run on that computer. (we are obv. still waiting on the rewrite)
The rest of the sensors need to provide enough detailed information for those 2 vital things to make useful & safe decisions/predictions while operating the vehicle.
Adding additional sensors that provide more information then is needed by the system is only going to have adverse effects on your entire fleet down the line.
If new sensor slow the the progress (just like carbon fiber did for Starship) it is time to scrap those sensors and rethink your approach.
Each sub-system has it's hardware and maintenance associated with it, then it has it's code and the maintenance that goes with that sub-system to get the useful data out of that hardware.
The "first principle" approach that Elon brings up often, in this application is passive cameras with active radar and ultrasonic sensors. Since they've revealed their vision in 2016, they've stuck to it, even when they announce that they are doing a rewrite of their software the underlying vision and principles of their solution remained the same.
As a throwback to "but Elon said" .... here is an interview with Wall Street Journal from September 2014.
Elon is quoted as saying
Obviously we are at the end of that 6 year window and will have to see what the 4D rewrite brings to see how close or how far Elon time was off.
Also, it is clear that the current implementation of Autopilot is not there.
paywalled link Tesla CEO Musk Sees Fully Autonomous Car Ready in Five or Six Years
non-paywalled link through Verge Elon Musk says self-driving car technology still has another 5 to 6 years to go
Last edited:
powertoold
Active Member
It seems Elon thinks about fundamentals and extrapolates his first principles assessment out into the future based on what he thinks is possible. He essentially is able to foresee the future viability of an approach or technology better than most of us, although he's often way too optimistic with the timeline. I think that's because to him, it seems like it's already possible, so why would it take so long?
I still have doubts that Tesla will achieve level 4.5 or 5 any time within the next 2-3 years, but even if it takes another 2-3 years, it would once again reinforce Elon's ridiculous ability to imagine the future and all the necessary details to make it happen.
I still have doubts that Tesla will achieve level 4.5 or 5 any time within the next 2-3 years, but even if it takes another 2-3 years, it would once again reinforce Elon's ridiculous ability to imagine the future and all the necessary details to make it happen.
Seeing a Falcon 9 land gives me chills every time, I watched the SN5 live stream on Tuesday before they scrubbed the launch.
Rockets have been solved, we have about 60 years of knowledge on how to launch rockets and objects into space and bring them back down to earth. There are lots of books on how to do it. It is a very costly endeavor but it is not brand new frontier. We have landed and returned vehicle from the moon. Rockets being complex in their own right does not mean it is more difficult than Autonomous driving, the approach you use to solve one does not in anyway mean the same approach to solve another.
Guess who also has 3 vital things? Waymo, they make their own sensors, they designed their own compute and write their own software. A single waymo car has more camera sensors than 3 model 3 teslas plus 6 more. They have perimeter Lidar, Radar and Ultra Sonic sensors that provide full 360 degree coverage . They also have HD maps, that is more data than a camera alone provides. And you called it a clusterfuck. The share arrogance.
I take it you have some informed sources for your claim? Who made you the arbiter on what amount of data is needed? This has not been completely solved yet. Even the right amount of sensors and combination varies from company to company. Heck Mobileeye is driving with cameras only, no radar or ultrasonic, does that mean Tesla does not need a radar? What remains true is that we need a combination of different sensors to provide more information and redundancy.
So additional sensors are now just like carbon fiber? They serve no purpose but to slow the progress? What if i told you waymo has been using lidar from the very beginning? What if i told you the first cars to win the Darpa autonomous vehicle challenge did so with lidars. A rocket is not in the same stratosphere in terms of cost compared to a vehicle. Stop conflating two absolutely different subjects, it serves no purpose but to obscure and muddy the discursion.
What i find fascinating and ironic is you are doing exactly what Elon Musk says goes against first principle approach to solving problems, you are drawing analogy to something else. Its the same stupid analogy i have seen people make and Elon Musk himself make. if we drive with just two eyes why can't we drive with just cameras. I can also drive safely with 1 eye. First principle approach necessitates us to think outside the box, additional sensors and HD maps simplify not add complexity. They also provide redundancy, if 1 camera fails or is blocked, you have coverage from other sensors.
What you are doing is so freaking transparent. If Elon Musk can solve rockets (it has already been solved) then surely he is right about autonomous vehicle. You are not making a case for less sensors or the validity of camera only approach, you are making a case for believing in Elon Musk. And again i say majority of people on this board believe in Elon's vision, less they would not buy Tesla stocks or buy Tesla vehicles or fork out 2 to 8 grand for the hope of autopilot being able to achieve L5 self driving capability.
What has been the point of various discussion is the aggressive timeline Elon Musk sets and when he will be able to deliver on the vision using the proposed hardware at the time. The hardware and software is constantly changing which means this is a moving target with no firm date in sight.
Last edited:
Daniel in SD
(supervised)
What is a "first principle" approach? It sounds like you're saying it's just sticking with the first thing you come up with?
DanCar
Active Member
So any company that prints something is proof? No that is not proof. If waymo says they are king of the earth are you going to believe it? Anyone can say anything, that is not proof. Elon has told many lies about self driving. Elon said automatic driving in city streets last year. Is that proof that there was automatic driving in city streets last year?
Like anything related to AP it's inconsistent.
So yes, you should be absolutely scared to use it on your car.
Occasionally I'll test it, but only in very controlled environments. In fact sometimes I test it while being in the car so I can brake immediately if needed.
In all my testing of Smart Summon it hasn't passed my 10 out of 10 tries doing the same smart summon. Sometimes it works, but it typically fails for some reason. I even with through the effort of adding the parking lot to OpenStreetMaps.
So I've shelved the testing until some future re-write of it or enhancement.
diplomat33
Average guy who loves autonomous vehicles
There is video proof to back up Waymo's claim of 5% driverless rides. We showed you plenty of videos of Waymo robotaxis driving around with no driver. There is no video proof to back up Elon's claim. That's the difference.
DanCar
Active Member
There is no video proof of 5%. There are a few videos that are stale, more than 6 months old.
diplomat33
Average guy who loves autonomous vehicles
Wrong. This video was uploaded to imgur just 2 days ago. It is very fresh. It is not 6 months old.
Imgur
willow_hiller
Well-Known Member
Waymo doesn't have an incentive to lie about their FSD. They don't need more investment, they've got plenty of cash from Alphabet.
But as always, we should be wary about things written online without verifiable proof. If we didn't have Teslas in our garages to personally test, a lot more people would think Tesla has L4 completed already.
But as always, we should be wary about things written online without verifiable proof. If we didn't have Teslas in our garages to personally test, a lot more people would think Tesla has L4 completed already.
diplomat33
Average guy who loves autonomous vehicles
True. Plus, if they were going to lie about 5%, you'd think they would have picked a bigger number that sounds more impressive.
The Roadster was the car that showed the world that an EV need not be a golf cart. The Roadster was a showcase for Tesla's drivetrain technology. The Roadster created the awareness of Tesla and created the public interest that made it possible to raise enough capital to build the Model S. An argument can be made that the Roadster was what made the Model S possible and that Tesla owes it's success to the Roadster.
DanCar
Active Member
Wrong, you post a single video in the last 6 months and claim that is proof Waymo has 5% driverless.
DanCar
Active Member
Sorry, alphabet isn't stupid. They aren't going to throw billions of dollars down the drain if they aren't seeing progress. Heads will role without progress, so there is plenty of incentive to show progress. What you see of Waymo's claims isn't much different from x years ago. For example they said they did their first driverless ride in 2015.
Last edited:
Similar threads
- Article
