Don't hold your breath for Blue Cruise Full Self Driving

[SpaceEVDriver]

In my opinion, it's nice to see an automaker listen to reality and stop with the foolishness of fully self-driving vehicles. Good on Ford for recognizing it's a fools' errand at this stage and that there are better things to spend their energy on.

Ford withdraws petition seeking U.S. approval to deploy self-driving vehicles

https://www.reuters.com/business/au...oval-deploy-self-driving-vehicles-2023-03-30/

Sponsored

 

[hybrid2bev]

Makes sense. Ford said in October 2022 they would focus on L2 and L3 drivers assistance when they shut down Argo AI. Getting to Level 4/5 is a money pit at this point.

 

[AliRafiee]

We are nowhere near ready. It still has decades to go.

 

[randomvoice]

Good.

Nothing wrong in being ambitious however, they need to be laser focused. BlueCruise has come a long way and keeps on getting better. Honestly, I trust it more than my Model 3's Auto Pilot since it hasn't phantom braked once so far. Let someone else figure out the FSD problem and license it down the road because it looks like a money pit right now with hardly a realistic difference. I just want a great 'aided' driving experience in my car for my commute to work. As long as it does that, I don't care if it doesn't drop me to my doorstep.

 

[SpaceEVDriver]

Good.

Nothing wrong in being ambitious however, they need to be laser focused. BlueCruise has come a long way and keeps on getting better. Honestly, I trust it more than my Model 3's Auto Pilot since it hasn't phantom braked once so far. Let someone else figure out the FSD problem and license it down the road because it looks like a money pit right now with hardly a realistic difference. I just want a great 'aided' driving experience in my car for my commute to work. As long as it does that, I don't care if it doesn't drop me to my doorstep.

I don't think Ford have stopped their development, just recognized that an attempt to push it out too early would be disastrous. There's no safe fsd at this time and there are too many unconstrained issues that make it unsafe for another 5-10 years at least.

We have trouble with "fsd" on our spacecraft and rovers where there are no other vehicles. Of course, we don't have the budget these private companies have, but it's a much easier problem and still is incredibly difficult to solve.

 

[Logal727]

FSD won't be a widespread reality until there is a standardized system where all cars can communicate with each other. This was supposed to be part of the 5GUW but... well.. you see how that went.

 

[Deleted member 16048]

I agree that FSD is not ready and it will take a very long time to get there. Though I think the question is do people want that in mass or do people still like actually driving?

My girlfriend doesn't like using BlueCruise because she likes to drive. I use it because it makes driving easier for me and less taxing on those long trips.

I think the Tesla fanboys fail to see that when you're using FSD, you have to pay more attention to the car than if they were doing the driving themselves. I don't think this makes things easier at all, and frankly that defeats the purpose of FSD. You have to have some trust in the system (But not fully trust). I feel BlueCruise can be trusted a ton more than FSD can be, and that puts it ahead in the race.

 

[Logal727]

I agree that FSD is not ready and it will take a very long time to get there. Though I think the question is do people want that in mass or do people still like actually driving?

My girlfriend doesn't like using BlueCruise because she likes to drive. I use it because it makes driving easier for me and less taxing on those long trips.

I think the Tesla fanboys fail to see that when you're using FSD, you have to pay more attention to the car than if they were doing the driving themselves. I don't think this makes things easier at all, and frankly that defeats the purpose of FSD. You have to have some trust in the system (But not fully trust). I feel BlueCruise can be trusted a ton more than FSD can be, and that puts it ahead in the race.

I would trust a proven system, but that would have take out the variables of human drivers that aren't using FSD. Like I said, it won't work until all vehicles are FSD.

 

[SpaceEVDriver]

I would trust a proven system, but that would have take out the variables of human drivers that aren't using FSD. Like I said, it won't work until all vehicles are FSD.

I completely agree. It also won't work with the current approach to AI/ML. They need to go through IRB review of a system that will record and analyze a trillion or so real hours of driving in all the possible contexts from rural to city; clear to stormy; into the sunrise or sunset; up and down steep inclines; etc., etc., etc.

We're just at the point of having the proper sensors on the vehicles and they don't even work in all situations.

People like to believe the claim of T having analyzed "millions of hours" of driving. That's not even enough to take a random sample of a single large city...

 

[Deleted member 16048]

I would trust a proven system, but that would have take out the variables of human drivers that aren't using FSD. Like I said, it won't work until all vehicles are FSD.

Yeah that's definitely part of it. I think even then the human still would have to give the system trust and that is the question. Will everyone do that even with it being proven? Maybe, but also maybe not since there are a lot of people that just never trust any technology. It will be interesting to see in our lifetime for sure.

The money pit side of it I believe is also true. Tesla is figuring that out by constantly raising its price. Yes it's getting "better" so you could say the value is growing, but also what was once promised at a lower price is the same promise they have now. They aren't truly adding anything of value to the "product". I think they made a big step backwards with moving to vision based parking assist. It's clearly not as accurate as the ultrasonic sensors and accuracy is huge when talking FSD and small space maneuvering. There is a reason our eyes use these extra sensors for assistance.

 

[Logal727]

Yeah that's definitely part of it. I think even then the human still would have to give the system trust and that is the question. Will everyone do that even with it being proven? Maybe, but also maybe not since there are a lot of people that just never trust any technology. It will be interesting to see in our lifetime for sure.

The money pit side of it I believe is also true. Tesla is figuring that out by constantly raising its price. Yes it's getting "better" so you could say the value is growing, but also what was once promised at a lower price is the same promise they have now. They aren't truly adding anything of value to the "product". I think they made a big step backwards with moving to vision based parking assist. It's clearly not as accurate as the ultrasonic sensors and accuracy is huge when talking FSD and small space maneuvering. There is a reason our eyes use these extra sensors for assistance.

Yeah, like was mentioned, there is no real convenience factor with FSD right now, it's just "wow this is cool". It wouldn't be until you can fully not pay attention and maybe get some work done in the back seat is when it would actually add some utility.

 

[SpaceEVDriver]

Yeah, like was mentioned, there is no real convenience factor with FSD right now, it's just "wow this is cool". It wouldn't be until you can fully not pay attention and maybe get some work done in the back seat is when it would actually add some utility.

At which point we're going to have reinvented...trains with private carriages. Crowd all the vehicles together and they'd get much, much better efficiency, at least on the freeway. Let a particular vehicle peel off whenever they need to and close the gap.

 

[Jimrpa]

I don't think Ford have stopped their development, just recognized that an attempt to push it out too early would be disastrous. There's no safe fsd at this time and there are too many unconstrained issues that make it unsafe for another 5-10 years at least.

We have trouble with "fsd" on our spacecraft and rovers where there are no other vehicles. Of course, we don't have the budget these private companies have, but it's a much easier problem and still is incredibly difficult to solve.

My knowledge is old, but when I was in that business, spacecraft had no “autonomy”. All orbital maneuvers were designed and validated on ground systems before commands were sent to the vehicle. Very old spacecraft didn’t even have automatic sun tracking for the solar arrays (can you say “safe hold”? ?) I understand that some of the Mars rovers have some degree of autonomy in local route planning, but I have no knowledge of that.

 

[SpaceEVDriver]

My knowledge is old, but when I was in that business, spacecraft had no “autonomy”. All orbital maneuvers were designed and validated on ground systems before commands were sent to the vehicle. Very old spacecraft didn’t even have automatic sun tracking for the solar arrays (can you say “safe hold”? ?) I understand that some of the Mars rovers have some degree of autonomy in local route planning, but I have no knowledge of that.

Yeah, our rovers on Mars now have some autonomous terrain navigation (at very low speeds). The OSIRIS-Rex spacecraft used natural feature tracking for autonomous navigation on its way to Touch And Go sampling at Bennu. The newer Mars landers and rovers also used autonomous landing systems to land safely.

Most of the spacecraft now also have the ability to autonomously find Earth if something causes them to get off pointing or if they have a safing event.

 

[Jimrpa]

Yeah, our rovers on Mars now have some autonomous terrain navigation (at very low speeds). The OSIRIS-Rex spacecraft used natural feature tracking for autonomous navigation on its way to Touch And Go sampling at Bennu. The newer Mars landers and rovers also used autonomous landing systems to land safely.

Most of the spacecraft now also have the ability to autonomously find Earth if something causes them to get off pointing or if they have a safing event.

You kids have it so easy these days! Back in the day, people dorked around for week with slide rules to develop geosynchronous stationkeeping maneuvers (just kidding - they actually used HP 11Cs ???)

Sponsored