Ameriflight Drone

ppragman said:
So I’m feeling a bit better (been doing some medical stuff) and would really like to talk about this stuff as it’s related to what I’m doing… I’ll be brief for now before going back to homework.

Suffice it to say,”this is coming” it’s right around the corner, your careers are probably safe for legal reasons and practical reasons, but I bet we see single-pilot 121 cockpits far faster than many folks would think. That’s ok, it’ll be really cool. Maybe the Airbus A420 will be single pilot - I’m only half kidding.

As for self-driving… it’s going to be a bit because it’s even harder than flying. By a lot, that said, that'll probably come first because of how many companies are working on it and the rewards if someone can get it right. We're close but probably not close enough.

I participated in a self-driving robot challenge / semester project last semester. Even lane identification is challenging to get to human levels. Not impossible but getting our crappy robot to drive around a taped off track was not trivial. There are lots of layers to the software, and they all have to work and not confuse each other.

Not only that, computer vision is really • hard. It’s what I’m studying in grad school - it’s not impossible - in principle it’s easy just a series of convolutional neural networks (CNNs), but it’s that AND you have to get it to work in a partially observable environment filled with other (unpredictable) agents - oh and the environment is changing constantly. There’s weather and road raging other drivers and poor vehicle maintenance, and grease on the cameras... yeah - it is not a trivial task.

That, and it’s not enough to be better than a human, that we can do - we need to be so much better than a human so that it’s indistinguishable from perfect before most people and regulators will trust it enough to not have other problems. You think mode confusion and complacency is bad in the airplane? Wait until Randy is using his Tesla to drunk drive home.

Tie that in with how unpredictable the actual operating environment is and it will be a bit before you don’t own a car and just use ubiquitous Johnny Cabs.

I guess this is all to say, "this is going to happen" it's going to be badass, no I don't think you guys will lose your jobs... but maybe the career won't exist in the same way as it does now (or to the extent it exists now) in 30 years.

I actually think that center controllers are a lot more likely to be automation targets too. That may actually be an easier hurdle to clear than automating the flight crew. Anyway, gotta have dinner then do stats homework.
Click to expand...

Stand by for a very opinionated posters expert opinion and subsequent laughing emoji. Questioning and doubting you on your actual graduate level studies... against his... opinion. Brace yourself.
 
Maximilian_Jenius said:
Stand by for a very opinionated posters expert opinion and subsequent laughing emoji. Questioning and doubting you on your actual graduate level studies... against his... opinion. Brace yourself.
Click to expand...
I mean, (i • hate that I’m going to say this) - Todd is not entirely wrong. I didn’t dig through the whole thread, but the tech as it exists right now is pretty f-ing amazing….

but it’s not as advanced as fanboys would have you believe. Also it’s worth mentioning that Tesla has been - well… to put it mildly • in how they’ve designed their UI, how they’ve responded to safety related issues, etc.

Fully autonomous is going to be a thing. Do I think the guy nose diving Twitter is going to be the one to do it? Maybe, but this • is much much harder than people think. Is the Tesla impressive? You damn right it is.

Is it as incredible as the fanboys say? Sometimes - which is “not good enough.” The problem is this is a sort of Pareto distributed amount of effort to get this right - that is to say the required work to get through the last 20% of engineering to get to “true” self driving is well over half the amount of effort of the entire endeavor.

That said we’ll see some major leaps and probably soon because there are a lot of really smart people working on this. How soon? Your guess is as good as mine, but I look forward to it. I want to live in Ian Banks’ “The Culture” - we’ll see if we get there.
 
Maximilian_Jenius said:
In other unrelated Tesla news...

Click to expand...


haha nice. I apparently didn't really retain much info about this situation. While the wife and I were cruising around socal last weekend, I casually mentioned this. She immediately freaked out and asked "well where was this?". I laughed and she said she just wanted to make sure it wasn't nearby. Turns out it was? Whoopsie.

Back to the subject, she did have a funny comment a while ago, when she said it was crazy she could make a phone call with her Apple Watch, and that we only dreamed of that watching inspector gadget as kids in the 80s. Obviously tech is advancing at an accelerating rate, compared to the last few decades. However I personally think the fear of this job, or even my previous military flying career, being replaced via automation is a little premature. I don't think my 7 year old will likely have the opportunity to do either of those things, but hey, he has no interest in aviation that I've seen but is very interested in "tech".
 
ppragman said:
I mean, (i • hate that I’m going to say this) - Todd is not entirely wrong. I didn’t dig through the whole thread, but the tech as it exists right now is pretty f-ing amazing….

but it’s not as advanced as fanboys would have you believe. Also it’s worth mentioning that Tesla has been - well… to put it mildly • in how they’ve designed their UI, how they’ve responded to safety related issues, etc.

Fully autonomous is going to be a thing. Do I think the guy nose diving Twitter is going to be the one to do it? Maybe, but this • is much much harder than people think. Is the Tesla impressive? You damn right it is.

Is it as incredible as the fanboys say? Sometimes - which is “not good enough.” The problem is this is a sort of Pareto distributed amount of effort to get this right - that is to say the required work to get through the last 20% of engineering to get to “true” self driving is well over half the amount of effort of the entire endeavor.

That said we’ll see some major leaps and probably soon because there are a lot of really smart people working on this. How soon? Your guess is as good as mine, but I look forward to it. I want to live in Ian Banks’ “The Culture” - we’ll see if we get there.
Click to expand...

I pretty much agree with you on this, contrary to Max’s smartass comments. The haters seem to forget that Tesla’s FSD is a Beta release product. Those of us who use it are testers. Does it work the vast majority of the time? Yes, it does. I can turn it on Atlanta and not touch it until I‘m in Florida. But that doesn’t mean it’s perfect already. They’re still working out the kinks. But when people make obviously uninformed comments like it not working in the mountain snow, yeah, I’m gonna call BS.
 
@SlumTodd_Millionaire what statistics are you referencing?

I 100% agree with @ppragman on this. The Tesla product is cool but it’s not level 5, and I question whether or not it will ever be if Elon refuses to reconsider his opposition to active sensing tech.
 
SlumTodd_Millionaire said:
They’re still working out the kinks.
Click to expand...

many people (myself included)think that having your customers work as beta testers in such a fashion on a safety sensitive product is kind of unethical or at least dubiously ethical. A website? Sure do AB tests, a massive semi-autonomous robot flying down the road? I am not too keen on that.

My problems with Tesla are almost entirely the workplace environment, the somewhat disingenuous marketing, disagreements I have with how they have made certain safety related decisions, and how their CEO behaves.

The tech is amazing. But there is also a lot of work before we get to “fully automated.”

Do I think they’ll do it? Probably if they can get out of their own way to do it. It would help if Elon stops taking up all the oxygen in the room screeching on Twitter and focused on running is company. But that’s a different sort of discussion

SlumTodd_Millionaire said:
But when people make obviously uninformed comments like it not working in the mountain snow, yeah, I’m gonna call BS.
Click to expand...

I mean “work” is relative, right? I think a lot of people expect it to do a lot more than what it is actually capable of. I think that’s partially due to some of that disingenuous marketing I spoke about earlier - but also every company does that.

Anecdotally, I know a few people with Teslas, literally every one of them rages about it. Much of the quality problems seem to have gone away in the last few years.

I have never driven one so I won’t comment on the experience of driving one, but it looks cool.

This tech though is really hard. I don’t think we’ll get to level 5 as fast as many of the fanboys think. That said, we will get there.

zmiller4 said:
and I question whether or not it will ever be if Elon refuses to reconsider his opposition to active sensing tech.
Click to expand...

I don’t know about this. You mean opposition to LiDAR right?

(• me in the heart for saying this) Elon may not be totally out to lunch for opposing LiDAR - humans don’t have LiDAR and we can drive. arguably not at the level of safety that we should, but we do it every day.

honestly I think we will get to level 5 - it’s just going to be a LOT more work before we’re there.

I hesitate to give a specific timeline because that’s always scary, but I wouldn’t be surprised if we were there in less than a decade. The laws are going to be weird though.
 
One other thing I’ll add is that the future is really unpredictable and things are changing really fast.

You guys remember 12 years ago or so iPads weren’t a thing and we all were using paper charts and many of us were calling flight service for a briefing. Hell I was doing that 6 years ago. In my first job I used a literal sectional to navigate across practically all of Alaska and in my second flying job we using weather radar as a “poor man’s moving map.”

The sectional down by PACD said “topographic data not available” close to False Pass when I was in flight school (roughly 17 years ago). ADS-B was only just being beta tested in Alaska. Drones weren’t a thing at all outside the military and now Ukrainian teenagers are bombing Russians with quad copters you can buy in target.

I’m not a Kerzweill-Ian (those beliefs are just the rapture for nerds) but the changes coming down the pike will be incredible and will remake this industry in ways we can’t predict.
 
ppragman said:
many people (myself included)think that having your customers work as beta testers in such a fashion on a safety sensitive product is kind of unethical or at least dubiously ethical. A website? Sure do AB tests, a massive semi-autonomous robot flying down the road? I am not too keen on that.
Click to expand...

Well, that’s the only way to really do it, because the volume of data that is needed just isn’t something you can do with a handful of test vehicles in California. They’ve done a good job by being very clear that it’s a beta test and using technology to assure driver attention. I see no ethical issues, personally.

I mean “work” is relative, right? I think a lot of people expect it to do a lot more than what it is actually capable of.
Click to expand...

I’m using it in the context that the claim was made in this thread. A claim was made that it simply won’t work in mountain snow. And that is blatantly false. Honestly, I was shocked that it did work when I could barely make out the road myself. But it did.
 
SlumTodd_Millionaire said:
I’m using it in the context that the claim was made in this thread
Click to expand...
Well I don’t really care what pissing contest you guys had here earlier so you could feel self-satisfied, I’m saying “work” is relative. There are things it “can’t do yet” and situations outside it’s wheelhouse. It will get there, but a great many criticisms of the technology are warranted - no poor performance in snow and mountainous terrain probably isn’t as warranted as people would have you believe - though from what I gather this was the case? Sure your point is reasonable. But also it has a long way to go yet and don’t oversell it.

Granted I’ve never driven one, so take my comments with a boulder of salt as well. Still, this stuff is much much more complex than people think.

If it were me I’d be curious to see how many times it had problems determining where the lanes were on a mountain drive in the snow. Also, I’d be curious to see that data for here in Anchorage- I haven’t seen a lane marking in 3 months…

Now here’s a genuine question - Todd, when your Tesla has a sort of issue figuring out where the lanes are etc - what does it do? Also can you get any metrics from your vehicle about how many times it gets confused etc on a drive or is that proprietary? What annunciations does it have etc? I know the early ones were giving non-aural alerts when they were disconnecting but I guess they fixed that about 3 years ago now (maybe 4?).
 
SlumTodd_Millionaire said:
I pretty much agree with you on this, contrary to Max’s smartass comments. The haters seem to forget that Tesla’s FSD is a Beta release product. Those of us who use it are testers. Does it work the vast majority of the time? Yes, it does. I can turn it on Atlanta and not touch it until I‘m in Florida. But that doesn’t mean it’s perfect already. They’re still working out the kinks. But when people make obviously uninformed comments like it not working in the mountain snow, yeah, I’m gonna call BS.
Click to expand...

arstechnica.com

Tesla tells investors it’s being investigated by the Justice Department

The carmaker's 10-K confirms reports from 2022 that the feds are unhappy.
arstechnica.com arstechnica.com
 
ppragman said:
Now here’s a genuine question - Todd, when your Tesla has a sort of issue figuring out where the lanes are etc - what does it do?
Click to expand...

It happens very rarely, and basically never on the expressway. When it happens on city streets, it will either slow down to figure it out, or will it emergency disconnect with loud warnings.

Also can you get any metrics from your vehicle about how many times it gets confused etc on a drive or is that proprietary?
Click to expand...

Ss far as I know, I can’t get that info, but I haven’t investigated to see if there’s a way.
 
SlumTodd_Millionaire said:
It happens very rarely, and basically never on the expressway. When it happens on city streets, it will either slow down to figure it out, or will it emergency disconnect with loud warnings.



Ss far as I know, I can’t get that info, but I haven’t investigated to see if there’s a way.
Click to expand...
Thank you, that is interesting
 
58EB6AD1-110F-4AB2-82EF-0DEB642718E8.jpeg
 
mikecweb said:
w3.ameriflight.com

Ameriflight Signs Agreement with Natilus to Purchase 20 Autonomous Cargo Aircraft - Ameriflight

January 26th, 2023 FOR IMMEDIATE RELEASE Media Contact: Jamie Smith Ameriflight, LLC Director of Corporate Communications js****@*********ht.com | 972.426.3066 ext. 76273 DALLAS, TX – Ameriflight, the nation’s largest Part 135 Cargo airline, announces agreement to purchase 20 autonomous cargo...
w3.ameriflight.com w3.ameriflight.com

View attachment 69358
Just put a penis tail on it so I can see the end and get it over with.
Click to expand...
If they're gonna go down that twisted path, they should have gone with BETA. Oh, well. Some billionaires gonna pay. Oh, wait! No they won't. They'll just extort the US government -and us taxpayers- à la the SVB "bailout". Social makes one strong these days. Eff the facts. It's all popular perception these days, maybe I need a refresher course.
 
Last edited:
You must log in or register to reply here.
Back
Top