I always wonder how they developed things like that in the 70s. Right now I don’t know a single engineer that could even know how to begin to develop a system like that. I feel like most of them would try and throw a neural net at the problem and call it a day.
>I don’t know a single engineer that could even know how to begin to develop a system like that
It's called "controls engineering", and there are plenty of talented people in the field. But you are correct that people who sought a career in websites and cellphones have never heard of it.
The Russian military is organized around a land defense of their homeland since they don't have the natural barriers the U.S. has (e.g. they last got invaded in 1941, we last got invaded in 1812.)
You might think they'd invest in fixed SAM systems but mobile SAM systems are so more survivable in a war because the radar for a SAM is highly visible. You really want to hide those systems in terrain and ambush planes going by.
Systems like that are most effective when they operate in a networked mode together with AWACS airplanes like
which use a "look down" pulse doppler radar which can see cruise missiles, UAVs and other low-flying threats. This is a great complement to the radar on a Patriot missile battery which is optimized to ballistic missile threats. Russia's latest S-500 system
I'm puzzled why we haven't created something that can compete with S-300 / S-400 / S-500.
At the very least the Patriot SAM should be redesigned to allow for vertical launching to allow shooting in all directions. If you point Patriot one way and you get ambushed from behind you're basically caught with your pants down.
At the very least you'd need four Patriot missile launchers pointing in all directions to prevent this scenario.
> I'm puzzled why we haven't created something that can compete with S-300 / S-400 / S-500.
In my opinion it is because the American way of fighting a war doesn’t need it. Let say there is a troublesome country who would want to fly airplanes to shoot at American troops. The war plans against such a threat would start with two steps: 1st supress their air defences. 2nd gain full air dominance.
Supression of air defences is a polite way to say that they pulverize anything which can shoot US airplanes from the ground or even detect US airplanes using a radar.
Gaining full air dominance is a polite way to say that they aim to pulverize all enemy airplanes, runways, taxiways, fueling equipment and maintenance infrastructure.
If that is your method of operation why would you need mobile, survivable air defenses. People who routinely plan on destroying clouds (the source of rain) don’t need an umbrella.
That's the AEGIS system. It works better when land-based, from what I've read; it's dramatically more accurate than the purely naval fleet-defence version.
AEGIS is an integrated, networked system; multiple sensors on multiple platforms (including AWACS) collaborate to detect and hit targets.
There was an AEGIS-equipped ship either in the Black Sea, or trying to get in through the Bosphorus (now closed to naval traffic). The presence of AEGIS ships in the Black Sea would be enough to explain the non-appearance of the Russian airforce.
That's why it was funny that Turkey was ostracized for buying Russian/Chinese SAMs ( was it S-400 last or a Chinese version). What else could they possibly buy? The Patriot system isn't even close, and doesn't seem to be working too well for the UAE ( but then again we don't know how S-300/400/500 would deal against drones and light missiles)
My guess is that Russia doesn't have that many S-xxx missile systems, and further that they are much more expensive than the Turkish drones. I imagine they're keeping their powder dry, against the risk of NATO planes appearing over Ukraine.
I'm guessing most mechanical and/or ECE programs still teach "control systems", and that might be where one begins, along with some "signals and systems" and RF classes.
I suspect that the engineers you know are more computer science "engineers" than people who went through a traditional engineering core.
And there's a fair cohort of us (note, US-based perspective here) that _did_ go through the traditional electrical engineering education, learned the hard math and analog/mixed-signal engineering needed to put together something like this, dipped our toes in the industry, and then realized all our CS buddies had double our base salaries for objectively easier work.
We've strongly disincentivized gaining mid-level expertise on hard engineering.
> and then realized all our CS buddies had double our base salaries for objectively easier work.
One thing I realised as a graduate student designing RF PCBs and digital circuits was that the debugger for reality sucks. It's really frustrating to be faced with a problem where you just can't get "more information by recompiling with debug flags" and can't easily (or reversibly!) change the system to make it easier to understand. Totally different skillset, and requires a different method of thinking (frankly, at times, with a lot more thought).
This is a huge part of it! I spent a couple years doing IC-level failure analysis, which involves cool toys like electron microscopes, ion beams, strong destructive chemistry… a lot of it was neat, but the feedback loops were hours at best, and often days. Compared with having CI just do this for you automatically, or having something like `guard` running locally, is utterly transformational by comparison.
The ABET mechanical and electrical engineering curricula combined would give you all the skills a team needed (perhaps with the aid of a physicist and chemist or two) to build this system.
That we have dissolved the word "engineer" to mean "ad salesman at scale" doesn't mean the traditional engineers aren't out there.
I'm teaching in an ABET accredited program using MIT textbooks and I'm not impressed. If engineers want to avoid learning how to integrate by parts or use the chain rule it seems like a great idea. Arubis' comment above seems on point.
Are you saying that the current program lets them avoid learning those things, or that they should be allowed to avoid learning them?
I'm not sure I've seen a calculus course that skipped those (granted, it's been more than a decade since I looked at calculus courses). They certainly seem to still be within the remit of the FE - ECE exam.
I don't think this is a new problem. Its largely a problem of the way its taught, particularly at the high-school level where professional teachers and politicians choose what is important in the curriculum (and yes, I know a lot of engineering programs require an "engineering calculus" taught at the college). Nothing has changed since Feynman's little rant, if anything looking at my kids schooling its worse than ~30 years ago when I was in HS. Although, i'm told I can't compare my HS education because I went to one of the best HS's for math/science in the country (one of first of what are now called "STEM magnets" long before that term was even coined) at the time.
But even so, my calc for engineering classes suffered from basically the same problems of focus. It was frequently hard to tell what was really important vs what was an interesting sidetrack I would never use again. Which is why I distinctly remember a number of my engineering classes having week long math refreshers as the first week of class for things like Laplace transforms/etc or a math refresher TA lead study groups the prof would strongly suggest people attend at the beginning if they couldn't solve some basic math problems written on the board the first day of class. Where I went, the first week or so was add/drop and many of the professors would toss in a "prereq" quiz to scare people into dropping who weren't strong enough in the prereqs as another method. And being an engineering class grades weren't "given". For some classes it was well known that there were going to be a lot of C's and D's because the prof wasn't going to dumb down his material if you couldn't keep up because they were believers in normal distributions (and I remember a physics professor who would show the distribution for tests/etc) and that A's really meant you were excelling.
Unfortunately we are told that we have to make engineering “not scary” and “fun” so we get more engineering majors and this is explicitly not to be a weedout class. I do try to work them hard in the recitations to compensate.
The HN crowd is primarily software engineers and their curriculum probably doesn't cover this. But a EE can specialize in this field with courses in radar systems, signal processing, control theory, etc.
Something in that vein that has always amazed me, even though its performance in the field left something to be desired, is the Norden bombsight [0].
"The Norden [used] an analog computer that continuously recalculated the bomb's impact point based on changing flight conditions, and an autopilot that reacted quickly and accurately to changes in the wind or other effects."
Spacecraft often use stars to orient themselves. The Voyager spacecraft have star trackers to keep the antenna pointing at Earth. They look for the Canopus star.
I guess if they can confirm that it is a specific star they know where on a I finite line they are. If that line happens to be through earth and the space craft they know where to point their antenna.
GPS and celnav (and other forms of navigation, radio or otherwise) are really not so different when you get down to fundamentals. At their core they are all about locating the intersection of multiple lines, circles, or spheres of position, depending on how many dimensions you're working with. The major differences are in how you come up with the LOPs.
The odd man out, if you will, is inertial nav, which has more in common with dead reckoning.
Oh, I'm pretty sure the stuff we "web software pukes" have to work with would drive all those old-timer Real Engineers crazy in its own way. It's a bit like how a middle school teacher doesn't necessarily have an easier job than a prestigious professor.
Yeah digging through 2000 page reference manuals and looking for 1 bit differences with nothing other than a serial port/memory debugger is a lot harder than putting up a web page. Now backend stuff at scale is similar complexity. I still do embedded, but I also help/design on the web front ends for said devices :)
That's true. (Although what's with "old-timer"? Most of the mech/aero/etc. engineers I know are under 35.) I'm sort of the bridge between the software and hardware folks where I work, and nobody understands what anyone else does. Which is fine - that's what specialization is. The difference I find is that most hardware folks are at least aware of that, while almost all programmers I've met outside of aerospace don't even know the basic engineering curriculum exists (as exemplified by the comment that started this thread).
It's funny because we were brainstorming on the hoop the other night and it hit me that this was the killer app for a "cheap and cheerful" inertial navigation system.
The Trident 2 submarine missile has a gyroscope in it that costs almost as much as the nuclear warhead because it is so resistant to drift that it can be spun up and locked on in port and maintain its position when the submarine is at sea.
If you tried INS with a smart phone or wiimote the errors in velocity estimation will build up quickly and you'll get terrible drift.
In the case of the spatial computing hoop though you know the hoop never gets far away from the user so you can damp the D.C. component of the velocity, work in coordinates relative the user's center of gravity and never notice any drift in the x or y coordinates. (I don't like damping the vertical so much but I think I can live with it.)
You may be exaggerating but there is some truth in it, and it applies to many areas of science. The growing specialization means your knowledge is deeper but lest vast. For many classes of problems you just use a pre-built solution as this is the easiest and most common way. But rest assured, the people who work on defense system know very well what they are doing.
Heat seeking missiles started development in 1940s. In the US, they began as an unfunded rogue project. The infrared guidance system was simple and very successful.
One thing a lot of people forget because war movies don't show them, is that by the end of the war allied and German aircraft were absolutely loaded with electronics.
The story of Radar on the ground is well known, but the story of systems like https://en.m.wikipedia.org/wiki/H2S_(radar) (so named apparently because it was a smell that no one had discovered before) is not so well known.
Similarly a lot of these British efforts were more at risk to their own commanders that the Germans at the beginning, because the engineers were having to go through utterly ridiculous hoops to justify basic physics to their superiors. Sometimes their scientific superiors...
https://youtu.be/GJCF-Ufapu8 is a very very good documentary about this kind of thing. Made before the scientists were dead, so lots of gossip.
Randall and Boot should be known by more people, but magnetrons are not as awesome as the nuclear bomb.
The proximity fuses were a marvel - tiny radars put into an antiaircraft shell. Such would never ever have occurred to me, as how could a radar system survive being fired out of a cannon? It doubled the effectiveness of flak.
