You're not a real programmer unless you could write an F16 autopilot program that can safely land on an aircraft

You're not a real programmer unless you could write an F16 autopilot program that can safely land on an aircraft.

Attached: INSVikramaditya.jpg (1200x800, 241K)

Other urls found in this thread:

en.wikipedia.org/wiki/Flight_dynamics_(fixed-wing_aircraft)
fr.wikipedia.org/wiki/Instrument_Landing_System
twitter.com/NSFWRedditGif

i'll write it in javascript

You're not a real shitposter if you don't make it the F35

Go back to your containment board

A flying aircraft or a stationary one? Both seem almost impossible to me from a pure physics standpoint.

Even humans can't get it right everytime.
Sometimes you hook doesn't catch the arresting cable

>not training a neural net to do it for you

Does the F16 have vertical landing? Otherwise I don't see how it could land on another aircraft

It is an interesting topic though. Now I am just wondering is it possible to achieve it alongside figuring out the number of variables we need to land a flying plane on a moving/stationary aircraft carrier.

>weather
>deck size?
>speed of plane
>speed of carrier
>alignment?

Maybe break it down to smaller problems?

OP never said aircraft carrier. I was talking about landing a plane on another plane.

Now do one to land by night with the carrier on minimal lighting

Start with state space control theory and its application to flight mechanics.

There are usually 9 state variables under consideration in mechanics of flight (velocities in 3 dimensions + 3 angles + the rate of change of the 3 angles) plus several input variables such as aileron/elevator angles. The matrices that are used to construct the state space representation of the system require extensive technical details of the aircraft which typically will not be available. However, once you get these matrices, optimal control theory can be applied to "land" the aircraft relatively easily assuming no random perturbations (e.g. like turbulence)

Incorporating random effects to create a "robust" landing algorithm is a far harder question to solve.

Give me 10000 carriers, about 1000000 planes, a huge open area where all the failed carriers will sink, and about a week

Attached: Artificial_neural_network.png (800x380, 19K)

en.wikipedia.org/wiki/Flight_dynamics_(fixed-wing_aircraft)

What about using machine learning? Just feed it thousands of images of the landing strip and also one label for the edge of the ship where you want it to land. Then have it break away and try again if the incoming speed is too high. Should work 80-90% of the time.

I had a CS assignment in college which was neat...

We were given real paper copies of an operations manual to the real NASA lunar lander from the 70's or 80's, with sections written about the math formulas governing constants, formulas, trajectories, etc... taking off, landing and more.

We wrote a computer program, filled with those constants and calculations and made it land a simulated craft.

>safely land on an aircraft
Kek

A
Fucking
Ramp

>real world applications
I'm not a code monkey, sorry

Define "safely"

You're not a real programmer if you lurk on this shitty site.

Thank you. Time to dust off my matrices/maths and get some reading done on the subject.

/thread

ez

append carrier to end of sentence?

No, on an aircraft :^)

Without any major fires

Does it have to be a shitty carrier like the pic or can it be an American carrier?

Only an Harrier could land on a plane

Hook up the flight path to a software PID controller, piece of cake.

They use remote guidance. It's more like the carrier lands the plane.

Can the fires be dealt with in a different function or do I have to avoid creating them in the first place?

Scala support it from the box.

import AircraftLander;
What next?

Now do it with without python and with only the mouse

what is CAT 3 standards.....

But the system its inherently stochastics. I think that the problem and, giving the multi-variable approach, will lead to some fuzzy logic controller as they are more robust. The first unknown variable for me will be inertia but can be studied and controlled over coriollis effect, but the wind and the 9 variables of it are what keeps me in the edge.

There are landing strips on aircraft?

there are 1.2 billion pajeets learning to code RIGHT NOW so they can move into your country

i'll make the logo

Yeah but like 0.5% of those are actually decent programmers.

Autonomous planes have already landed on carriers.

/thread

Yes but never on an aircraft

Link?

its wouldn't be that hard. i think the hardest part will be getting solid data from the airplanes/carriers horrible and very proprietary systems.

You don't need optimal control, just constantly adjusting. PID was already mentioned here, it should compensate for random errors and wind. We had a model plane using it that could adjust to weights dropped on its wings, it was a reduced case though, just a model in front of a blowing fan.

You're not a real programmer unless you can make a retarded looking aircraft fly own it's own.

Attached: boeing bird of prey.webm (640x360, 1.88M)

The F-16 didn't have a naval variant so I reckon the closest you can get to a safe landing is crashing into a safety net.

You're wrong.

The system would be pretty complex. For example it would have be smart enough to realize when to abort a landing, i.e. because of sudden strong crosswinds. And what if you have problems with the aircraft? Further complex decisions to make.

Never underestimate the human brain, it's awesome what people can do.

Nice try, Chinese spy.

if (closeEnoughToCarrier){
land();
}

>closeEnoughToCarrier is a boolean, not a method
>using polling

How could you fuck up two lines of code so badly..?

Where land() is a NPM-provided function?

>autonomous
That requires quite a bit of work. See picture.

Won't be allowed. You're probably going to have to use assembly, but there's a small chance for C.

Attached: milstd882esw.png (1665x1528, 243K)

>lands in water

>F16 land on an aircraft.
Do we have airborne aircraft carriers yet? Putting aside the launch/land shenanigans, I imagine the fuel required to keep it airborne has kept it infeasible for all this time. We would probably need something that creates upward force as a constant due to its existence as opposed to force created from a reaction.
The only things I can think of off the top of my head that would allow for that is are balloons and creating vacuum chambers inside the carrier filled with that less dense air.
A nuclear reactor might make it possible too, but there's obviously a bit of resistance to putting one of those high up in the atmosphere.

there was a proposal for an 747 aircraft carrier but it was abandoned because it had none of the benefits of actual carriers, was too expensive, and just served to move the "airstrip" a few hundred km inland which could easily be done by tankers anyway.

Is this Admiral huesos?

Planes did "land" on other planes before. Well, technically they docked: the US specifically created a program to develop a system that allowed planes to dock, shut down and restart the engine, and undock with the B-36 intercontinental bomber, so that they could carry their own escort far beyond the fighters' maximum range.

>what is ils approach
fr.wikipedia.org/wiki/Instrument_Landing_System

Attached: ILS-APPR.gif (384x179, 30K)

F16 can't land on any carrier it has not hook and is unable to do vertical landing

You have to ruin everything don't you

It's sad because the US military probably has enough budget to be able to afford to waste thousands of plains and carriers on something as stupid as that.

Just ask the hardware guys to make the carrier submersible so it can put out the fires by taking a dip underwater. Fires are a hardware problem anyway.