Navy Drone Successfully Lands on Aircraft Carrier | Hacker News Reader

Navy Drone Successfully Lands on Aircraft Carrier

blogs.wsj.com

76 pointsmelling32 comments

Calling it a 'drone' kind of understates the accomplishment here. :) Navy drones have already taken off from and been recovered onto ships, in the VTOL configuration.

This is essentially an unmanned jet aircraft. It recently successfully managed a catapult-assisted takeoff from a carrier.

Now it has completed the other end of the deal, an arrested landing. (i.e. it had to catch its tailhook on one of the arresting wires, execute a proper glide path descent onto a moving platform, etc.)

While this is an accomplishment, I'm much more impressed by difficult manned arrested landings like these:

https://www.youtube.com/watch?v=4gGMI8d3vLs

But it is just a matter of time before that type of landing is part of the historical archive.

Not only is this an unmanned aircraft, it is an autonomous (as opposed to remotely piloted) aircraft. This means that instead of a human controlling the landing, the aircraft was instructed to land and executed the descent and landing on its own.

Anyone who's been involved in carrier operations will tell you that landing a jet on the deck is non-trivial even with experience. The fact that an aircraft can do this autonomously is quite impressive.

I have to disagree. You are being fooled into thinking that is a stealth fighter or bomber. Its not.

It's a drone which is at mini-scale to the actual aircraft you may be referring to.

Yes. We have stealth bomber looking drones at 1/3 scale to the actual thing.

And to your original point, this is obviously not a first, drones have been taking off and landing from aircraft carriers for years.

Why the washington post is reporting this now is a better question. They are clearly out of the loop.

> drones have been taking off and landing from aircraft carriers for years

Fixed wing? Such as?

An unmanned jet is a drone, just like the D-21 which flew in 1960s. The big deal with the X-47B is that it landed autonomously and not remotely piloted.

I really don't think Congress is exercising its budgetary oversight correctly if it is approving nearly trillion dollar purchases of manned fighter aircraft to be delivered a decade or more in the future. We're pretty clearly within a transition from manned to unmanned combat aircraft, and given the lack of compelling threats flying top-end fighter aircraft (or fielding top AA defenses), there's no reason we couldn't keep the F-16/F-15/F-18 in service (and whatever F-22s we already purchased) for another 10-20 years, then replace entirely with drones.

Drones aren't just cheaper, they're better. Smaller, higher G loading, less radar cross-section, no need for SAR when one goes down, potential for "in extremis" one-way missions (intentionally sacrificing a $5mm reusable drone to accomplish a mission is often fine), etc. The only thing they don't do is create as many senior officers with flight experience.

(They also don't cost as much money, so to the extent you believe every dollar spent on defense, needed or not, is a dollar stolen from humanity, ...)

There is one thing a human can do that these new fancy drones can't yet. Look out the window and identify if a target is hostile or not. My understanding (I used to be in Naval flight training) is that this is still a huge part of the work done by pilots that these drones don't handle well.

It previously struck me that an advantage of drones is that the pilots aren't under the time / personal danger pressure of needing to make snap decisions. In practice this turns out not to be the case, and -- as you say -- human pilots may make better judgment calls.

If that is an issue surely the answer is more assisted drones. Better to have someone sitting in safety viewing a high def and possibly magnified camera output to identify things that someone taking a quick look in the cockpit while under fire.

Latency and mission. If you don't have line of sight with the drone, you've got multisecond latency on input, that's not suitable for fighter aircraft, though is not deal breaking for recon and air-to-ground (bombs or missiles). So fighter drones require the pilots to have line of sight to be effective in combat, which means you need control of ground or sea assets within range of the air combat theater. That's not a given in war. Though, at present, this is probably a given for the current activities of the US DOD, you can't base future planning on current conditions.

That said, F-35 is a clusterfuck. Military contractors have had so little oversight, and are such terrible planners, (particularly of software) that most projects that depend on a significant software component are guaranteed to have cost and time overruns, with little repercussion except for the poor program manager that was thrown into the deep end and drowned.

NCTR (Non-cooperative target recognition) has advanced rapidly to the point that AESA radars can provide the geometry of a locked target and identify the type of aircraft BVR. There's also old-fashioned IFF.

That seems like something which matters only in two situations:

1) Normal US style war, where we have complete domination of the skies and airwaves. No problem using high bandwidth comms back to base to get video in front of lawyers and flag officers to decide on the merits of a given target. False positives way worse than false negatives. (Also, for this kind of mission, even an F-18 is overkill; you could probably fly refitted F-4 or F-16 when the enemy is potentially a civilian airliner or whatever.)

2) "Real" war, where losing is a possibility (vs. just not winning). Comms may be disrupted. In that environment, the drone needs to be autonomous, but could make IFF decisions more loosely; within a bounding box, it's reasonable to assume any military aircraft which doesn't IFF as friend is in fact the enemy and can be shot down with no additional confirmation. Since the US prefers to engage BVR in that kind of situation, the drone has every bit as much data (from radar, emissions, etc.) as a manned pilot. False negatives and false positives are more comparable (maybe 10:1 vs. 10000:1 in severity).

Drones are the future for a lot of missions, but considering the amount of money thrown around, no Congressman is going to turn the opportunity to secure billions in defence spending for their state/district.

Defence spending has always been driven by domestic pressures - anyone remember the 'scandal' when Airbus (with a better plane) secured a contract for US transport airlift?

I don't see why these need to take off and land horizontally. Give them the capability to take off and land on their tails, and the size of the carrier can shrink dramatically. The reason why we don't use tail sitting craft is because they are too awkward for human pilots to land. Drones wouldn't be so limited.

The reason you don't use tail sitters is because they have much less payload capacity. In that configuration the engine has to be able to output thrust greater than the entire weight of the aircraft. Something like the 737 has a thrust to weight ratio of only 0.25 (you need greater than 1 to take off vertically).

A more direct comparison: A fully loaded standard F-35 has a thrust to weight ratio of 0.87 (that is with the afterburners on). In comparison the short take of and landing version of the F-35 loses 43% payload of its and 25% of it's fuel capacity (and thus range) [1].

1 - http://en.wikipedia.org/wiki/Lockheed_Martin_F-35_Lightning_...

> The reason you don't use tail sitters is because they have much less payload capacity.

I wasn't thinking of abandoning horizontal flight entirely. Being able to land a drone aircraft after it has spent its munitions and fuel should be handy.

> "The reason why we don't use tail sitting craft is because they are too awkward for human pilots to land."

This isn't true. Taking off anything vertically requires their thrust-to-weight ratios be higher than 1.0 (otherwise try as it might, it's going to sit on the ground).

Modern, latest-generation fighter aircraft have a TWR somewhere between 0.75 and 1.15. If you pointed them upwards, they will barely be able to lift off the ground (if their TWR is above 1.0 at all), much less gain enough speed to generate lift with their wings.

Taking off horizontally allows you to get lift "for free" with the wings. Accelerating horizontally means you're not fighting gravity the whole time. Lifting off vertically, assuming the aircraft has enough TWR to make it realistic, is a terrible waste of fuel. Modern fighters only have a combat radius of ~300-350mi before they have to turn around and go home. The more fuel that is expended taking off or landing, the less actual flight (and fight) time they have.

Catapult launches also allows you to save fuel on the aircraft by "borrowing" the necessary energy from the launcher. Ditto arrested-hook landings that allow you to borrow stopping force from the aircraft carrier - otherwise you'd have to carry enough fuel to exert this force yourself.

It's certainly possible to design craft that can take off and land vertically - the physics are relatively simple. The problem is making them practical - the more fuel/weight an aircraft spends just taking off and landing, the less weight can be dedicated to useful tasks, like missiles, bombs, and flying far.

The real reason generating lift with large wings is better is that it's much more efficient to generate a force by accelerating a large volume of fluid a small amount, than vice versa as with a turbine engine. This is why helicopters exist at all, because of the efficiency of large rotors (wings). Ref http://en.wikipedia.org/wiki/File:VTOL_DiscLoad-LiftEfficien... (note log scale).

> Modern, latest-generation fighter aircraft have a TWR somewhere between 0.75 and 1.15. If you pointed them upwards, they will barely be able to lift off the ground (if their TWR is above 1.0 at all), much less gain enough speed to generate lift with their wings.

Taking off horizontally allows you to get lift "for free" with the wings.

Well, more specifically, you are using the atmosphere as reaction mass. Wings direct air down. It's not free lift, it's free reaction mass.

I'm not saying that you should eliminate catapult launches. But a ship large enough to have catapults can still be a whole lot smaller than a modern US Navy aircraft carrier. Any aircraft that can land vertically will also be able to take off vertically. (Though with less efficiency when fully loaded, if at all.)

There's a huge amount to be gained by being able to stage aircraft from much smaller ships. How about building submarines a bit larger than the old Soviet Typhoons, and having aircraft on those? Short of that, how about the ability to build squadrons of multiple redundant cruiser-sized carriers?

Perhaps such schemes would be more valuable potential opponents of the US, than they would be to the US.

Yes, on the whole, the system will be less fuel efficient than horizontal take off and landing.

> There's a huge amount to be gained by being able to stage aircraft from much smaller ships.

These are called amphibious assault ships like the America and Tarawa classes. They can serve helicopters and V/STOL fixed-wing aircraft. No catapults.

http://en.wikipedia.org/wiki/America-class_amphibious_assaul...

http://en.wikipedia.org/wiki/File:USS_Saipan_LHA-2_amphibiou...

I suspect you may see submarines configured to launch drones vertically with rocket assist and recover them using something smaller than a flight deck. Subs already deploy SEAL teams and there has been at minimum experimental use of remote piloted minisubs — this would be a perfect complement.

This heralds the end of the aircraft carrier (a good thing, since they're really quite obsolete conceptually).

It's actually more useful to have larger carriers than smaller ones. The obvious difference is top speed as drag does not increase linearly with length. But you also gain the ability to do takeoffs and landings in a wider range of weather conditions as well as having redundant nuclear reactors and dentists etc. But, perhapse most importantly the larger the aircraft carrier the larger aircraft it can support and there are many advantages to having larger aircraft.

Really? Larger aircraft have a larger cross-section for enemy radar and missiles, and are likely to have worse thrust/weight ratios and maneuverability (square-cube law). Combat range and ammo capacity scales with size up to a point, but if you can make a smaller drone with the same range and armament as a normal fighter then it would surely be more useful in most respects.

Your don't need to take my word for it both carriers and modern fighter craft have continued to get larger over time. http://en.wikipedia.org/wiki/Supercarrier and the next gen the Navy is constructing is not getting any smaller. But, don't forget the navy uses a lot of non steath aircraft for logistics and other thing to the point where they seriously considered building a floating 2 mile runway (http://en.wikipedia.org/wiki/Mobile_offshore_base).

As to why it's useful for combat aircraft to be larger. First radar cross sections have little to do with the overall size of the craft it's mostly a question of shape an material assuming everything is staying within the same order of magnitude. Second while strait up thrust/weight is useful, top speed, range, and munitions capacity are generally more important because if your fast enough you can bank at over 10g's which will cause you to pass out long before the aircraft is in trouble. So while maned aircraft are still limited by the pilots all of the rest of those tend to improve as you scale up aircraft size.

PS: In many ways Cost and the desire for redundancy do limit the size of both aircraft carriers and combat aircraft.

> Navy drones fly autonomously with the planes’ robot brain making the split second calculations necessary to conduct an arrested landing on the deck of a moving ship.

Is anyone else a little freaked out that this is one more step towards aerial ED-209's?

Me. A combination of freaked out, and feeling silly for being freaked out.

Somehow, aerial drones seem like less of a stretch than ground drones. I'm surprised that tanks aren't being replaced by drone tanks. All the same reasoning applies. I think the first time land battles happen between drones, we'll get the message.

I would think that it'd be easier to mess with communications to and from the tank. Or that they'd get stuck in a valley or something where radio communications aren't very good. Also, navigating on the ground is trickier than flying in some ways, depending on the terrain.

They've also got drones tested for mid-air refueling. Work is being done on both having the drones hook up to be refueled, and for drone refuelers. IIRC, the last test got them flying in formation correctly, but they didn't actually connect the hose.

That's a big drone. I expected a drone to be smaller but that has the size of full brown aircraft.

Your move, Skynet.

That is AWESOME! It's a shame we won't get to see the code for it anytime soon.

Reply on news.ycombinator.com