When will it be able to land? This seems to me to be a simpler problem than having a car drive in traffic. True, carrier landing is very difficult for humans, but all the necessary pieces seem to already exist for a robot to do it. It’s my understanding that we can already automatically land drones on static runways, that they can deal with turbulence and bad weather. The only additional variable is the moving runway. With a couple of beacons at the end of the flight deck, I don’t see why a robot couldn’t home in on the arrestor cable. What am I missing here?
According to this Reuters Article: “It is scheduled to undergo two weeks of testing aboard the carrier leading up to a landing on the ship, in which a plane’s tailhook grabs a wire that will slow it and keep it from plunging overboard.”
Isn’t the answer that it technically isn’t a robot, since it requires a human controller/pilot flying it remotely, and therefore a carrier landing is rendered even more difficult due to a more “distant” POV for the pilot?
I would imagine for that very reason they would program landing procedures into the drone. Remote pilot gets the bird close and then hits the “autoland” button or whatever.
Missed edit: ALSO, a drone that thinks for and lands itself is more secure. If I were a bad guy I’d wait until the drone is vectored for a landing and then jam the controlling radio transmission turning it into a kimikaze drone.
One of this issues is that land-based runways tend not to move very much. Carriers are a different story, pitching and rolling and stuff.
There’s also a considerable challenge in getting an unmanned aircraft to operate on a carrier deck: F/A-18s have a pilot that can watch a sailor direct them to park over there, pull a u-turn, throw it into reverse, etc., all with the situational awareness of what else is happening around them. With a UAV, being able to move quickly and accurately around the deck without driving off the side of the boat is more of a challenge, if for no other reason than the operator may only see the world through a soda straw (perhaps a single camera on the UAV).
That would have to be a pretty complex piece of software. If this were true then why don’t they have current manned warplanes using this feature. I’ll WAG that due to the constantly changing nature of the mobile landing surface that software may not be failsafe enough to be implemented.
It’s easy enough to focus a transmitter and receiver to mitigate this. But more importantly, if there’s a transmitter close enough to a carrier to accomplish this then the people on the carrier have much much bigger problems.
As to the op’s question, the X47B was so successful at land tests they cut the tests short. Having a computer land a plane on a carrier will most likely exceed anything humans can do. It’s all about reacting to changes quickly which is something computers do better than humans.
. I have no idea what this says about me.