The smaller drones are not usually autonomous. See the Starlink alleged incident.
Inertial guidance is popular but very expensive to do accurately with laser gyros. I'm surprised there haven't been more "terrain following" systems.
There's probably always going to be a continuum between manned and unmanned platforms, and a discussion about SEAD.
Ukraine says it happened and Musk does too - is ‘alleged’ needed?
What other agenda could be going on here? They are both hiding some Russian capabilities from us?
For one, by making it harder to jam in the first place. Starlink with its extremely directional antennas is a good example - an opponent would need an equally massive fleet of satellites or high-altitude ECM planes to jam it, and the latter ones can easily be targeted by anti-radar rockets.
This is why the US government has been pushing insane amounts of money into SpaceX... Starlink is the future of interconnected wars.
Maybe, but what do you figure the US military will use?
The same way the F-35 does I guess. Besides, it's pretty common for AI systems to overtake human capabilities these days, so when jammed they can just carry on.
Sure, they do mistakes but humans do these too and the advantage of not carrying 80kg of fragile human and all the life support systems onboard is quite significant. It makes the thing much cheaper, it removes the need to come back thus doubles the range, it makes the thing smaller thus harder to detect and destroy, it doesn't have to limit its manoeuvres to human levels this makes the thing much more agile.
No, because that's in part "let the human pilot make decisions".
> Besides, it's pretty common for AI systems to overtake human capabilities these days...
Not in the realm of "should I shoot that thing?" sort of decisions.
It's not like pilots are making political decisions. They pilot and shoot predefined targets, avoid hostile actions. AI is capable of doing this.
> Not in the realm of "should I shoot that thing?" sort of decisions.
On the contrary, AI is very capable of making that decision. There are no philosophical dilemmas or children in the skies and even if there were we are at the point where we can tell the device not shoot children. There will be mistakes but human pilots makes mistakes too.
I would not currently trust an AI to handle those very well.
What would an AI have done in this situation? What should it have done? "Russian pilot deliberately fired missiles at a Royal Air Force surveillance plane in international airspace over the Black Sea last year": https://apnews.com/article/uk-russia-fighter-jet-missile-bla...
What makes you think that AI can't incorporate those into decision making? Pilots do these through instruments anyway.
The fact that state-of-the-art AI already fails at much simpler decisions.
In the case of the Black Sea incident, the potential consequences include global thermonuclear war.
It's always hilarious to see the confidently incorrect comments by a bunch of ignorant software developers. The Dunning–Kruger effect is on full display here.
Nothing is harder than self-driving cars. Nothing. We'll colonize Mars before we have a solid solution to that problem. Why? Self-driving cars have to coexist with human drivers and human infrastructure. Nobody in aerospace has that problem. If they did, they'd run screaming for the hills.
That simply doesn't happen when I fly my quads. "How would I write code to dodge an attacking drone? How would I modify my drone to drop a grenade or a Molotov cocktail, or otherwise cause a large amount of grief to people below? How would I build a SLAM model that allows the drone to do this without intervention from the ground?" None of these engineering problems bug me the way driving a car would. They are all addressable with multiple degrees of freedom, both literally and figuratively.
Meanwhile, on the road:
"Hmm, the light at this intersection is out. There's a cop with an angry look on his face, flapping his arms at me like a dying chicken. What does he want me to do, exactly?"
"Huh, here I am in Seattle, and it looks like they have chosen to mark the stripes on the road with some sort of paint whose complex impedance at optical frequencies is identical to that of rainwater. I'm sure glad I'm driving, and not my lane-keep assistant, which I had to turn off because it tried to steer me into the median the last time it snowed."
"Whoa, where'd that ambulance come from. The law says I have to move right, but the only way I can get out of his way is to move left, and in any case, that's what the car ahead of me is doing. What to do, what to do."
In most of the airborne scenarios you mention, doing nothing is a perfectly viable, fail-safe solution when confronted with a situation the hardware or software can't handle. If we approach driving that way, a few miscreants can brick an entire city, intentionally or otherwise.
I'm not surprised Karpathy tapped out, let's put it that way.
Sixth-generation tactical aircraft (the successors to the F-35) are likely to be optionally manned. They will be able to operate with remote pilots and/or autonomous control for high risk strike missions but most of the time will still have human crews on board.
In the age of the Strategic Corporal, they absolutely are.
I'm not sure how accurate that is historically.
Canada's first losses in a combat zone since Korean War.
> "Let's just make sure that it's, that it's not friendlies, is all"
> Twenty-two seconds later, he reported a direct hit. Ten seconds later, the controller ordered the pilots to disengage, saying the forces on the ground were "friendlies Kandahar".
There's little evidence autonomous combat fighter AIs are better than humans at tough calls of this nature. They may be someday, but given the state of the art in self-driving, that day probably hasn't arrived.
AESA derivatives as bidirectional communication devices seem like they will render jamming a lot less effective. Simply by virtue of being able to pump radar levels of power into communication.
And current trends seem to be converging on flocks of drones, of which one or two can be specialized with uplink. Or simply babysat by stealthy HALE platforms like the RQ-180.
You'll have to blanket an area with ungodly amounts of energy to fully jam point-to-point, highly directional links, especially for close range hops.
You say that as if it isn't incredibly easy to do to the point we have entire enforcement orgs built around trying to keep people from unintentiinally doing just that.
Of course whether the F-35 platform makes sense for that role is a different question. There are probably great niches for a drone F-35 (e.g. targeting anti-air installations), but Ukraine shows that having lots of $1000-$100,000 drones might be more valuable on the battlefield than one $75,000,000 drone.
A big reason why the US will continue to prefer drones measured in tons instead of grams is that cheap drones are most useful when you have boots on the ground, which the US likes to avoid. But with the budgets available it's not like they have to choose between F-35 sized drones and Dji Mini sized drones, they can just get both.
Those drones seem to look more like cheaper loitering weapons platforms than F35s but who knows what happens when the other side isn't so stupid.
I highly doubt it. At least I hope so.
Nobody dreams of becoming a drone pilot. At the end of the day it's a cubicle job.
Maybe they view it a a job with a lot of the upside but none of the safety downsides of rocketing around the earth at 1000 miles an hour half a world away from home...
Which the US is wholly aware of given the "Loyal Wingman" program, which the F-35 is designed to work with: commanding unmanned fighters/aircraft as support or missile trucks if what have you
The closest analog would be converting cessnas. Half a million dollars plus drone kit = 1 million dollar bomber => 75 bombers or one F-35.
At that price point one might start thinking about sending the bombers first (to deplete air defenses), then the F-35s...
And even aside from that, self-driving cars still regularly are observed to make stupid mistakes. When warplanes start making stupid mistakes, the consequences are going to be a lot more dire.
'Advancements in AI' have reached a point where they can affect everything from your smartphone to a box of cereal.
With the advancements in AI in recent years, it will probably get harder and harder to justify carrying a biological being and all the support systems onboard. At some point, a warplane capabilities might become irrelevant and the only important aspect would the the implications of carrying a solder onboard and making the downing of the device much more politically significant than downing a drone.