Elon Musk Predicts F-35s Would Have 'No Chance' Against Drone Fighter Planes (businessinsider.com) 298
"Tesla CEO Elon Musk suggested that Lockheed Martin's F-35 Lightning II, the costly stealth jet considered to be pinnacle of U.S. military aviation, 'would have no chance' if pitted against a drone [fighter plane] that is remotely piloted by a human," writes Business Insider.
Long-time Slashdot reader DeadlyBattleRobot shares their report: At the U.S. Air Force's Air Warfare Symposium in Florida, Musk said there should be a competitor to the F-35 program, according to a tweet by Lee Hudson, the Pentagon editor at Aviation Week... The Air Force conference at which Musk made his comments included senior US military officials and pilots. Speaking with Space and Missile Systems Center Commander Lt. Gen. John Thompson, Musk said autonomous drone warfare "is where it's at" and "where the future will be," according to Defense News.
"It's not that I want the future to be this. That's just what the future will be," Musk added. "The fighter jet era has passed. Yeah, the fighter jet era has passed. It's drones."
Long-time Slashdot reader DeadlyBattleRobot shares their report: At the U.S. Air Force's Air Warfare Symposium in Florida, Musk said there should be a competitor to the F-35 program, according to a tweet by Lee Hudson, the Pentagon editor at Aviation Week... The Air Force conference at which Musk made his comments included senior US military officials and pilots. Speaking with Space and Missile Systems Center Commander Lt. Gen. John Thompson, Musk said autonomous drone warfare "is where it's at" and "where the future will be," according to Defense News.
"It's not that I want the future to be this. That's just what the future will be," Musk added. "The fighter jet era has passed. Yeah, the fighter jet era has passed. It's drones."
No chance at all (Score:5, Insightful)
Unless the F35 had some form of jamming equipment. I guess it's lucky for remotely controlled drones that no-one has invented such a thing.
Re: No chance at all (Score:5, Insightful)
Exactly. All the maneuvering superiority is pointless if you canâ(TM)t communicate with the thing.
And, even if comms were not jammed, you still need to be able to âoeseeâ the F-35 to engage it. The only time they would have an advantage is if the drone can get a visual or thermal lock.
Now, if they solve the communication and observability problems, a swarm of heavily armed stealth drones would be frightening indeed.
Re: (Score:2)
"And, even if comms were not jammed, you still need to be able to âoeseeâ the F-35 to engage it. "
'It'? They are cheap, you'll send a dozen.
Re: No chance at all (Score:5, Insightful)
But:
Just last week I watched an episode of Dogfights. One flight that was showcased was an Isreali ace (Giora Epstein) who at one time was outnumbered 10 to 1 in a dogfight (his buddies had bugged out) and was flying an airplane that was inferior to what he was fighting against. Yet, if I remember correctly, he downed 5 enemy aircraft and almost got a 6th.
I worry more about the possibility of mass produced (cheaper but competent) enemy aircraft with well trained pilots.
Re: (Score:2)
Yet, if I remember correctly, he downed 5 enemy aircraft and almost got a 6th.
It is called: target rich environment
In computer games I used to do that all the time. Lucky we played in LANs so everyone saw I was not cheating.
Obviously RL is a huge difference, especially the required battle awareness.
Re: (Score:3)
He could see enemy planes twice as far out as they could see him. He had amazing situational awareness. He knew now to get the most out of his aircraft He was patient and waited until is enemy made a mistake.
Interestingly, those are all things that drone AI's are excellent at compared to humans.
Re: (Score:3)
I suspect the dividing line between missiles and drones is going to get blurry. Drone electronics aren't any more expensive than missile guidance systems. Mass-produced drones don't need to carry multiple missiles, so why not just put in a warhead? This will make the thing smaller and stealthier too. The attack on the Saudi oil facilities last year used drones as cruise missiles. A smaller model could be used like a SAM without giving away ground positions, especially if it can loiter in the air waiting for
Re: (Score:2)
"And, even if comms were not jammed, you still need to be able to âoeseeâ the F-35 to engage it. The only time they would have an advantage is if the drone can get a visual or thermal lock."
While the Eyeball Mk-I is still popular, there was this invention back in WWII, something about using radio waves to locate a target at night ...
Seriously, we are currently in the era of missile combat. Radar lock is obtained at huge distances. The AIM-120D (current most popular long-range air-to-air missile)
Re: (Score:2)
What's the problem with seeing it? They're only stealthed from the front and to a much lesser extent the side and rear, the bottom and top have almost no protection. They've known since the tail end of the '90s that the path of a "stealth" aircraft flying through a cellphone network can be detected by analyzing the towers' own intranetwork communications. Once the drone has been directed close enough to establish visual and/or thermal contact the flying money-pit is dead.
Re: (Score:2)
When a stealth fighter opens its missile bay it glows up on radar like a christmas tree.
The only point where an F-35 and others have an advantage is that they are integrated into an Awacs or Hawk-Eye command center. So a standard drone would not come close to them.
And don't forget: the Airforce is already experimenting with drones as wingmen, being one or two hundred kilometers away from the controlling aircraft.
They won't be remote controlled anyway. The remote operator would basically only confirm targets
Re: (Score:2)
Re: (Score:2)
Why not use an AI? AI's controling aircrafts already exist for fighter games.
Because the real world isn't a video game, and when your AI shoots down an airliner full of passengers, it makes people angry.
Re: (Score:2)
"Because the real world isn't a video game, and when your AI shoots down an airliner full of passengers, it makes people angry."
The AI doesn't care.
Re: (Score:3)
Call it 'pilot error', the same as the other 20+ airliners that have been shot down over the years.
Re: No chance at all (Score:5, Funny)
I find it hard to imagine that Musk wasn't thinking of AI for this scenario, given how much work Tesla is putting into its car AIs.
Tesla CEO Elon Musk suggested that Lockheed Martin's F-35 Lightning II, the costly stealth jet considered to be pinnacle of U.S. military aviation, 'would have no chance' if pitted against a drone [fighter plane] that is remotely piloted by a human", writes Business Insider.
Re: (Score:3)
I sincerely doubt the military, and even Lockheed, considers the F-35 the "pinnacle of U.S. military aviation." The F-35 is what you get when as a cost-cutting measure, Congress requires the USAF, Navy, and Marine Corps to buy the same aircraft for aerial dogfighting, bombing, carrier takeoff and landing, VTOL, and close air support missions. The thing is a frankenstein of conflicting miss
Re: No chance at all (Score:2)
You do realize they are not just radar stealth but themal as well right? They minimized the thermal plume cross section by changing the shape and design on the gen-1 F-117. Otherwise its not as great a stealth design.
Re: (Score:2)
"You do realize they are not just radar stealth but themal as well right?"
But not much stealth if you are 200 yards behind it.
Re: (Score:2)
"Really? At supersonic speeds, do you really think you're relying on eyeballs or cameras to find a target?"
360 degree cameras have been there for a loooong time.
Re: (Score:2)
Re: (Score:2)
It's not stealthed at supersonic speeds, the shock wave can be detected even from orbit.
Re: (Score:3, Insightful)
Unless the F35 had some form of jamming equipment.
(facepalm)
Yep, the whole point of the F35 "stealth" program was to build an aircraft that continuously broadcasts its position via a wide spectrum radio signal.
Clue: A "jamming" signal is incredibly easy for a missile to lock onto.
Re: (Score:2, Interesting)
You might like to face palm yourself, you are not as clever as you think you are.
1. The drone control signal is also easy for a missile to lock onto.
2. You don't go active with jamming until something locks onto you. At which point stealth is irrelevant.
Sometimes, someone knows more than you.
Re: (Score:2)
You might like to face palm yourself, you are not as clever as you think you are.
1. The drone control signal is also easy for a missile to lock onto.
2. You don't go active with jamming until something locks onto you. At which point stealth is irrelevant.
Sometimes, someone knows more than you.
You haven't thought that all the way through.
If your F35 starts broadcasting it's position whenever it detects a certain signal then I just flood the sky with that signal.
(followed by missiles).
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re:No chance at all (Score:4, Interesting)
Early in WW2 the Germans utilizes radio signals to pinpoint bombing of British cities. It was remarkably effective in comparison to bombing done by the RAF which was based on aerial reconnaissance. The British figured out the German systems (Knickebein, X-Gerät and Y-Gerät) and the jamming was so effective that the Luftwaffe bombers ended up bombing well off target in areas where no damage was sustained. The Most Secret War by R.V. Jones is a great summary of those efforts and it details how German methods of operation made it much easier for the British to predict, detect, and counter those systems. The British had Y-Gerät solved for the first time the Germans used it.
Re: (Score:2)
Jammers have off switches. The F35 has a pretty sophisticated (active) electronic warfare suite.
https://www.f35.com/about/capa... [f35.com]
Thanks for Making It Even Easier (Score:2)
Yep, the whole point of the F35 "stealth" program was to build an aircraft that continuously broadcasts its position via a wide spectrum radio signal. Clue: A "jamming" signal is incredibly easy for a missile to lock onto.
Wouldn't that also mean that the continuous, high-bandwidth data stream that the drone has to send back to its human pilot so they can see what they are doing would also be "incredibly easy for a missile to lock onto"? I guess jamming is not needed, just one radio-seeking air-to-air missile.
Re: (Score:2)
Unless the F35 had some form of jamming equipment. I guess it's lucky for remotely controlled drones that no-one has invented such a thing.
You've heard of infra-red right. As in heatseekers. Not jammable. Also the F-35 is a bag of shit. It's not stealth, it's stealthy ish from some angles and only when it's not carrying external stores (ie when it has close to zero offensive power). On top of that it has tiny little shitty stubby wings that won't allow it to manoever nearly as well as it's competitors. If the thing gets in a tangle with anything its pretty much fucked. 1 viper with a visual would rip it apart with juns. No missiles needed.
Re: (Score:2)
Re: (Score:2)
You do realize that a cruise missile is an AI-controlled drone which completes its attack without the guidance of a human, which just happens to also be the missile, right? A fairly stupid AI, but the things have been around since the 1970s and are responsible for the deaths of hundreds of thousands. That horse left the barn a long time ago.
Re: (Score:3)
"You've heard of infra-red right. As in heatseekers. Not jammable."
https://www.military.com/equip... [military.com]
Re: (Score:2)
LOL I was about to post these
https://en.wikipedia.org/wiki/... [wikipedia.org]
It's really amazing seeing people like stealth_finger who delude themselves into thinking they know shit...
Re: (Score:3)
I love Slashdot.
Poster #1: "this doesn't exist."
Poster #2: "here's an example of it existing."
Poster #3: "but but but something else."
Re: (Score:2)
Re: (Score:2)
Considering that most aircraft kills are done a long way off, and the F35 is SPECIFICALLY designed for over the horizon kills and not in close dog-fighting, "jamming" the signals from many km away might be a bit difficult.
The autonomous part, which I assume kicks in on signal loss, is the tricky part. The pilot is more there for accountability than anything else, indeed I would imagine much of the modern jets systems are pretty automated already.
Re: (Score:2)
Good luck jamming a laser targeted at a satellite or ground station
I doubt it (Score:4, Informative)
Re:I doubt it (Score:4, Insightful)
"Remote control" in this context obviously means that a human controller tells the drone *what* to do, not *how* to do it, then it can execute those instruction with digital speeds, a human can't compete with that. The drone will also be able to execute maneuvers that a human could never do.
Re:I doubt it (Score:5, Informative)
The problem though is how fast can the information needed to make the decision and communicate it be made? Latency is a HUGE deal here.
IF we are discussing dog fights, then the drone loses almost hands down due to latency over a human in the aircraft opponent. Unless you get the missile shot off as they come into range and assuming nobody decides to disengage because they are at a disadvantage, you are going to quickly be in a low speed turning fight. In this situation, the winner is generally the one who can turn (out maneuver) the other. Fighter pilots say it this way "Rate kills" if you can turn faster than your opponent, you win. Drones would be at a distinct disadvantage here due to the communications latency. A human pilot need only change something in the turn and his opponent won't be able to react to that change as quickly, effectively lowering the turn rate.
Don't fool yourself into thinking that some AI can respond faster either. The sensing required and the processing power required to make such decisions is way too big, heavy and power hungry.
Re: (Score:2, Interesting)
Re: (Score:2)
There is no way a human pilot can withstand the G-Force that a drone piloted craft could. Having no pilot makes the fighter aircraft orders of magnitude more dangerous. Add AI into this mix and human remote operators will just be there for the admin
Yes, but... As I discuss elsewhere.. High G turns are not usually how a dogfight is won. High G maneuvers are only going to happen at the beginning of a fight, when energies and airspeeds are high and each combatant is deciding if they wish to join the fight or not. Many times the fight ends with a stand off missile shot followed by a high G turn to get away. Being a drone able to sustain higher G loads doesn't change this much.
Once you get into close quarters, combat is not about pulling G's but how fa
Re: (Score:3)
combat is not about pulling G's but how fast you can turn.
I am confused: How is "how fast you can turn" not exactly the same as "pulling G's?"
Re: (Score:2)
At dog fight distance It could go visual on the target (sensor integration like a self driving car would allow it to have better battlefield awareness). It doesn't need to survive, just crash into it.
A drone is n
Re: (Score:2)
Re:I doubt it (Score:5, Insightful)
How remotely? As far as I know, for every 300 KM the remote pilot will have to deal with 1 second of latency, one second of unknown is a lot.
It's radio waves so they travel at 300,000 km/s not 300...
Re:I doubt it (Score:4, Interesting)
Re: (Score:2)
Re: (Score:2)
The whole idea of the "loyal wingman" concept is to combine the best of both worlds -- have a human in a fighter close enough to be immune to jamming that commands a squadron of drones flying with him.
They won't be cheap and easily replaceable though. There's only so much cost/weight savings gained by ripping out the human life support, and while pilot training is expensive it's still nowhere near the cost of the airframe. The reason drones are seen as "cheap" is that the ones we use today mostly are low-pa
Re: (Score:2)
The only other option is to make them as cheap as possible knowing they'll get shot out of the sky, with the goal being to overwhelm with numbers, but that just puts you in a war of industrial capacity.
...and the F35 is revealed for what it always was: An expensive boondoggle.
Re: (Score:2)
Well, the F-35B did turn out to be the best STOVL fighter ever made, so there's that. The Marines did get a big win for their flattops. Pity the Air Force and Navy have to pay for it.
Re: (Score:2)
It is well known, and companies must be working on this for a long time. I don't know why we need Musk to say this.
Re: I doubt it (Score:2)
300000km, you know. The earth moon distance.
Roundtrip time latency is double that.
Re: (Score:2)
He is Armchair Fleet Marshal of the Space Force, his calculations make sense.
Re: (Score:2)
You're off by a bit. Light travels at 300,000 km/s.
Re: (Score:2)
Re: I doubt it (Score:2)
Theres a simple way to solve that: motherships. Many air forces have aerial radar platforms likes the US AWACS or E-2 Hawkeye that can direct fighters to targets. You simply have another aerial platform with control stations orbiting at a safe distance to control the unmanned fighters. The drones themselves can be land or carrier based.
Re: (Score:2)
Removing the human limitations... (Score:2)
Piloted planes are limited by the pilot's fleshy limitations.
A drone plane can do more extreme manoeuvres than a normal plane - that will give it an advantage. Also the pilot support systems take up space, space that could be used for ammunition or other features you can't fit on a normal plane.
On board tracking, targeting AIs might also give it a responsiveness advantage - the remotely piloted drone has a latency disadvantage.
I guess when a drone enters active situations, you can also have more than one pe
Re: (Score:3)
You're still thinking of the drone as a single unit though.
Once you have ten smaller, more maneuverable, cheaper drones all coordinating their movements, a human piloted jet loses all value, they'll come from all sides, close off all your possible options.
There's no need to have humans piloting or targeting, a computer can calculate trajectories much better, you just have humans around to define broader goals and maybe confirm or deny kill decisions.
The only reason humans still have any chance against AI li
Re: (Score:2)
Autonomous Drones? (Score:2)
Elon's speech seemed to say that he think that the F-35's replacement will be a remotely controlled drone with some autonomous assistance.
I think that we're pretty far away from having a fully auto piloted drone, though. If Elon thinks that the PR blowback of a Tesla on autopilot crashing into someone is big, just wait to see what happens when an armed drone accidentally blows up some civilian bystanders because it's sensors got confused during a dogfight.
Re:Autonomous Drones? (Score:4, Informative)
Probably the same thing that happens when a human-piloted jet bombs some civilians based on bad intel, they'll call it collateral damage and move on.
Re: (Score:3)
when an armed drone accidentally blows up some civilian bystanders
Considering Russia is deliberately bombing hospitals in Syria [nytimes.com], and has been doing so since the day it started helping the Syrian regime, and the world doesn't even raise an eyebrow, I don't think a drone accidentally hitting some civilians will be that big a deal.
Re: (Score:2)
when an armed drone accidentally blows up some civilian bystanders
Considering Russia is deliberately bombing hospitals in Syria [nytimes.com], and has been doing so since the day it started helping the Syrian regime, and the world doesn't even raise an eyebrow, I don't think a drone accidentally hitting some civilians will be that big a deal.
America has been accidentaly been hitting civilians with drones for years and no one gives a shit.
Re: (Score:3)
America has been accidentaly been hitting civilians with drones for years and no one gives a shit.
The difference is it's an accident when the U.S. does it and it is investigated whereas when Russia deliberately bombs hospitals it either claims no knowledge of the event, despite there being documentation showing the pilots know they are bombing hospitals [nytimes.com], or claim they are going after "terrorists". And by terrorists they mean anyone who disagrees with the Syrian dictator.
Further, when civilians are hit, the U.S. owns up to it and at least makes a good faith effort not to deliberately targeted civilians.
Re: (Score:3)
Syria is playing the victim here. They put missiles and weapons caches in hospitals and then cry foul when its bombed. Neither side is that stupid.
The misinformation is strong in this one. These are hospitals which are fully marked and have patients in them. In fact, the UN gave Russia and Syria a list of hospitals, and their locations, and immediately afterwards both Russia and Syria started bombing those exact locations.
No, there are no weapons caches in these places. Only civilians who have been targeted by Russia and Syria.
Re: (Score:2)
Elon's speech seemed to say that he think that the F-35's replacement will be a remotely controlled drone with some autonomous assistance.
I think that we're pretty far away from having a fully auto piloted drone, though. If Elon thinks that the PR blowback of a Tesla on autopilot crashing into someone is big, just wait to see what happens when an armed drone accidentally blows up some civilian bystanders because it's sensors got confused during a dogfight.
He can barely get a car to drive itself properly yet, maybe master that before we try getting fighter jets to do it.
Re: (Score:2)
There have been fully autonomous drones for quite some time. Technically "drone" used to mean an autonomous aircraft. And you don't have civilian bystanders to dogfights.
After all, what are missiles but autonomous single-use drones?
Is where it's at. .. (Score:3)
Re: (Score:2)
Hypersonic drones will probably be the next big thing. Very difficult to intercept, the next big arms race.
I don't know why this is news though. Musk says a lot of stuff, most of it bollocks.
Not so fast (Score:3)
Musk apparently doesn't understand the physics of flying or how this dog fight thing happens.
Where a drone would be a difficult opponent in some situations, IF the F-35 could be seen from it's pilot's vantage point on the ground, the ONE thing the F-35 brings to the party is stealth. It may not be as stealthy as some fighters, but it's going to be pretty hard for a drone to follow using sensors like radar. Visual following using, say a contrast tracker, is a hit or miss proposition and forget IR tracking, that's part of the stealth thing these days.
That's not to say drones are not a threat to the F-35, they are. It's just that in this case, the conventional ground to air missiles are going to be most effective and if you think about it, really are just specialized drones already. We have developed tactics and countermeasures for missiles which are pretty effective - including stealth.
Flying a drone remotely quickly becomes a difficult task when you are maneuvering. Sure, flying straight and level for hundreds of miles or circling some point on the globe is easy to automate and not too demanding on the data links in terms of data rates and latency, but when you need to be precisely controlling an aircraft in an unpredictable flying environment, the amount of data you need and latency start to become a significant problem. Dogfighting requires very quick reaction times, the quicker the better, so pilots train to see aspect changes (the observed aircraft is changing it's orientation or even better the deflection of the control surfaces that cause tha aspect change, and learn to react... "He's rolling to the right, means he's going to be making a right turn." even before they move to the right. This will be VERY difficult to do with a drone over a video link with the added latency it will require. The human piloted aircraft will have a distinct advantage in close quarters when the energies are low and the winner of the fight is about how fast can you turn in the right direction where a drone will have the disadvantage of communications delays.
So, Musk is a bit crazy here, and is likely boasting about things he really doesn't fully understand.
That's not to say drones don't have a place in an air to air conflict, just not as individual combatants. Where a drone will be *very* useful and dangerous is in the possible uses as remote sensing platforms and communications link providers. Drones can work in concert with manned aircraft to launch weapons, provide remote and cooperative sensing making the manned aircraft more effective or able to stay out of harm's way by staying out of weapons range, but launching missiles from a drone instead. But in a dogfight, those drones will generally lose...
Re: (Score:2)
A combination of remote control plus local AI on the drone could be very interesting. Without a human onboard, drones can pull far higher acceleration, and the AI would only need to handle decisions on the timeframe of the latency of the control link, not anything long-term. It helps that AI can have reaction times far faster than a human, and far faster than any plane or drone can manoeuvre.
You'd have to actually program that AI, and find a way to fuse the AI and human control in a way that handles natural
Re: (Score:2)
So, Musk is a bit crazy here, and is likely boasting about things he really doesn't fully understand.
Musk talking about shit he don't understand? Nope, don't believe it for a second. XD
Re: (Score:2)
Musk apparently doesn't understand the physics of flying or how this dog fight thing happens.
Doesn't understand the physics of flying? Better not let him near a rocket then.
Not controversial claims, consistently, Elon makes (Score:4, Insightful)
But PayPal was inevitable. Transitioning away from fossil fuels. Privatizing launch and orbiting vehicles...the man has always been in a position of enormous privilege and sufficiently modest to study his interests and hire teams he can discern know more than himself.
But on this topic, not a lot posters are considerately framing "what's in front of you". In 2005, I suspected what ratio drones versus piloted craft comprised US defenses/offenses would be classified and I'm no military hobbyist. Drones aren't new. And Musk's "opinion" is not radical in any way.
Too many posts are assuming the simplest of navigational controls (read vulnerable and without supplement or contingency) to sic a flock of drones on a squadron of piloted aircraft or emerging [collective noun for drones{bee related suggested}] of drones. Cooperative bot games are being publically presented, so what's being developed with classified techniques and methods (and their combination) is far forward of it. Most, if not all, drones will be piloted, but the assisting technology will undoubtedly approach autonomous function and task completion if that connection is compromised.
Musk's obsession with technology to have his cars self-drive is ONE of his greatest blunders. Wouldn't an electric car that is "faster than falling" be just fine without having it assist? But NOoooooo. Can't be as inexpensive as it might have been...must add on all that jive that is many years out from reliable and safe. NO system of guidance is demonstrated as it will have to negotiate being surrounded by other systems, them all signal bouncing for orientation.
It's ALWAYS: Look how this lone vehicle negotated not striking a pedestrian or building or confusing a tree with a fork in the road. Unless you're counting carriers in series and their assisted maintenance of proximity, which is, ya know, A TRAIN. Independently self-guiding drone swarms and traffic systems are parallel problems and, imo, the task will be achieved in military applications in the sky with plenty of room to manuever before reliable systems prioritizing non-collision will reliably work on the ground.
Well, what do you think those "Navy UFOs" are? (Score:2)
If not autonomous drones the size of fighter jets. Every nation state in the world with a big enough budget surely has been working on just this concept for at least the last 10 years.
No chance at all (Score:2)
Drones... (Score:2)
One of the biggest limitations in fighter design is the ability of the pilot to handle the forces associated with high speed manoeuvring, a drone would not have that limitation and also wouldn't require all the equipment required to support a pilot so could also be lighter, assuming the same level of technology to produce the rest of the airframe and the engines etc a drone should be able to outperform a manned aircraft.
However there are potential problems to consider...
Latency of the control connection - i
Fifty years (Score:3)
(Or maybe that is how long it will take for customers to complete the repayments?)
So it is a reasonable bet that within the next 50 years there will be sufficient development of UAVs and target detection to utterly and completely kick the F-35's arse.Probably well within 20 years - maybe even less than 10.
It is also more than reasonable to assume that those drones will be cheap to the point of expendable and that their numbers will mean that even a 50% attrition rate against an F-35 would be acceptable losses.
He must be right (Score:2)
There never was a fighter jet era (Score:2)
Obvious (Score:2)
Global Air Forces Already Heading that Way (Score:3)
Different Missions (Score:2)
Moneyball (Score:2)
F-35's cost six to seven times what a Reaper does, and their cost per flight-hour is TEN TIMES that of a Reaper.
I'm not aware of a US drone outfitted with interceptor role missiles like Sidewinders or AMRAAM. Such a thing could be done though.
Could a hotshot pilot in a latest-generation superiority fighter defeat even half his plane's cost in dedicated interceptor drones?
How many replacements for the dead pilot are available?
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
What on earth is a car manufacturer doing on a Warfare Symposium?
Space-X clearly wants in on that sweet military industrial complex money.
Re: (Score:2)
Musk said autonomous drone warfare "is where it's at"
Emphasis mine.
Re: (Score:2)
Musk said autonomous drone warfare "is where it's at"
Emphasis mine.
Musk is crazy. Carrying a sensor package that could collect enough data to dogfight would be a huge (literally) deal and then to toss in enough processing power to deal with that amount of data would add significant weight and power consumption to his drone, enough to outweigh any pilot. Now you want to get in a dogfight with this? I don't think so.
Drones have a place in air combat, just not in a dogfight.
As a sensor or communications platform that has basic defensive capability? Sure, excellent choi
Re: (Score:2)
F-35 does not need any sensor package or processing power?
Re:Depends (Score:5, Interesting)
G forces only matter for the first few moments of a dogfight.
Once everybody hits their maximum sustained turn rate, which is a function of thrust, drag, weight and lift, the G-Forces are actually pretty low, and usually not a serious factor. A turning fight is full throttle, pulling to your best rate AOA and then figuring out what your opponent is going to do and reacting quickly. Turning fights happen well under max-G loads because even in modern fighters there isn't enough thrust to keep the energy up when at the high AOA's. HIgh G loads produce MASSIVE drag and require huge amounts of thrust to sustain, so what usually happens is everybody makes their first turn and by the time that's done, everybody is at full throttle, rolling to keep your opponent on a line that you can pull them through your nose (or keep the other guy from doing it to you) while keeping the AOA at the ideal point to maximize your turn rate.
Fighter pilots have a saying "Rate kills", the guy with the best turn rate and quickest reaction time wins (assuming everything else is the same) Drones won't have a huge advantage on rate and will have a significant disadvantage on reaction time. Who's going to win? Humans... Humans learn, adapt and can employ tactics that are advantageous to them, drones cannot adapt as quickly.
So, unless the drone has a higher thrust to weight ratio (which if it's carrying enough sensors to replace a pilot and the processing power to sift through the data it collects to make decisions on is highly unlikely) it's not going to be at much of an advantage because a pilot can only handle 6-9 G's for short periods.
Re: (Score:3)
Sensors are fairly light weight compared to what's required to keep a human being alive. Not to mention that our bodies take up an awful lot of space that you can't sensibly distribute across the body of a plane, so that's not exactly a huge disadvantage. What's left is reaction speed, and to be blunt, no human being can hold a candle to the reaction speed of a computer.
Re: (Score:2)
One on one in dogfights, yes. However, if drones are much cheaper to build, then quantity, to reuse Stalin's phrase, has a quality all its own. If my memory is correct, the Air Force has done tests with the quantity of cheaper planes greater than the quantity of superior planes. I don't know what the actual ratio turned out to be quantity won.
Re: (Score:2)
The F-35 already has lots of cameras. I don't think it would take a huge amount of processing power to make a competent dog fighter. You wouldn't have the creativity of a human pilot (yet), but standard air combat is fairly algorithmic. Most importantly, computers are way cheaper than trained human fighter pilots.
Re: (Score:2)
I presume what you're trying to say is that game AI has to be handicapped to not walk all over humans in flight sims already, and that's been true for ages. Further, that's assuming that the AI pilots are limited to human levels of G forces, where the only limit the drone AI has to follow is "don't rip the wings off". Machines already do much of the work for the human pilot, not just the obvious stuff like flying but also target aquisition and identification, so there's no reason why autonomous drones can't
Re: (Score:2)