I reality it would probably be more like how the software for voting machines is developed. In secret by large companies who's only specialty is doing anything they can get the government to cough up money for.
There are actually already some UAV AI competitions that accept student teams, though they're typically not competing in a dogfighting sense, more in a "follow a path, search for an object like x and take a picture of it", etc.
For those in the aerospace industry, are we seeing any movement towards unmanned commercial airplanes? Are the impediments mostly technical or social? Any rough timelines?
Indeed. As shown by the recent distracted-pilots-flew-past-Minneapolis fiasco, modern planes have become so automated that one of the biggest problems for pilots is fighting boredom:
There's probably too little freight to justify development of an unmanned commercial plane. I think the trivial advantages for passenger planes would be outweighed by the misdirected incentives. :)
There are several problems for an unmanned commercial airplane. None of them are insurmountable, but they add up to making it easy to see how this will probably be the last sector that gets automated.
In a combat aircraft the advantages for getting rid of the pilot are numerous: less weight & volume for the pilot and their associated support kit, eliminating the pilot allows for plane geometries that are just not possible with a human onboard, eliminating the pilot enables maneuvers that would turn a pilot to jelly, etc. The pilot is probably the most expensive component of a combat aircraft. Eliminating the pilot opens up new mission profiles (e.g. you don't have to worry too much about the aircraft getting back to base.)
For a commercial aircraft the benefits of going unmanned are not as pronounced and in some cases are non-existent. In addition to the lack of benefits, you introduce additional costs related to liability (you can always blame pilot error for a problem, but if there is no pilot then it is the airlines fault for either not maintaining the system properly or for selecting the system in the first place) and the fact that passengers are not going to trust this system anytime soon.
The impediments are almost completely social, but the benefits are so slight that it does not make much sense to bother aiming for completely unmanned. What you are going to see (and in fact what has been happening for years now) is the increasing automation of cockpit tasks. There will always be a human being "at the controls" for the forseeable future, but what that actually means and how much active control they are exercising is going to be slowly changing.
You can always blame pilot error for a problem, but if there is no pilot then it is the airlines fault for either not maintaining the system properly or for selecting the system in the first place
Great observation. The primary advantage I saw (reduced pilot errors, responsible now for ~3/4 of all aviation accidents) is a liability for the airline and the manufacturer.
> eliminating the pilot enables maneuvers that would turn a pilot to jelly
I think this is a common misperception. I believe the operational risk is that manuevers above 6g (varying on pilot and G-suit) reduces blood flow to the brain (by forcing it into the legs) and cause a blackout. See http://en.wikipedia.org/wiki/G-LOC Blackouts, even for a few seconds, can be devastating in the middle of a dogfight. I believe there is software in modern fighter jets for just when pilots blackout but I can't find it on google right now.
I didn't interpret evgen's comment as literally turning a pilot to jelly. (Hrmm, there's an interesting visual.) For practical purposes, a pilot that's blacked out = jelly.
Oh, I was responding in general. Most articles, the one (linked included) talk about physical limitations about human pilots. I always thought it was something more gory until I found out about blackouts a few years ago.
The big difference is that redout happens at substantially lower G force, partially because there is no way to use skeletal muscles to fight it. This is partially why both the Immelman and the Split S use a "climb" to effect their vertical movement.
I think what kwantam meant by lower G-forces is that an experienced pilot can use abdominal/extremity muscles and a fancy flight suit to take around 9G before losing consciousness, but redout will hit at around -3G and there is not much you can do about it. A maneuver that you could easily do in normal flight becomes impossible to sustain if you are inverted, for example.
It's probably 20 years old by now and I think I taped it back then (with very poor reception!), but Nova did a great episode called "Top Gun and beyond" about the limitations of human pilots in fighter aircraft.
I get the feeling with people are unwilling to have unmanned trains, then unmanned planes are a no go (except by the military who can make a logical cost-benifit-risk decision in this case).
"Learn how hackers can bring down planes into buildings at 5"
Considering that military planes are used with the intent to kill, it's not as much of an issue as when passengers are aboard. Plus, most of these fighter planes are going to be flying over war zones where lots of people die all the time.
Actually, a lot of the unmanned military planes are used in the US (some are stationed down the highway from where I work). I just think society isn't ready for unmanned mass transport, but the military has some systemic advantages in being allowed to go that way.
When I moved from Vancouver to San Francisco I found it completely bizarre that they required Drivers for BART. It took me a couple years to get over that.
The DLR in London is unmanned too, that ferries commuters into London's second financial district. If that's not a terrorist target I don't know what is, so steps must have been taken to secure it.
I was going to mention the DLR. Although there is no driver (a very good thing IMHO), there is always a guard who closes the doors and instructs the train to move to the next station. The guard is currently vital because you don't want a totally automated train to move off if someone is (eg) trapped in the doors.
The metro in Copenhagen is unmanned and fully automated. Doors open and close by software. I'm sure there must be other places that run with a similar system?
Skytrain, in Vancouver, has sensors to determine if someone is stuck in between doors, or, alternatively, has hopped onto the tracks. The trains at Peak time are leaving every 45 seconds, so the system has been fairly well tested over the last 23 years.
Radar guided missiles are easily defeated with modern ECM. Despite what the salesmen will tell you, in a real fight you have to get in and knife the f*&kers.
The same happens with remotely piloted vehicles. If you ever go up against even a modestly sophisticated adversary they will show phenomenal pluck and ingenuity. See serbian downing of an F-117 for an example.
Human pilots will be needed for some time to avoid having your toys fall out of the sky or stolen and used against you.
I agree, the article talks about the competition investing in UAV's but it fails to mention the average modern military's ability to defend themselves against UAV's.
Didn't they say the same thing about swords when guns were invented, about infantry when artillery was invented, and about battleships when aircraft carriers were invented?
It was true back then, too - it was a good 25 years between when aircraft carriers were invented in WW1 and when they sank their first vessel at Coral Sea. But new technology opens up new tactics and new economies of scale that can fundamentally change the rules of battle.
What if you didn't care about having your toys fall out of the sky, for example? Make them so cheap that you can literally blanket the sky with them. Send up thousands of computer-piloted missile-aircraft. If they get shot down, who cares? If they shoot each other down, who cares? If they get stolen by the enemy and used to shoot down your own missile-craft...well, that's what was happening before anyway.
But you could throw up a curtain so dense and so shifting that it's impossible for anything to fly through them. And that's exactly what a potential adversary of the U.S. wants: nullify American air power so the f#&kers on the ground can knife our guys. It doesn't matter if they shoot down 10 of their own missile-craft for every one of ours. If they can put up 100x as many, that still equals no aircraft in the sky for Uncle Sam.
There was a recent wargame that followed a strategy much like this, except that instead of computer-controlled aircraft it used human-controlled small boats launching cruise missiles (that other automated-pilot air weapon):
Nearly pointless correction: when they sank their first vessel at Coral Sea – that was the first direct battle between carriers, not the first sinking of an opposing vessel. They did a bang up job at Pearl Harbor if nothing else.
Already happening. Radar/IR vs. ECM vs. ECCM is an ongoing 'cobra vs. mongoose' co-evolution that won't settle in favor of either offense or defense.
US air superiority is a system that includes AWACS (enabling better situational awareness), GPS (enabling JDAM/SDB for example), tankers, and a constellation of factors other than "mano a mano" with bullets. UAVs are already an integral part of the mix.
The Serbian downing of the F-117 almost certainly had more to do with a spy in NATO relaying flight path info on a moonlit night, making visual contact possible. But as is the case of all black bag operations, the real story will likely stay hidden.
Or, somewhat more boringly, the spies were posted outside the airbases in Italy with a mobile phone and instructions to call when the F-117's took off.
The first is a forum for junior military officers and professional analysts for discussing tactics and strategy. The articles all have by-lines, mostly by Captains and Majors (i.e. recently promoted Captains). The editorial board is made up of professionals. You will never read any classified information here. You may read some good war stories, but mostly you will read professional treatments of modern tactics by professionals for professionals.
On the other extreme many consider debka a conduit for the Mossad. Good luck separating the information from the disinformation on this site, but the articles can be a good read, if you can deal with the wilderness of mirrors.
In between these extremes keep in mind that intelligence analysis is about connecting the dots. Even the professionals seldom know anything for sure, and sources and means of gathering information are usually more important than any individual datum. If it makes it to print (or html), there’s probably a reason.
So where does this leave http://www.strategypage.com ? Let’s look at the staff bios on the about us page. Of the four "staff" exactly one has any military/intelligence background at all. "...author (over 20 books), wargame designer (over 100 designed and publisher of over 500), defense advisor (since the 1970s), pundit (since the 1970s) and general troublemaker. Served in the US Army artillery (1961-64) as a Spec/4." His website, http://www.jimdunnigan.com , lists some of his books, and more bio info including "Various Department of Defense and State Department agencies called on me to give lectures or just have a chat." I guess that was his "defense advisor" gig. He’s certainly a dedicated amateur. If luck had broken his way a little better, perhaps he could have been Tom Clancy. There’s also a handy "write for us" page. If you write well enough and don’t make any flaming technical errors, you can make some coin.
In short, I don’t think this site would even attract placing any real disinformation. So how about the article itself?
First, no by-line.
Secondly the protagonist’s name, "Zoltan". Sure sounds made up to me, but that’s just my opinion.
The way the piece is put together has the feel of a composite, somebody who did some research and threw this story together. If a real reporter got the real story from a real Colonel (and covered up his identity with a pseudonym) he could have written for a more prestigious publication or wire service and made better money. Still...not entirely implausible.
For now I’m sticking to my theory of the shoot-down only because accounts I read right after the incident described the wreckage of the aircraft landing largely intact (a little hard to believe since the fuselage is composite) with bullet holes and, if I recall correctly, the pilot ejected and survived (a little less likely after a missile strike). Hence my belief in heavy machine gun fire from a pursuit aircraft. Also the Russians really wanted to get their hands on one of these babies, even a crashed specimen. So it’s easy to imagine their risking an agent inside NATO to pull this off.
Reading this stuff on the web is not like reading hacking tips. The background is more important than the foreground, and you can never be sure.
I understand your point about the background of the website. However, in this case I thought it was a fairly well known story, and there are quite a number of other accounts.
eg: ""At times, they acted like amateurs," Dani said, listing some ways the Serbs managed to breach NATO communications security, including eavesdropping on pilots' conversations with AWACS surveillance planes.
"I personally listened to their pilots' conversations, learning about their routes and bombing plans," Dani said."
"A network of spies and observers would report F-117A takeoffs from Italy’s Aviano Air Base. Coupling this intelligence with tweaked radars, the Serbians fired several surface-to-air missiles almost blindly, hoping for a hit."
I thought that it was shot down by SAM missiles was generally accepted. The NATO spy part was more interesting, because there was reasonable support for the Serbs within some parts of NATO, so that isn't out of the question.
The "people listening for aircraft" theory makes a lot of sense, too, though - it's easy, and it's been proven to work before.
Thanks for the reply. The airforcetimes article is really very good. I wish you had originally posted it instead of the strategy page article. The usatoday link seems to be MIA.
Today, no passenger wants the computer to control the plane. They are right.
Today planes can take off and land using only software and radio signals, even when the pilot can't, like no visibility when cloudy.
The problem is that computer understanding of the surroundings is NULL. Today there is no software that could understand(well) that there are birds(or small planes) on the way that are going to be eaten by your engines, and damage it.
The only technologies that work are active, like LIDAR, this is going to change, but first on vehicles with less degrees of freedom. Trains(1FD) before ships(2FD) before cars before planes.
Today there is no software that could understand(well) that there are birds(or small planes) on the way that are going to be eaten by your engines, and damage it.
This is just fearmongering. Birds, sure, but airplane engines are designed to ingest small birds and for the most part can handle it reasonably well. Small planes still must have a transponder, and TCAS systems will identify and warn of ADS-B signals ("squawk"), even if the smaller plane has not been equipped with a TCAS transceiver.
I'm talking about groups of birds here, they are very real. E.g, the pilot of a big plane saved the lives of the passengers when a big group(thousands) of flamingos take off from Doñana national park in Spain, he did what the book says not to do, put the plane in full-power and fly over them.
There is a rule in engineering, never underestimate a 900km/hr projectile. And I have seen myself engines birds test that are mandatory on every turbine.
Entirely social. Commercial jets are already equipped with software that can take off, navigate, fly, and land the jet. Generally more efficiently than a pilot can. The job really is an easy one for software.
Unless something like this happens: http://en.wikipedia.org/wiki/United_Airlines_Flight_232 . Then there's absolutely no replacement for a human being or three in the cockpit, and the public and the FAA know it. A pilot's salary or two is not too much to pay to protect the lives of several hundred people once in a while, even if it's otherwise unnecessary.
The military believes it, too, and the jets they fly have a higher likelihood of encountering life or death situations such as having pieces shot off and needing to get home. While there may not be pilots in the jets in the next generation, they'll be flying them for a while yet, and commanding them for a long time. The battlefield is too difficult, complex, and important for anything else. Once someone gives the order to drop a bomb, they'll trust software to execute the order efficiently and correctly. But it's going to be a human being giving the order for a long time. Possibly forever.
There are some jobs that won't be done by software until it can pass a Turing test. Software may be able to do the job in the vast majority of cases, and in some societies that may be good enough. Socially speaking, in this society, it isn't.
How do you figure that? The Civil War demonstrated nothing of the kind. It only showed that if you split both the populace and the military establishment in two, the side that has more railroads, factories, and soldiers is going to win.
I think the fact that the National Guard consists of nothing but fewer, lesser-trained soldiers with aging, outdated weapons/technology is a pretty good indicator that the population would never be able to take on the full might of the federal military branches.
If you put all the power the state can muster on one pile and that of the population on another, resistance against the state has been impossible – I would guess – for a few centuries (or more). But that is – again – not how it works. There are other players involved. Parts of the state might not be willing to use all that power. Other states (sometimes with vastly more power) will constrain the ability of the state to use all its power. And so on.
The consequence of all this is that revolutions against tyrannies are possible without firing a single shot or using any violence at all. Even though, just looking at the raw power of the two sides, the state could easily crush the population. Just think of the fall of the Berlin Wall.
"...tell me, how good does a combat robot have to be to kick in your door, spray the house with mace, and hit anything that coughs with a taser until it stops moving and can be encased in spray webbing for later pickup and processing? How goes does a robot have to be to detain anybody who smells like pot? How good does a robot have to be to break up a political demonstration? Not that good. If a robot is good enough to fight wars, its good enough to do political policing. It might not solve crimes, but it can bust heads that protest and capture people who’s position is known and who have been identified as enemies of the state."
I feel like my eventual children, or maybe my children's children, will see a world where Americans don't go to was physically (or at least in large numbers). This whole war thing is going to be really weighted against the poorer countries who have to leave real people to die.
I see terrorist tactics (blowing up buildings, schools, etc.) as dominating their efforts.
That said, I think that if I were to do a defense startup now, it'd be towards detecting explosives, guns, etc. in large groups of people. If that technology falls behind, war's about to get a whole lot uglier.
"I see my children's children living in a world without war."
Do you honestly believe that? I don't see how it's even possible unless we sedate everyone in some kind of matrix like false utopian state. It's good to dream though.
True, but the environment in Western Europe is very different from that of the rest of the world, and it's not clear that we can scale up the conditions of Western Europe to 6 billion people, at least with present tech, and especially if we're nearing peak oil.
If peak oil is true, I would predict quite the opposite.
Western Europe is a pretty small slice of the world, and they've had peace for half a century, which on the scale of multiple centuries, is hardly cause to celebrate the dawning of the age of aquarius. I'm sure this isn't the first 50-year period without war in Western Europe.
Alliances can easily decrease the risk of small scale war, but this is at the cost of systematic risk (alliance vs alliance war or world wars). Perhaps we've only managed to decrease the risk of medium-sized wars at the cost of increased risk for that very rare, very big one. It wouldn't be too different from the recent financial crisis or power distribution networks in the US.
Western Europe has lived for decades under the U.S. Strategic Umbrella.
Once (if) that goes away, Western Europe will either have to pony up for it's own defense, or be held hostage to outside powers with their own agendas (like Russia blackmailing Europe with gas supplies)
I don't see the struggle between national interests going anyway anytime unless and until the nation-state itself is rendered obsolete. People need stuff, and that creates conflict. Different groups of people will need different stuff. And let's not forget ideological problems -- Western Europe isn't exactly a model of social integration.
It only takes one bad actor to make all participants gear up for the potential to use force, and we've found that once geared up, the actual use of force is not far behind.
It's a prisoner's dilemma played out on a global stage. But you'd find the same situation with three thousand guys living in caves somewhere. War is a natural consequence of groups of unique and intelligent humans.
The struggle between national interests does not require war. War is counter-productive to national interests. A great conflict is currently playing out between the USA and China in Africa but not a shot has been fired and there is no reason for a shot to be fired.
Well it's worse than that -- all armed conflict requires is a group of armed people determined to use violence. What terrorism has shown us is that it doesn't even require nation-states.
So your nation can be so inept as to not be able to control things inside it's borders (many nations fit this bill). If armed camps build up inside them and attack other countries, guess what? You're at war -- or something so close to war as to be indistinguishable from it.
If you'd like the world to work like some super-big version of your local country, where most conflicts are solved by some civil system, then you're going to have to have popular choice of representatives, mixed local, national, and world control, an independent court system built on something besides treaties, etc. The UN is nowhere near any of that.
There is the theory that democracies only compete via commerce. If you buy into that, then perhaps a world full of democracies would be the most peaceful. But I don't see that happening either any time soon.
It's just a long-shot. A real long-shot. I'd argue that in order to do away with war you would have to do away with 99.999% of the impetus for using violence under any circumstances. It'd be a world of extreme pacifists. Would that be such a great thing? Doesn't chaos -- whether war, sickness, disasters, etc -- contribute to evolution and growth? Do we really want to live in stasis? Or -- to put it in religious terms -- is heaven really such a great thing?
Not if you count the London bombings of 2007 or Madrid in 2004. The perpetrators were at least in part European citizens.
In any case, it is one thing to have stability in an affluent, post-religion area like Western Europe, quite another to have it worldwide. Many nations in the Mid East, for example, are comparable to medieval Europe - except that if the current regimes fall (as they might well, if oil loses its value) their subjects are more likely to create a Muslim Brotherhood-type regime than a liberal republic.
It worked for western europe: Great Britain and Portugal claiming rights to trade with Spanish colonies in America motivated a lot of wars; the USA independence war from GB was motivated by economic reasons, etc. But it is not enough.
I would also add the end of nationalism and colonialism to the requirements for world peace. Although you could say that without the former there isn't the latter.
How do you protect your ideology of freedom and [pseudo?] democracy then from those bent on violent oppression. From what I can see a nation of absolute pacifists can be subdued by a single violent person - there has to be some point at which you'd make a physical stand?
I can see the courageous absolutism of Gandhi working well on a small scale inspiring others to rebel against the status quo - but against a nation bent on destruction of your own people, say, what then.
What I find intriguing is that in a situation of combat between computer-piloted UAV it is purely technological advance that gives the advantage. That then is dissociating war from direct violence and making technology the battleground rather than military aggression and effective killing.
Sadly I don't believe, for now, that we can end human warring in favour of "robot wars".
I can see the courageous absolutism of Gandhi working well on a small scale
You can't extrapolate from Gandhi to a generalized theory of resistance. His strategy worked well against the British Empire at a time that it had already decided that empires were more trouble than they're worth, and besides, we only came here to play cricket anyway. Any other imperial power would simply have shot him.
Yeah, I don't think Ghandi's strategy would have worked against Stalin. In the purges he had people murdered in broad daylight pretty regularly and it didn't even ruffle his mustache.
> From what I can see a nation of absolute pacifists can be subdued by a single violent person - there has to be some point at which you'd make a physical stand?
Policing will probably always be needed, but that is different. The police are not at war with criminals.
The reason wars between countries are so rare is that the U.S. has military bases in 132 countries. Countries with U.S. military bases do not fight each other. If you dream of world peace, you should be supporting the American Defense Department and Pax Americana.
While I agree with the fact that there are many US military bases around the world and they contribute to the equilibrium, your statement sounds a bit like slaves won't fight between them because they serve only one master.
Cecil Rhodes (http://en.wikipedia.org/wiki/Cecil_Rhodes) had the same idea about Pax Britannica. The problem is that a lot of countries don't like having foreign invaders based on their soil.
1) Falkland's war between Argentina and Great Britain in 1982.
2) Greco-Turkish war in Cypress.
3) Indian-Pakistani wars (although this example is stretching it, as US influence in India is quite low)
4) The First and Second Cod War
5) The Turbot War
There will be war for as long as there are debates on the internet: until everyone agrees about everything important enough to fight over.
World peace could arise as the result of some extremely democratic/libertarian/scientific worldwide monoculture that agreed on everything but what they were willing to settle with words. I think that an unlikely end to the story, though. The easiest road to world peace is for someone to win. And that's not a happy ending in my book. I'll take war over that. It's nasty, but nowhere near as bad as tyranny.
Put another way, there will be war as long as there is crime. I hope that's forever. Living in a society that had actually done what it takes to eliminate crime completely would not be worth it to me. I feel the same way about a world without war.
What's not entirely clear is which countries will be poor and which ones will be rich two generations down the road. What if it's your children's children who are left to die?
It think we do have a chance to end wars once and for all as economies get more interconnected. Traditional geostrategic thinking is obsolete.
Not really. 200 software developers can write the code for every piece of flying drone that the U.S. has. While pilots where employed almost 1:1 for each piece of aircraft. Not to mention all the crew, flight controllers and maintenance people; I don't think a drone undergoes the same rigorous safety checks and standards as a manned plane. And as the military depends on more drone, the cheaper they will become.
One interesting development is that, with the growing number of UAVs, there aren't enough pilots to fly them all, so new operators are no longer required to have flying experience.
I heard an interview with "Wired for War"'s author who claimed that controls for some of these systems are modeled after X-box/PlayStation controls, both to capitalize on the ergonomics research done by console makers and to make it easier for young recruits who are already used to operating them.
"The wars of the future will not be fought on the battlefield or at sea. They will be fought in space, or possibly on top of a very tall mountain. In either case, most of the actual fighting will be done by small robots. And as you go forth today remember always your duty is clear: To build and maintain those robots." (http://en.wikipedia.org/wiki/The_Secret_War_of_Lisa_Simpson)
It's fun to imagine that somewhere sometime a scrawny geek told a chiseled-jaw pilot "Go away or I will replace you with a very small shell script"
That may have been my late father's fantasy. He was the lead engineer at Boeing Military for much of the B-52's avionics. He was not a man much given to resentments and anger, but he strongly disliked USAF pilots, and it was clear the other engineers did as well. They apparently are quite the prima donnas, and arrogant in the extreme.
This engineers vs. pilots trope goes back to the days of the space program. Tom Wolfe talked about it in _The Right Stuff_.
I am sure at the end of the 19th century horse drawn carriage makers faced a dismal future as well. Some professions simply die down and become irrelevant.
Interestingly enough this already happened to the Airforce in the late 60s. Airforce was running manned spy satellite missions, but because of the cost and the danger involved, those missions were being canceled in favor of unmanned satellite missions. So this is basically a repeat for Airforce.
Maybe we can spend the trillions from fighter jets and pilot training on health care?
The war machine expenditure in the USA is more than EVERY OTHER country on earth COMBINED (and it's deficit spending).
The problem is the war machine is always chomping at the bit to get into a war so they can use their toys. Did you know Japan has a military? And they are protesting against their constitution which prevents them from getting involved in other wars. They WANT to go to war, somewhere, anywhere. It's crazy.
Nope. Fighter planes are a special case because they're both significantly cheaper and with better performance without a pilot on board. I don't see infantry replaced any time soon.
Writer Lewis Page points out that the Infantry are the only branch of the military who can do something other than blow stuff up from a distance. Demand for that type of operation is only going to increase.
Meanwhile the UK govt is cutting Infantry battalions and investing 10s of billions of GBP in Eurofighter and Nimrod. Because the Taliban's airforce and submarine fleet are obviously the biggest threats we face...
Note also that it's only the USAF who require trained pilots to operate UAVs... Everyone else uses automated landing systems (etc) and regular troops fly them. It's one of the featured jobs in the British Army's recruiting campaign, there's a guy with a handset similar to an Xbox controller scouting ahead of his unit as they patrol.
Yes, and after that, we'll have to start teaching ourselves multiplication again, so we can develop a manned plane that can break the stalemate that software will get into:)
They would never use Linux for Avionics, because it would be far too expensive to get such a large and complex system certified for that kind of work.
Commercial planes use linux for non-critical functions like the in-flight entertainment. I can't think of anything that would be on a fighter plane that would be considered "non-critical" enough to not require a high-integrity system though.
Vxworks is typical, but Linux does happen sometimes. Most often for ground stations or auxiliary tools, sometimes for continuity between the development and deployed environment, sometimes because it's just a research project and certification isn't an issue.
I wouldn't claim it's common or that you'd be likely to see it in a deployed environment, but it's not totally unheard of. Throw Linux & UAV in Google - you'll see what I mean.
"t's not just that most of the those American air force generals began their careers as fighter pilots. No, the reason is more practical. American air superiority has largely been the result of superior pilots. The U.S. didn't always have the best aircraft, but they always had the most talented and resourceful pilots. And that's what gave the U.S. its edge. Will that translate to software piloted fighters? Research to date seems to indicate it will."
I really need some real non-american references to believe in such a thing.
In WW2 German fighters were technologically superior. In Korea Russian MiGs were superior (the Sidewinder missile was developed specifically to combat MiGs that could cruise at an altitude American fighters couldn't reach, swoop down to attack, then return to safe altitude). Even in the Cold War, the MiG-25 could out-run any American fighter.
Whether the quality of American pilots was decisive in any of those conflicts is debatable, but the first part is certainly true. Many observers believe the Eurofighter Typhoon to be the most advanced fighter in the world today.
"Talented and resourceful"isn't quite right. What the US has generally had, though, was the best trained pilots and feedback. When the kill ratio starts favoring the enemy, the US has typically analyzed the problem and improved training.
I referenced Top Gun and Beyond in an earlier post: if you can find the video, it goes into this.
Pretty amazing stuff. Most missions in Iraq and Afghanistan are flown via satellite from the US, truly remote controlled war. (Submitted it to HN at the time, but it only got 2 points ;))
Sooner or later someone will make one cheap enough for police forces to buy. We ought to put some thought into whether that is going to be a good thing in the long run.
On the plus side:
- faster response to 911 calls, since UAVs can avoid traffic, and if cheap enough,
they could be salted about the place in a powered off state.
- less need for police to risk their lives, and consequently
better policing of gang-ridden areas.
On the minus side:
- police become (even more) disconnected from the general
public, potentially developing a culture of impunity
- spread of surveillance
As someone who's worked with these systems, we faced quite a bit of this pilot-as-luddite resistance at first. Unfortunately this article doesn't really capture that many pilots are thankful that some flying computer will get blown away by a popup SA-6, rather than them. Nor does it convey that, while batches of UAVs can perform complicated tasks autonomously, they have not yet achieved air dominance.
When I worked in the City we came across a lot of trader-as-luddite resistance to new software too. Many feared being automated out of jobs. Those who got with the programme tho' found they were able to single-handedly managed more complex positions. I think we'll see the same sort of thing here, perhaps a single pilot with the right technology (controlling multiple UAVs?) will be able to carry out a mission that would previously have taken an entire squadron. As the article says, the Pentagon can buy UAVs faster than it can train fully-qualified fighter pilots.
The article states that countries that want to attack the US are excited about this. The article forgets to mention that the US is doing most of the research into this field.
While you are correct that we don't have any actual information to show that in this thread, it would seem plausible (if you compare the amount of money that the U.S. spends on research and the air force vs. e.g. Iran) that America's doing most of the heavy lifting here.
Remote-controlled war planes will probably return to being thought of as a niche/specialist weapon after the first time they are launched against an enemy with some actual engineering know-how. (Consider battery/solar-powered Tesla coils lofted on weather balloons, by the hundreds. Hell, spoofing GPS will probably do the trick.) Any signal can be jammed - especially a critically real-time one.
Barring Strong AI, the ultimate command to fire will remain under human (remote) control.
Navigation also requires a radio signal (unless you can shrink the mechanical inertial guidance system used in submarines to a flyable size; this is not only difficult but would likely raise the cost of the drones to heights extraordinary even by US military standards.)
Navigation does not require a radio system. Terrain Contour Matching aided by by Inertial Navigation Systems allows for fully autonomous navigation - They had this working with 1970s era Technology (Try and recall what a computer was like in the Mid 1970s and continue to be amazed at what engineers could get to work in a missile)
It was a big deal in Canada in the Late 80s/Early 90s - the United States used to test their (hopefully unarmed) cruise missiles over Canada - presumably we had a terrain profile that was useful to them. Or maybe it was the long stretches of unpopulated environment.
Now consider what your typical iPhone would be capable of doing. (Compass, GPS, Terrain Maps, Camera, Radio Signal...)
129 comments
[ 10.3 ms ] story [ 220 ms ] threadMore stuff like this on HN please.
http://www.startribune.com/local/68083742.html
In a combat aircraft the advantages for getting rid of the pilot are numerous: less weight & volume for the pilot and their associated support kit, eliminating the pilot allows for plane geometries that are just not possible with a human onboard, eliminating the pilot enables maneuvers that would turn a pilot to jelly, etc. The pilot is probably the most expensive component of a combat aircraft. Eliminating the pilot opens up new mission profiles (e.g. you don't have to worry too much about the aircraft getting back to base.)
For a commercial aircraft the benefits of going unmanned are not as pronounced and in some cases are non-existent. In addition to the lack of benefits, you introduce additional costs related to liability (you can always blame pilot error for a problem, but if there is no pilot then it is the airlines fault for either not maintaining the system properly or for selecting the system in the first place) and the fact that passengers are not going to trust this system anytime soon.
The impediments are almost completely social, but the benefits are so slight that it does not make much sense to bother aiming for completely unmanned. What you are going to see (and in fact what has been happening for years now) is the increasing automation of cockpit tasks. There will always be a human being "at the controls" for the forseeable future, but what that actually means and how much active control they are exercising is going to be slowly changing.
Great observation. The primary advantage I saw (reduced pilot errors, responsible now for ~3/4 of all aviation accidents) is a liability for the airline and the manufacturer.
I think this is a common misperception. I believe the operational risk is that manuevers above 6g (varying on pilot and G-suit) reduces blood flow to the brain (by forcing it into the legs) and cause a blackout. See http://en.wikipedia.org/wiki/G-LOC Blackouts, even for a few seconds, can be devastating in the middle of a dogfight. I believe there is software in modern fighter jets for just when pilots blackout but I can't find it on google right now.
It's got to exist on the web somewhere :-)
"Learn how hackers can bring down planes into buildings at 5"
When I moved from Vancouver to San Francisco I found it completely bizarre that they required Drivers for BART. It took me a couple years to get over that.
That reminds me of the 'Soft Walls' concept for fencing aircraft to where they belong:
http://softwalls.eecs.berkeley.edu/
It's like DRM for flight routes!
Radar guided missiles are easily defeated with modern ECM. Despite what the salesmen will tell you, in a real fight you have to get in and knife the f*&kers.
The same happens with remotely piloted vehicles. If you ever go up against even a modestly sophisticated adversary they will show phenomenal pluck and ingenuity. See serbian downing of an F-117 for an example.
Human pilots will be needed for some time to avoid having your toys fall out of the sky or stolen and used against you.
It was true back then, too - it was a good 25 years between when aircraft carriers were invented in WW1 and when they sank their first vessel at Coral Sea. But new technology opens up new tactics and new economies of scale that can fundamentally change the rules of battle.
What if you didn't care about having your toys fall out of the sky, for example? Make them so cheap that you can literally blanket the sky with them. Send up thousands of computer-piloted missile-aircraft. If they get shot down, who cares? If they shoot each other down, who cares? If they get stolen by the enemy and used to shoot down your own missile-craft...well, that's what was happening before anyway.
But you could throw up a curtain so dense and so shifting that it's impossible for anything to fly through them. And that's exactly what a potential adversary of the U.S. wants: nullify American air power so the f#&kers on the ground can knife our guys. It doesn't matter if they shoot down 10 of their own missile-craft for every one of ours. If they can put up 100x as many, that still equals no aircraft in the sky for Uncle Sam.
There was a recent wargame that followed a strategy much like this, except that instead of computer-controlled aircraft it used human-controlled small boats launching cruise missiles (that other automated-pilot air weapon):
http://en.wikipedia.org/wiki/Millennium_Challenge_2002
US air superiority is a system that includes AWACS (enabling better situational awareness), GPS (enabling JDAM/SDB for example), tankers, and a constellation of factors other than "mano a mano" with bullets. UAVs are already an integral part of the mix.
There are a few details on http://www.strategypage.com/htmw/htada/articles/20051121.asp...
First a primer on websites reporting military/intelligence affairs: two sites that occupy the ends of the credibility spectrum, http://smallwarsjournal.com and http://www.debka.com.
The first is a forum for junior military officers and professional analysts for discussing tactics and strategy. The articles all have by-lines, mostly by Captains and Majors (i.e. recently promoted Captains). The editorial board is made up of professionals. You will never read any classified information here. You may read some good war stories, but mostly you will read professional treatments of modern tactics by professionals for professionals.
On the other extreme many consider debka a conduit for the Mossad. Good luck separating the information from the disinformation on this site, but the articles can be a good read, if you can deal with the wilderness of mirrors.
In between these extremes keep in mind that intelligence analysis is about connecting the dots. Even the professionals seldom know anything for sure, and sources and means of gathering information are usually more important than any individual datum. If it makes it to print (or html), there’s probably a reason.
So where does this leave http://www.strategypage.com ? Let’s look at the staff bios on the about us page. Of the four "staff" exactly one has any military/intelligence background at all. "...author (over 20 books), wargame designer (over 100 designed and publisher of over 500), defense advisor (since the 1970s), pundit (since the 1970s) and general troublemaker. Served in the US Army artillery (1961-64) as a Spec/4." His website, http://www.jimdunnigan.com , lists some of his books, and more bio info including "Various Department of Defense and State Department agencies called on me to give lectures or just have a chat." I guess that was his "defense advisor" gig. He’s certainly a dedicated amateur. If luck had broken his way a little better, perhaps he could have been Tom Clancy. There’s also a handy "write for us" page. If you write well enough and don’t make any flaming technical errors, you can make some coin.
In short, I don’t think this site would even attract placing any real disinformation. So how about the article itself?
First, no by-line.
Secondly the protagonist’s name, "Zoltan". Sure sounds made up to me, but that’s just my opinion.
The way the piece is put together has the feel of a composite, somebody who did some research and threw this story together. If a real reporter got the real story from a real Colonel (and covered up his identity with a pseudonym) he could have written for a more prestigious publication or wire service and made better money. Still...not entirely implausible.
For now I’m sticking to my theory of the shoot-down only because accounts I read right after the incident described the wreckage of the aircraft landing largely intact (a little hard to believe since the fuselage is composite) with bullet holes and, if I recall correctly, the pilot ejected and survived (a little less likely after a missile strike). Hence my belief in heavy machine gun fire from a pursuit aircraft. Also the Russians really wanted to get their hands on one of these babies, even a crashed specimen. So it’s easy to imagine their risking an agent inside NATO to pull this off.
Reading this stuff on the web is not like reading hacking tips. The background is more important than the foreground, and you can never be sure.
I understand your point about the background of the website. However, in this case I thought it was a fairly well known story, and there are quite a number of other accounts.
eg: ""At times, they acted like amateurs," Dani said, listing some ways the Serbs managed to breach NATO communications security, including eavesdropping on pilots' conversations with AWACS surveillance planes.
"I personally listened to their pilots' conversations, learning about their routes and bombing plans," Dani said."
http://www.usatoday.com/news/world/2005-10-26-serbSerbian-st...
And:
"A network of spies and observers would report F-117A takeoffs from Italy’s Aviano Air Base. Coupling this intelligence with tweaked radars, the Serbians fired several surface-to-air missiles almost blindly, hoping for a hit."
http://www.airforcetimes.com/news/2008/04/airforce_nighthawk...
I thought that it was shot down by SAM missiles was generally accepted. The NATO spy part was more interesting, because there was reasonable support for the Serbs within some parts of NATO, so that isn't out of the question.
The "people listening for aircraft" theory makes a lot of sense, too, though - it's easy, and it's been proven to work before.
(Also, Zoltan is a reasonably common Hungarian name: http://www.google.com.au/search?q=name+Zoltan or http://nameberry.com/babyname/Zoltan - as you are no doubt aware, Hungarian names are quite common is some parts of the former Yugoslavia, and Zoltan Danni is of Hungarian decent)
Today planes can take off and land using only software and radio signals, even when the pilot can't, like no visibility when cloudy.
The problem is that computer understanding of the surroundings is NULL. Today there is no software that could understand(well) that there are birds(or small planes) on the way that are going to be eaten by your engines, and damage it.
The only technologies that work are active, like LIDAR, this is going to change, but first on vehicles with less degrees of freedom. Trains(1FD) before ships(2FD) before cars before planes.
This is just fearmongering. Birds, sure, but airplane engines are designed to ingest small birds and for the most part can handle it reasonably well. Small planes still must have a transponder, and TCAS systems will identify and warn of ADS-B signals ("squawk"), even if the smaller plane has not been equipped with a TCAS transceiver.
There is a rule in engineering, never underestimate a 900km/hr projectile. And I have seen myself engines birds test that are mandatory on every turbine.
Unless something like this happens: http://en.wikipedia.org/wiki/United_Airlines_Flight_232 . Then there's absolutely no replacement for a human being or three in the cockpit, and the public and the FAA know it. A pilot's salary or two is not too much to pay to protect the lives of several hundred people once in a while, even if it's otherwise unnecessary.
The military believes it, too, and the jets they fly have a higher likelihood of encountering life or death situations such as having pieces shot off and needing to get home. While there may not be pilots in the jets in the next generation, they'll be flying them for a while yet, and commanding them for a long time. The battlefield is too difficult, complex, and important for anything else. Once someone gives the order to drop a bomb, they'll trust software to execute the order efficiently and correctly. But it's going to be a human being giving the order for a long time. Possibly forever.
There are some jobs that won't be done by software until it can pass a Turing test. Software may be able to do the job in the vast majority of cases, and in some societies that may be good enough. Socially speaking, in this society, it isn't.
Lawn-motor robots? Perhaps arduino-controlled pistol gun with belt feed bullets?
I am worried about the future ability of a population's to resist a technological tyranny.
If you put all the power the state can muster on one pile and that of the population on another, resistance against the state has been impossible – I would guess – for a few centuries (or more). But that is – again – not how it works. There are other players involved. Parts of the state might not be willing to use all that power. Other states (sometimes with vastly more power) will constrain the ability of the state to use all its power. And so on.
The consequence of all this is that revolutions against tyrannies are possible without firing a single shot or using any violence at all. Even though, just looking at the raw power of the two sides, the state could easily crush the population. Just think of the fall of the Berlin Wall.
No one should consider themselves informed on the subject of robotic tyranny without having read this:
http://vinay.howtolivewiki.com/blog/global/the-second-amendm...
"...tell me, how good does a combat robot have to be to kick in your door, spray the house with mace, and hit anything that coughs with a taser until it stops moving and can be encased in spray webbing for later pickup and processing? How goes does a robot have to be to detain anybody who smells like pot? How good does a robot have to be to break up a political demonstration? Not that good. If a robot is good enough to fight wars, its good enough to do political policing. It might not solve crimes, but it can bust heads that protest and capture people who’s position is known and who have been identified as enemies of the state."
I see terrorist tactics (blowing up buildings, schools, etc.) as dominating their efforts.
That said, I think that if I were to do a defense startup now, it'd be towards detecting explosives, guns, etc. in large groups of people. If that technology falls behind, war's about to get a whole lot uglier.
Articles like this are interesting, but a little sickening in that they assume that continually increasing defense spending is actually a good idea.
I'd work for Swoopo long before I would work for a defense startup.
Do you honestly believe that? I don't see how it's even possible unless we sedate everyone in some kind of matrix like false utopian state. It's good to dream though.
If peak oil is true, I would predict quite the opposite.
I really doubt that. (Well, as long as we limit ourselves to times when there were Homo Sapiens in Europe.)
Also, see: http://en.wikipedia.org/wiki/Democratic_peace_theory
Name one.
http://www.unrv.com/tacitus/tacitus-germania-3.php
Once (if) that goes away, Western Europe will either have to pony up for it's own defense, or be held hostage to outside powers with their own agendas (like Russia blackmailing Europe with gas supplies)
I don't see the struggle between national interests going anyway anytime unless and until the nation-state itself is rendered obsolete. People need stuff, and that creates conflict. Different groups of people will need different stuff. And let's not forget ideological problems -- Western Europe isn't exactly a model of social integration.
It only takes one bad actor to make all participants gear up for the potential to use force, and we've found that once geared up, the actual use of force is not far behind.
It's a prisoner's dilemma played out on a global stage. But you'd find the same situation with three thousand guys living in caves somewhere. War is a natural consequence of groups of unique and intelligent humans.
So your nation can be so inept as to not be able to control things inside it's borders (many nations fit this bill). If armed camps build up inside them and attack other countries, guess what? You're at war -- or something so close to war as to be indistinguishable from it.
If you'd like the world to work like some super-big version of your local country, where most conflicts are solved by some civil system, then you're going to have to have popular choice of representatives, mixed local, national, and world control, an independent court system built on something besides treaties, etc. The UN is nowhere near any of that.
There is the theory that democracies only compete via commerce. If you buy into that, then perhaps a world full of democracies would be the most peaceful. But I don't see that happening either any time soon.
It's just a long-shot. A real long-shot. I'd argue that in order to do away with war you would have to do away with 99.999% of the impetus for using violence under any circumstances. It'd be a world of extreme pacifists. Would that be such a great thing? Doesn't chaos -- whether war, sickness, disasters, etc -- contribute to evolution and growth? Do we really want to live in stasis? Or -- to put it in religious terms -- is heaven really such a great thing?
In any case, it is one thing to have stability in an affluent, post-religion area like Western Europe, quite another to have it worldwide. Many nations in the Mid East, for example, are comparable to medieval Europe - except that if the current regimes fall (as they might well, if oil loses its value) their subjects are more likely to create a Muslim Brotherhood-type regime than a liberal republic.
http://www.ted.com/talks/lang/eng/steven_pinker_on_the_myth_...
I can see the courageous absolutism of Gandhi working well on a small scale inspiring others to rebel against the status quo - but against a nation bent on destruction of your own people, say, what then.
What I find intriguing is that in a situation of combat between computer-piloted UAV it is purely technological advance that gives the advantage. That then is dissociating war from direct violence and making technology the battleground rather than military aggression and effective killing.
Sadly I don't believe, for now, that we can end human warring in favour of "robot wars".
You can't extrapolate from Gandhi to a generalized theory of resistance. His strategy worked well against the British Empire at a time that it had already decided that empires were more trouble than they're worth, and besides, we only came here to play cricket anyway. Any other imperial power would simply have shot him.
That's what I said. Worked small scale but won't work large scale = can't extrapolate. Did I miss something?
Policing will probably always be needed, but that is different. The police are not at war with criminals.
1) Falkland's war between Argentina and Great Britain in 1982. 2) Greco-Turkish war in Cypress. 3) Indian-Pakistani wars (although this example is stretching it, as US influence in India is quite low) 4) The First and Second Cod War 5) The Turbot War
World peace could arise as the result of some extremely democratic/libertarian/scientific worldwide monoculture that agreed on everything but what they were willing to settle with words. I think that an unlikely end to the story, though. The easiest road to world peace is for someone to win. And that's not a happy ending in my book. I'll take war over that. It's nasty, but nowhere near as bad as tyranny.
Put another way, there will be war as long as there is crime. I hope that's forever. Living in a society that had actually done what it takes to eliminate crime completely would not be worth it to me. I feel the same way about a world without war.
It think we do have a chance to end wars once and for all as economies get more interconnected. Traditional geostrategic thinking is obsolete.
It's fun to imagine that somewhere sometime a scrawny geek told a chiseled-jaw pilot "Go away or I will replace you with a very small shell script"
I heard an interview with "Wired for War"'s author who claimed that controls for some of these systems are modeled after X-box/PlayStation controls, both to capitalize on the ergonomics research done by console makers and to make it easier for young recruits who are already used to operating them.
indeed.
That may have been my late father's fantasy. He was the lead engineer at Boeing Military for much of the B-52's avionics. He was not a man much given to resentments and anger, but he strongly disliked USAF pilots, and it was clear the other engineers did as well. They apparently are quite the prima donnas, and arrogant in the extreme.
This engineers vs. pilots trope goes back to the days of the space program. Tom Wolfe talked about it in _The Right Stuff_.
Interestingly enough this already happened to the Airforce in the late 60s. Airforce was running manned spy satellite missions, but because of the cost and the danger involved, those missions were being canceled in favor of unmanned satellite missions. So this is basically a repeat for Airforce.
The war machine expenditure in the USA is more than EVERY OTHER country on earth COMBINED (and it's deficit spending).
The problem is the war machine is always chomping at the bit to get into a war so they can use their toys. Did you know Japan has a military? And they are protesting against their constitution which prevents them from getting involved in other wars. They WANT to go to war, somewhere, anywhere. It's crazy.
Meanwhile the UK govt is cutting Infantry battalions and investing 10s of billions of GBP in Eurofighter and Nimrod. Because the Taliban's airforce and submarine fleet are obviously the biggest threats we face...
Note also that it's only the USAF who require trained pilots to operate UAVs... Everyone else uses automated landing systems (etc) and regular troops fly them. It's one of the featured jobs in the British Army's recruiting campaign, there's a guy with a handset similar to an Xbox controller scouting ahead of his unit as they patrol.
Commercial planes use linux for non-critical functions like the in-flight entertainment. I can't think of anything that would be on a fighter plane that would be considered "non-critical" enough to not require a high-integrity system though.
I wouldn't claim it's common or that you'd be likely to see it in a deployed environment, but it's not totally unheard of. Throw Linux & UAV in Google - you'll see what I mean.
I've definitely seen Linux (re?)booting on in-flight entertainment systems.
http://news.ycombinator.com/item?id=898214
I really need some real non-american references to believe in such a thing.
Whether the quality of American pilots was decisive in any of those conflicts is debatable, but the first part is certainly true. Many observers believe the Eurofighter Typhoon to be the most advanced fighter in the world today.
If nothing else you'll learn about John Boyd, one hell of a model American.
I referenced Top Gun and Beyond in an earlier post: if you can find the video, it goes into this.
Pretty amazing stuff. Most missions in Iraq and Afghanistan are flown via satellite from the US, truly remote controlled war. (Submitted it to HN at the time, but it only got 2 points ;))
On the plus side:
they could be salted about the place in a powered off state. better policing of gang-ridden areas.On the minus side:
public, potentially developing a culture of impunity - spread of surveillanceNavigation also requires a radio signal (unless you can shrink the mechanical inertial guidance system used in submarines to a flyable size; this is not only difficult but would likely raise the cost of the drones to heights extraordinary even by US military standards.)
It was a big deal in Canada in the Late 80s/Early 90s - the United States used to test their (hopefully unarmed) cruise missiles over Canada - presumably we had a terrain profile that was useful to them. Or maybe it was the long stretches of unpopulated environment.
Now consider what your typical iPhone would be capable of doing. (Compass, GPS, Terrain Maps, Camera, Radio Signal...)
Everyone would have died on a computer plane.