5 ms·

> The whole concept of manned jet fighters is laughably obsolete today. We can scarcely build an autonomous car that doesn't crash into static road barriers. C

> The whole concept of manned jet fighters is laughably obsolete today. We can scarcely build an autonomous car that doesn't crash into static road barriers. Call me skeptical.

antpls 8y ago
I was told unmanned flying in the 3D space is a lot easier to achieve than on a 2D plane with many obstacles. There is no road barriers in the air.
- nradov 8y ago
  Ok now detect and avoid a missile coming at you. That's a lot tougher than steering around a road hazard.
  - akmarinov 8y ago
    Isn’t missile lock detection already a feature on current planes?
    - nradov 8y ago
      Sort of. Some aircraft can detect targeting radar or missile launch thermal signatures to a limited extent from some directions. But it isn't very reliable or precise. Human eyeballs are still more effective in many cases.
      - antpls 8y ago
        Genuine question : how does the pilot can see the missile if the missile is chasing the plane or under the plane?
        tim333 8y ago
        They often don't. Fighters have missile detections systems of various types. https://en.wikipedia.org/wiki/Missile_approach_warning_system https://en.wikipedia.org/wiki/Missile_approach_warning_syste...
        antpls 8y ago
        That proves the direct parent is wrong about human eyeballs being more effective than detection systems.
  - antpls 8y ago
    I have no idea of the actual challenge, but given the missile itself is unmanned, a human would less be likely to escape if he is under pressure, stressed, or tired than an AI. Here the AI and human would have access to the same data to make a decision and would be able to equally control the plane's movements. Surely the AI is going to out-perform the human in this task.
    - mieseratte 8y ago
      > Surely the AI is going to out-perform the human in this task. Therein lies the problem. We don't have good AI to achieve these types of tasks and are not nearly close enough to achieving it for military applications. The notion that we can just make an autonomous killing machine at this stage is absolutely laughable marketing hype.
      - antpls 8y ago
        If you invest a few billions dollars to the task, you will have your specialized AI with all the sensors needed to out-perform human at this task in less than a decade. There is a reason Google employees refused to work for military.
        nradov 8y ago
        Your naive faith in software developers is touching, but this is entirely speculative. Until someone actually does it we don't know how much it will cost or how long it will take. Meanwhile we need to make plans based on technologies that are actually available, or at least have a clear development path.
        antpls 8y ago
        > Your naive faith in software developers is touching, but this is entirely speculative. Until someone actually does it we don't know how much it will cost or how long it will take. What about using those 1.4 trillions dollars to support R&D, businesses and help people to live longer and have a good life? What about investing 1.4 trillions in genome research? in cancer research? in renewable and nuclear energy production? Sorry, but I prefer to be a naive software developer helping connecting people together around the world than someone who helps random people to get kill. We are not in the cold war anymore, people don't communicate with papers. We can solve problems by other means than sending missiles to each others, we can see beyond the military propaganda of the countries. We know when something is possible but not preferable : killing drones are one of those project. Technically feasible, ethically not acceptable.
  - yogthos 8y ago
    No need, you just use numbers. Imagine a drone carrier platform that carries thousands of smart drone missiles and releases them in the target area. The only goal of the missiles is to lock onto anything that moves and hit it. Sure, you'll take a few out, but it's a game of numbers in the end.
    - nradov 8y ago
      A drone carrier platform that carries "thousands" of smart drone missiles would be too large and heavy for practical operations, or the missiles would be too small to have a useful range. The basic laws of physics still apply.
      - yogthos 8y ago
        If you couldn't have a single platform with useful range, you could just make many platforms each carrying 100s of missiles instead. Each missile could itself have a clustered warhead. These platforms would still have superior maneuverability to jet fighters, and you could put unshielded nuclear engines on them as Russia is doing right now with their new cruise missiles https://arstechnica.com/science/2018/03/best-bad-idea-ever-why-putins-nuclear-powered-missile-is-possible-and-awful/ https://arstechnica.com/science/2018/03/best-bad-idea-ever-w... This is something that's simply not possible to do if you have a human pilot who requires massive shielding and can only handle small Gs.
        Const-me 8y ago
        > as Russia is doing right now with their new cruise missiles Russia is not doing anything with missiles. They only have bad 3D renders. I don’t think modern Russia is able to design & manufacture even conventional jet planes: https://en.wikipedia.org/wiki/Sukhoi/HAL_FGFA https://en.wikipedia.org/wiki/Sukhoi/HAL_FGFA
        yogthos 8y ago
        https://abcnews.go.com/International/russia-tested-nuclear-missile/story?id=54123222 https://abcnews.go.com/International/russia-tested-nuclear-m...
- tim333 8y ago
  There's been some progress eg: In simulations " But a new artificial intelligence system, ALPHA, has been besting expert pilots in combat simulations, even when the A.I. is given a handicap." https://www.bbc.co.uk/news/technology-36650848 https://www.bbc.co.uk/news/technology-36650848 And with a pilotless F16 "During the exercises, the F-16 planned and executed an air strike according to "mission priorities and available assets." The F-16 also managed "dynamically react to a changing threat environment" while managing "capability failures, route deviations, and loss of communication" https://www.popularmechanics.com/military/research/a26028/f-16-drone-have-raider-ii/ https://www.popularmechanics.com/military/research/a26028/f-...

> The whole concept of manned jet fighters is laughably obsolete today. We can scarcely build an autonomous car that doesn't crash into static road barriers. Call me skeptical.

antpls 8y ago

I was told unmanned flying in the 3D space is a lot easier to achieve than on a 2D plane with many obstacles. There is no road barriers in the air.

nradov 8y ago
Ok now detect and avoid a missile coming at you. That's a lot tougher than steering around a road hazard.
- akmarinov 8y ago
  Isn’t missile lock detection already a feature on current planes?
  - nradov 8y ago
    Sort of. Some aircraft can detect targeting radar or missile launch thermal signatures to a limited extent from some directions. But it isn't very reliable or precise. Human eyeballs are still more effective in many cases.
    - antpls 8y ago
      Genuine question : how does the pilot can see the missile if the missile is chasing the plane or under the plane?
      - tim333 8y ago
        They often don't. Fighters have missile detections systems of various types. https://en.wikipedia.org/wiki/Missile_approach_warning_system https://en.wikipedia.org/wiki/Missile_approach_warning_syste...
        antpls 8y ago
        That proves the direct parent is wrong about human eyeballs being more effective than detection systems.
- antpls 8y ago
  I have no idea of the actual challenge, but given the missile itself is unmanned, a human would less be likely to escape if he is under pressure, stressed, or tired than an AI. Here the AI and human would have access to the same data to make a decision and would be able to equally control the plane's movements. Surely the AI is going to out-perform the human in this task.
  - mieseratte 8y ago
    > Surely the AI is going to out-perform the human in this task. Therein lies the problem. We don't have good AI to achieve these types of tasks and are not nearly close enough to achieving it for military applications. The notion that we can just make an autonomous killing machine at this stage is absolutely laughable marketing hype.
    - antpls 8y ago
      If you invest a few billions dollars to the task, you will have your specialized AI with all the sensors needed to out-perform human at this task in less than a decade. There is a reason Google employees refused to work for military.
      - nradov 8y ago
        Your naive faith in software developers is touching, but this is entirely speculative. Until someone actually does it we don't know how much it will cost or how long it will take. Meanwhile we need to make plans based on technologies that are actually available, or at least have a clear development path.
        antpls 8y ago
        > Your naive faith in software developers is touching, but this is entirely speculative. Until someone actually does it we don't know how much it will cost or how long it will take. What about using those 1.4 trillions dollars to support R&D, businesses and help people to live longer and have a good life? What about investing 1.4 trillions in genome research? in cancer research? in renewable and nuclear energy production? Sorry, but I prefer to be a naive software developer helping connecting people together around the world than someone who helps random people to get kill. We are not in the cold war anymore, people don't communicate with papers. We can solve problems by other means than sending missiles to each others, we can see beyond the military propaganda of the countries. We know when something is possible but not preferable : killing drones are one of those project. Technically feasible, ethically not acceptable.
- yogthos 8y ago
  No need, you just use numbers. Imagine a drone carrier platform that carries thousands of smart drone missiles and releases them in the target area. The only goal of the missiles is to lock onto anything that moves and hit it. Sure, you'll take a few out, but it's a game of numbers in the end.
  - nradov 8y ago
    A drone carrier platform that carries "thousands" of smart drone missiles would be too large and heavy for practical operations, or the missiles would be too small to have a useful range. The basic laws of physics still apply.
    - yogthos 8y ago
      If you couldn't have a single platform with useful range, you could just make many platforms each carrying 100s of missiles instead. Each missile could itself have a clustered warhead. These platforms would still have superior maneuverability to jet fighters, and you could put unshielded nuclear engines on them as Russia is doing right now with their new cruise missiles https://arstechnica.com/science/2018/03/best-bad-idea-ever-why-putins-nuclear-powered-missile-is-possible-and-awful/ https://arstechnica.com/science/2018/03/best-bad-idea-ever-w... This is something that's simply not possible to do if you have a human pilot who requires massive shielding and can only handle small Gs.
      - Const-me 8y ago
        > as Russia is doing right now with their new cruise missiles Russia is not doing anything with missiles. They only have bad 3D renders. I don’t think modern Russia is able to design & manufacture even conventional jet planes: https://en.wikipedia.org/wiki/Sukhoi/HAL_FGFA https://en.wikipedia.org/wiki/Sukhoi/HAL_FGFA
        yogthos 8y ago
        https://abcnews.go.com/International/russia-tested-nuclear-missile/story?id=54123222 https://abcnews.go.com/International/russia-tested-nuclear-m...
tim333 8y ago
There's been some progress eg: In simulations " But a new artificial intelligence system, ALPHA, has been besting expert pilots in combat simulations, even when the A.I. is given a handicap." https://www.bbc.co.uk/news/technology-36650848 https://www.bbc.co.uk/news/technology-36650848 And with a pilotless F16 "During the exercises, the F-16 planned and executed an air strike according to "mission priorities and available assets." The F-16 also managed "dynamically react to a changing threat environment" while managing "capability failures, route deviations, and loss of communication" https://www.popularmechanics.com/military/research/a26028/f-16-drone-have-raider-ii/ https://www.popularmechanics.com/military/research/a26028/f-...