You seem to assuming a very simple wing structure here. Even without warpable wings (which can be implemented using flexible materials), you can make the flapping hinge adjustable to change the sweep pattern. While I don't know of any existing implementations of a hover ability, there's a lot of work being done on the subject and it certainly isn't impossible. I think we're back to the same issue as before: it can't be done off-the-shelf, but the hurdles aren't impossible to get over with some more research. Like the research the military's funding here.
When you've got a UAV, the software has to do the tricky work. Even with remote control, it's difficult to execute a landing with only a camera view of what's going on. Without that human element, the chances of a failed landing skyrocket. If all you have is a camera, yes. When you add in a 3-axis accelerometer and infrared or ultrasonic proximity sensors, it becomes a lot more reasonable. Software is actually pretty good at doing things like landing aircraft, it's the inadequate sensor data that's the problem. That's why, I admit, landing in trees might be farfetched for the moment, but all the others are entirely plausible.
Deserts are actually worse. Putting aside how easy it is for a 6 inch ornithopter to sink into the sand, it will be incredibly easy to spot for anyone on patrol in the area. Yes, if there happens to be someone on patrol in the particularly secluded patch of desert you chose, and they don't assume it's a piece of trash or, for that matter, a dead bat. And why would a six-inch ornithopter be in particular danger of sinking? Something light but with a large surface area is in less danger, not more. At no point do they say that the thing will be completely autonomous, just that it will be able to navigate on its own. I don't doubt that the commanders will maintain an ability to switch to manual flight control.
And as regarding the money issue, I still think my point stands, but if you read the effing press release that TFA is based on, you'll see that UMich is just using its 10M to work on miniaturizing the electronics; there are at least two more universities involved in the overall project, each of which I would guess have their own grants. http://www.ns.umich.edu/htdocs/releases/story.php?id=6409
On an unrelated note, I'd really like to know who made those 3d models, and who made the ludicrous decision to put them up with the press release. There are so many problems with the design of that model that I actually am a little concerned with their competence to do what they're talking about.
Taking off and landing automatically on runways. Have you heard of these magical new devices called helicopters? Beyond that, one of the benefits of ornithopters over fixed-wing aircraft is that they have the potential to take off like birds do; i.e., almost vertically.
The most common result? Capsizing. Have you ever actually used one of those RC dragonflies? I've landed them plenty of times without capsizing, and when I have flubbed the landing, all I had to do was flip it over and it was ready to go again. Do you really think that they couldn't manage an autorighting system with $10 million to play with?
I'd like to see their toy make regular landings on level ground. Until that's solved, rooftops and trees are out. And forested clearings are exposed areas where the enemy might spot your drone. (Assuming you can even find a forested clearing in our current theaters of operation.;-)) Setting aside the landing issue, since many designs (yes, not traditional fixed-wing, but that's why the military is looking at alternatives) can make regular landings, if you land far enough away from populated areas the risk will be low. And I mentioned deserts as well as forests, if you'll notice.
While yes, 10M isn't that much by the standards of many military devices, for the sorts of prototyping and tech development we're talking, over the next five years, that's not an unreasonable sum, especially since I'm sure they're putting lots more money into other groups researching very relevant tech. I also significantly doubt that the military is looking for a 10M a pop recon drone. There's too much risk of damage. We're not talking about a Predator, here, which is the size of a Cessna (and which, btw, only costs about $3 million), and which only has to worry about AA fire.
we don't have UAVs that can land and takeoff unattended inside enemy territory This isn't a technological issue. We have plenty of designs capable of taking off and landing autonomously wherever it's needed; I don't know whether any are currently deployed, but there's no reason we couldn't.
Yet the military thinks that this magical ornithopter is going to manage takeoff and landing unattended? (Which is significantly complicated by its wing design.) While this is complicated by its wing design compared to a helicopter, it's actually easier compared to a fixed-wing drone. You don't need an extended runway; hell, those little WowWee dragonfly RC toys can take off from ground in about five feet.
On top of that, the military really expects that these things will lay out in the open (where they can get sunlight) and go completely undetected? On rooftops, in trees, in forested clearings or desert away from people - yes, I think it could find places out of the way to sit and recharge during the day.
You're right that those specs aren't possible with current technology, but I suspect that's why they're giving the University of Michigan $10 million to try to improve the current technology.
If you want a military recon drone that only has enough battery capacity to fly for ten minutes and can barely lift more than its own weight before you add the camera and long-range transceiver, sure. 10 million isn't out of line to do what the military's asking for, here, because they're asking for quite a bit. As for using real bats, interfacing control circuitry into a brain to a sufficient degree to get more than "turn right, now turn left" is a hell of a lot more complicated of a problem then miniaturizing batteries and electronics packages.
Depending on the wing design and control software, ornithopters can actually be a lot more maneuverable than fixed wing aircraft. If they want an ability to go indoors, fixed-wing is pretty much out, and helicopters both burn energy like crazy and tend to a bit more obvious.
Since the article says it's supposed to be long-range, my guess is that the mission profile would be to sit somewhere out of the way and charge during the day, then do its recon at night. All of those methods are very useful, because it means the drone could stay in an area and continue to operate without human intervention nearly indefinitely.
Oh, I completely agree that there are excellent instrumental reasons to found a system of government on the basis of natural rights; I acknowledged as much in one of the postings or other in this thread. I was just objecting to the original poster's assertion of the fact of natural rights without anything to back it up.
Personally, I believe in a form of rule utilitarianism, but I also think that that system is entirely subjective. I'm a materialist (in the philosophical sense), and scientifically I don't see any evidence for people being anything other than a particularly smart kind of animal. I think that there's an emotion-based faculty for moral thinking hard-wired into our brains because it's useful in allowing individuals to survive in group situations, and as such we can't help but assign moral value to some actions. I hope that we can use reason to shape those moral feelings in directions which maximize human happiness and fulfillment. I have no basis for structuring my ethics around happiness other then the fact that I and the vast majority of other people seem to want it, but there you have it. I don't see how anyone else is basing their ethics on anything more solid.
I genuinely am interested in how to develop an objective moral system without some kind of supreme being. How do you see Kant as doing that? I fully admit that I may just not get his argument - it's hard reading, and it wouldn't be the first time that something completely legitimate just went way over my head.
I've read some Kant, and debated extensively with a Kantian. The same problem applies: he makes these hand-waving assertions about how reason will inevitably guide us to an objective moral standard, but doesn't actually take us through any such pathway. Universalizability is a fine maxim, but I still see no reason why I'm bound to follow it.
I understand full well what I mean, and I don't intend to qualify it. You're the one who's confusing a descriptive argument with a normative one. Do I want to be your slave? No. But if all of society rose up and made me your slave, there's not much I could do to prevent it. Just because I don't like a state of affairs doesn't mean it isn't factual. And just because I don't like a state of affairs doesn't mean that there is some cosmic law forbidding it.
I'm also not saying that "majority rules" is some kind of natural state. People have to agree to abide by majority decisions. In the absence of such an agreement, people will do whatever they can force or convince others into allowing them to do.
I know that what you're saying is the cornerstone of natural rights philosophy. I disagree with natural rights philosophy. It's easy to say that with intelligence and education the nature of our objective rights becomes clear, but it's much harder to demonstrate in practice how or why this is the case.
And no, I don't have to see something to have reason to think it exists. I'm not that extreme of a verificationist. But I don't see any evidence, other than a long-standing belief among many people to that effect, that there are such things as objective rights out in the world. If you have such evidence, I'd love to see it. And no, quoting the Declaration of Independence won't do. I recognize that the founding fathers were believers in natural rights philosophy, but I also think that it's ok to disagree with the founding fathers. If reasoning self-evidently points any rational observer to the belief that there are natural rights, then show me the reasoning? Tell me what book to read, what philosopher to learn about, that will tell me why I should agree with you. That's all I'm asking.
I'm troubled by the confidence with which you "correct" people on the nature of rights. It's fine to go around saying that "rights exist apart from law", but last time I checked that was a position, not a fact. Maybe you're right, and maybe you're wrong. Personally, I don't keep my rights stored in the linen closet in case they're needed, and I have yet to run into one on street, or hell, see any evidence for their existence at all. As far as I can see, people are people, and they have to decide how to deal with each other. We can either all just do whatever we feel like, to the extent that we can force or convince others to go along with it, or we can lay down some ground rules regarding what we can and cannot do to one another. By that understanding, rights are what we make of them. To so confidently assert that you magically know what people's rights are closes down the democratic process, because you're not the only one out there who thinks he knows what his "natural" rights are, and those others aren't necessarily going to agree with you. Setting down certain rights as inviolate (i.e., in constitutions) is an excellent idea, because it curbs the sudden impulses of society away from actions they'll later regret. But to assert so confidently that you know the nature of things merely blinds people - he's trying to take away my RIGHTS! Burn him! - and makes them unable to rationally assess the situation.
Hell, maybe it's better as a rule that people think they have natural rights, as a kind of super-constitution. However, I'd hoped for a little more complex of a dialog here, or at least some reasons why I should believe what you say.
I suppose that makes some sense, though I wouldn't have thought that the RCS motors would be able to get the capsule far enough fast enough. Regarding landing by rocket power, though, I thought that while there is a debate between airbags or retrorockets, either one is used in conjunction with the parachute system. The retrorockets would be designed to slow down a parachute-assisted landing, which would be IIRC around 18 mph. You're going to end up going much faster than that from a ~350 foot fall, which is the minimum we'd be talking about. Carrying enough fuel inside the capsule to provide that much delta-v seems impractical, especially for a purely emergency purpose. It seems like you'd also be looking at significant stabilization issues trying to keep the capsule from tumbling, which the RCS motors wouldn't have a lot of time to fix.
The summary doesn't describe the system itself very well - if that was how it worked I'd agree it'd be idiotic. The "vehicle" the summary mentions is actually just a separate rocket engine attached to the nose of the capsule. If something goes wrong, the astronauts don't have to go anywhere; the bolts holding the capsule onto the main Ares launch vehicle blow, and the escape rocket fires, lifting the entire Orion capsule off the Ares rocket and high enough into the air to get clear of the launch pad and any unpleasant explosions. Then the escape rocket separates from the capsule, while the capsule is hopefully high enough to land softly by parachute. For more info (and pictures), see here: http://en.wikipedia.org/wiki/Launch_escape_system and here: http://www.astronautix.com/craft/orionlas.htm.
The shuttle would be somewhat safer with ejection seats, but the rocket-based launch escape system the article's talking about really is a better option. Ejection seats wouldn't do a whole lot of good if there was a failure on the launch pad, and you'd really need some kind of ejection pod system for bailing at high speed and altitude, which would add a lot of weight, complexity, and possible points of failure. As for the existing return to launch site abort plan, as well as the other various abort plans, what's wrong with them? Other than requiring a generally intact shuttle, that is. Anyway, Orion's planned abort system gets around all those problems. While I don't necessarily agree with the whole Constellation plan, putting the crew on top of the stack is a good choice.
Slashdot Mirror
When you've got a UAV, the software has to do the tricky work. Even with remote control, it's difficult to execute a landing with only a camera view of what's going on. Without that human element, the chances of a failed landing skyrocket. If all you have is a camera, yes. When you add in a 3-axis accelerometer and infrared or ultrasonic proximity sensors, it becomes a lot more reasonable. Software is actually pretty good at doing things like landing aircraft, it's the inadequate sensor data that's the problem. That's why, I admit, landing in trees might be farfetched for the moment, but all the others are entirely plausible.
Deserts are actually worse. Putting aside how easy it is for a 6 inch ornithopter to sink into the sand, it will be incredibly easy to spot for anyone on patrol in the area. Yes, if there happens to be someone on patrol in the particularly secluded patch of desert you chose, and they don't assume it's a piece of trash or, for that matter, a dead bat. And why would a six-inch ornithopter be in particular danger of sinking? Something light but with a large surface area is in less danger, not more. At no point do they say that the thing will be completely autonomous, just that it will be able to navigate on its own. I don't doubt that the commanders will maintain an ability to switch to manual flight control.
And as regarding the money issue, I still think my point stands, but if you read the effing press release that TFA is based on, you'll see that UMich is just using its 10M to work on miniaturizing the electronics; there are at least two more universities involved in the overall project, each of which I would guess have their own grants. http://www.ns.umich.edu/htdocs/releases/story.php?id=6409
On an unrelated note, I'd really like to know who made those 3d models, and who made the ludicrous decision to put them up with the press release. There are so many problems with the design of that model that I actually am a little concerned with their competence to do what they're talking about.
The most common result? Capsizing. Have you ever actually used one of those RC dragonflies? I've landed them plenty of times without capsizing, and when I have flubbed the landing, all I had to do was flip it over and it was ready to go again. Do you really think that they couldn't manage an autorighting system with $10 million to play with?
I'd like to see their toy make regular landings on level ground. Until that's solved, rooftops and trees are out. And forested clearings are exposed areas where the enemy might spot your drone. (Assuming you can even find a forested clearing in our current theaters of operation.
While yes, 10M isn't that much by the standards of many military devices, for the sorts of prototyping and tech development we're talking, over the next five years, that's not an unreasonable sum, especially since I'm sure they're putting lots more money into other groups researching very relevant tech. I also significantly doubt that the military is looking for a 10M a pop recon drone. There's too much risk of damage. We're not talking about a Predator, here, which is the size of a Cessna (and which, btw, only costs about $3 million), and which only has to worry about AA fire.
Yet the military thinks that this magical ornithopter is going to manage takeoff and landing unattended? (Which is significantly complicated by its wing design.) While this is complicated by its wing design compared to a helicopter, it's actually easier compared to a fixed-wing drone. You don't need an extended runway; hell, those little WowWee dragonfly RC toys can take off from ground in about five feet.
On top of that, the military really expects that these things will lay out in the open (where they can get sunlight) and go completely undetected? On rooftops, in trees, in forested clearings or desert away from people - yes, I think it could find places out of the way to sit and recharge during the day.
You're right that those specs aren't possible with current technology, but I suspect that's why they're giving the University of Michigan $10 million to try to improve the current technology.
If you want a military recon drone that only has enough battery capacity to fly for ten minutes and can barely lift more than its own weight before you add the camera and long-range transceiver, sure. 10 million isn't out of line to do what the military's asking for, here, because they're asking for quite a bit. As for using real bats, interfacing control circuitry into a brain to a sufficient degree to get more than "turn right, now turn left" is a hell of a lot more complicated of a problem then miniaturizing batteries and electronics packages.
Depending on the wing design and control software, ornithopters can actually be a lot more maneuverable than fixed wing aircraft. If they want an ability to go indoors, fixed-wing is pretty much out, and helicopters both burn energy like crazy and tend to a bit more obvious.
Since the article says it's supposed to be long-range, my guess is that the mission profile would be to sit somewhere out of the way and charge during the day, then do its recon at night. All of those methods are very useful, because it means the drone could stay in an area and continue to operate without human intervention nearly indefinitely.
Oh, I completely agree that there are excellent instrumental reasons to found a system of government on the basis of natural rights; I acknowledged as much in one of the postings or other in this thread. I was just objecting to the original poster's assertion of the fact of natural rights without anything to back it up.
Personally, I believe in a form of rule utilitarianism, but I also think that that system is entirely subjective. I'm a materialist (in the philosophical sense), and scientifically I don't see any evidence for people being anything other than a particularly smart kind of animal. I think that there's an emotion-based faculty for moral thinking hard-wired into our brains because it's useful in allowing individuals to survive in group situations, and as such we can't help but assign moral value to some actions. I hope that we can use reason to shape those moral feelings in directions which maximize human happiness and fulfillment. I have no basis for structuring my ethics around happiness other then the fact that I and the vast majority of other people seem to want it, but there you have it. I don't see how anyone else is basing their ethics on anything more solid.
I genuinely am interested in how to develop an objective moral system without some kind of supreme being. How do you see Kant as doing that? I fully admit that I may just not get his argument - it's hard reading, and it wouldn't be the first time that something completely legitimate just went way over my head.
I've read some Kant, and debated extensively with a Kantian. The same problem applies: he makes these hand-waving assertions about how reason will inevitably guide us to an objective moral standard, but doesn't actually take us through any such pathway. Universalizability is a fine maxim, but I still see no reason why I'm bound to follow it.
I understand full well what I mean, and I don't intend to qualify it. You're the one who's confusing a descriptive argument with a normative one. Do I want to be your slave? No. But if all of society rose up and made me your slave, there's not much I could do to prevent it. Just because I don't like a state of affairs doesn't mean it isn't factual. And just because I don't like a state of affairs doesn't mean that there is some cosmic law forbidding it.
I'm also not saying that "majority rules" is some kind of natural state. People have to agree to abide by majority decisions. In the absence of such an agreement, people will do whatever they can force or convince others into allowing them to do.
I know that what you're saying is the cornerstone of natural rights philosophy. I disagree with natural rights philosophy. It's easy to say that with intelligence and education the nature of our objective rights becomes clear, but it's much harder to demonstrate in practice how or why this is the case.
And no, I don't have to see something to have reason to think it exists. I'm not that extreme of a verificationist. But I don't see any evidence, other than a long-standing belief among many people to that effect, that there are such things as objective rights out in the world. If you have such evidence, I'd love to see it. And no, quoting the Declaration of Independence won't do. I recognize that the founding fathers were believers in natural rights philosophy, but I also think that it's ok to disagree with the founding fathers. If reasoning self-evidently points any rational observer to the belief that there are natural rights, then show me the reasoning? Tell me what book to read, what philosopher to learn about, that will tell me why I should agree with you. That's all I'm asking.
I'm troubled by the confidence with which you "correct" people on the nature of rights. It's fine to go around saying that "rights exist apart from law", but last time I checked that was a position, not a fact. Maybe you're right, and maybe you're wrong. Personally, I don't keep my rights stored in the linen closet in case they're needed, and I have yet to run into one on street, or hell, see any evidence for their existence at all. As far as I can see, people are people, and they have to decide how to deal with each other. We can either all just do whatever we feel like, to the extent that we can force or convince others to go along with it, or we can lay down some ground rules regarding what we can and cannot do to one another. By that understanding, rights are what we make of them. To so confidently assert that you magically know what people's rights are closes down the democratic process, because you're not the only one out there who thinks he knows what his "natural" rights are, and those others aren't necessarily going to agree with you. Setting down certain rights as inviolate (i.e., in constitutions) is an excellent idea, because it curbs the sudden impulses of society away from actions they'll later regret. But to assert so confidently that you know the nature of things merely blinds people - he's trying to take away my RIGHTS! Burn him! - and makes them unable to rationally assess the situation.
Hell, maybe it's better as a rule that people think they have natural rights, as a kind of super-constitution. However, I'd hoped for a little more complex of a dialog here, or at least some reasons why I should believe what you say.
I suppose that makes some sense, though I wouldn't have thought that the RCS motors would be able to get the capsule far enough fast enough. Regarding landing by rocket power, though, I thought that while there is a debate between airbags or retrorockets, either one is used in conjunction with the parachute system. The retrorockets would be designed to slow down a parachute-assisted landing, which would be IIRC around 18 mph. You're going to end up going much faster than that from a ~350 foot fall, which is the minimum we'd be talking about. Carrying enough fuel inside the capsule to provide that much delta-v seems impractical, especially for a purely emergency purpose. It seems like you'd also be looking at significant stabilization issues trying to keep the capsule from tumbling, which the RCS motors wouldn't have a lot of time to fix.
The summary doesn't describe the system itself very well - if that was how it worked I'd agree it'd be idiotic. The "vehicle" the summary mentions is actually just a separate rocket engine attached to the nose of the capsule. If something goes wrong, the astronauts don't have to go anywhere; the bolts holding the capsule onto the main Ares launch vehicle blow, and the escape rocket fires, lifting the entire Orion capsule off the Ares rocket and high enough into the air to get clear of the launch pad and any unpleasant explosions. Then the escape rocket separates from the capsule, while the capsule is hopefully high enough to land softly by parachute. For more info (and pictures), see here: http://en.wikipedia.org/wiki/Launch_escape_system and here: http://www.astronautix.com/craft/orionlas.htm.
The shuttle would be somewhat safer with ejection seats, but the rocket-based launch escape system the article's talking about really is a better option. Ejection seats wouldn't do a whole lot of good if there was a failure on the launch pad, and you'd really need some kind of ejection pod system for bailing at high speed and altitude, which would add a lot of weight, complexity, and possible points of failure. As for the existing return to launch site abort plan, as well as the other various abort plans, what's wrong with them? Other than requiring a generally intact shuttle, that is. Anyway, Orion's planned abort system gets around all those problems. While I don't necessarily agree with the whole Constellation plan, putting the crew on top of the stack is a good choice.