The material of the tinfoil makes not much of a difference, but the shape certainly does. If you come up with a good stealth design, you should be able to fool any monostatic sonar (receiver and emitter close to each other) quite a bit.
The robot is as harmful to the bats as one bat is to the other. Its sonar beam is pretty narrow, pulsed in time, and has a short range only. A flying bat would get out of its reach within seconds at worst.
The situation with the dolphins and the sonar is different, because sound (of the same frequency) carries much further in water than in air. That and the fact that for electromagnetic waves it is the other way round is the reason why the Navy uses sonar and the Army doesn't.
There are certainly situations with so many bats so close to each other that they will likely jam each other, e.g. in a large colony. However, in most of these situations, the bats are "at home", so they don't need and - as anecdotal evidence suggests - probably don't read their sonar.
Since bat sonar is directional on the emission and reception side, as well as - for "resonably" specular targets - on the reflection side, is range-limited by absorption, and - as you point out correctly - there is a time window for reception, I expect that jamming is not much of a problem and bats probably do not require much sophistication in dealing with it.
For these reasons, I am not inclined to throw precious resources at this. Maybe I'll hand out a master thesis project one day to come up with an estimate of how close bats have to be packed for jamming to become a problem. But I expect you would end up with a "sardines in a tin" packing.
You are correct in pointing out that a single sensor cannot give "good" 3d results; they may be "accurate", but not unambiguous (all points on the surface of a prolate spheroid correspond to the same mouth-to-ear time-of-flight).
The array information from the individual cilia is - in the first place - exclusively a frequency-domain information, because the inner ear splits the incoming signal into its frequency components. The earshape can add a spatial dimension to this, because it can cause the ear to be sensitive in different directions for different frequencies.
So it is not a question of "either or", ear shapes and frequency analysis by the cilia work together.
Radar uses electromagnetic waves, in-air sonar uses ultrasound. Ultrasound frequencies are by definition above the audible spectrum, so you don't hear them.
One would hope, however, that we don't start relying too much on SONAR becuase it appears to be of limited functionality, can potentially cause noise pollution, and even alter the migration patters of bats themselves;) Either that or maybe it would drive dogs and cats crazy.
Don't worry about the noise pollution or the effect on other animals. Ultrasound has a very limited range in air. Because of this, bats can cope with other bats being around and probably don't even need to take any special precautions.
the technology is only going to be good at things that reflect sound.
In air, all "usual" (i.e., not specially prepared) solids and liquids will reflect ultrasound very well. The only problem are flat, "specular" surfaces, where the reflected sound is directed away from the source for shallow grazing angles. So sonar works best with "rough" surfaces
or when it moves around in clever way that avoids being confused by the invisiblity of such "mirrors" from certain viewing angles.
As a general comment, I would like to remind you that bats are an "existence proof" for the power of in-air sonar. These animals are active in 3-dimensional space, they are versatile, often predatory, and they can achieve all with biosonar as a sufficent far sense. As to your specific points:
1) sonar can be used both in active and passive mode. You are correct that range information is not easily obtained in the latter case, but range is not everything and you can learn a lot form listening to what is going on around you. Using an active sense and bringing your own energy sources can also be an advantage, if there are none around.
2) sonar can be used for many other things than building 3d models. You may for instance tell what kind of environment you are in, whether you are likely to collide with an obstacle, etc., all without a 3d map.
3) Ultrasound in air has a limited range due to absorption and spreading losses. However, when operating at a close range (say within 10-15 meters) you can get excellent signal-to-noise ratios (say 40 to 80 dB). If you don't move too fast, such a range should do.
4) The question is what you want to do. Sonar can easily outperform a camera, if there is no light, or if the air is filled with thick smoke, in murmky water, and so on. Other issues are the cost of processing, data storage, and so on. Your fance stereo processing isn't exactly cheap computationally, so far a small consumer electronics product, sonar may offer a much better deal, for instance.
First of all, it is still not fully clear if bats take special actions to avoid "jamming" each other. Ultrasound doesn't carry far (due to absorption) and ultrasonic emissions by bats (and the CIRCE head) are pretty directional, i.e., sound goes mostly into one direction. So the space that is "jammed" by a bat (or a robot) is really small. In the temporal domain, there are a lot of pauses, too. So one sonar system (bat or robot) really influences only a small volume for short time intervals, which should not hamper others all that much.
A few terrestrial animals other than bats use sonar. An example is the Oilbird (Steatornis caripensis) in South America as well as some swiftlets. Owls also use sound to track down their prey, but they do not produce "sonar pings" them themselves. Instead, the listen to sounds that the pre-produces.
You are right, receiving a signal is not enough. You need to receive the right signal and then process it in the right way. The question is where to process. Traditionally, simple receivers were used and sophisticated processing performed afterwards was then supposed to get all the information from their output. That never worked. What we learn from bats is that the processing has to start early on, i.e., what signal ("sonar ping") to use, how to spread its energy in the environment, where to point the ear in the environment and for which sounds from where to be sensitive. Remember that by virtue of the data processing lemma in information theory, you can only throw information away when processing sensor output. However, if you influence the process where information is created, you have control over what information comes into the system.
Producing an accurate 3d landscape representation is probably not the way to go either. If you design a robot, this is rarely what you want. A robot has to navigate around and to attain some goals, so it should extract from its sensors exactly the information that it needs to do that. 3d Map are unnecessary detours in most cases, because you still have to interprete the map then.
Slashdot Mirror
The material of the tinfoil makes not much of a difference, but the shape certainly does. If you come up with a good stealth design, you should be able to fool any monostatic sonar (receiver and emitter close to each other) quite a bit.
The robot is as harmful to the bats as one bat is to the other. Its sonar beam is pretty narrow, pulsed in time, and has a short range only. A flying bat would get out of its reach within seconds at worst.
The situation with the dolphins and the sonar is different, because sound (of the same frequency) carries much further in water than in air. That and the fact that for electromagnetic waves it is the other way round is the reason why the Navy uses sonar and the Army doesn't.
There are certainly situations with so many bats so close to each other that they will likely jam each other, e.g. in a large colony. However, in most of these situations, the bats are "at home", so they don't need and - as anecdotal evidence suggests - probably don't read their sonar.
Since bat sonar is directional on the emission and reception side, as well as - for "resonably" specular targets - on the reflection side, is range-limited by absorption, and - as you point out correctly - there is a time window for reception, I expect that jamming is not much of a problem and bats probably do not require much sophistication in dealing with it.
For these reasons, I am not inclined to throw precious resources at this. Maybe I'll hand out a master thesis project one day to come up with an estimate of how close bats have to be packed for jamming to become a problem. But I expect you would end up with a "sardines in a tin" packing.[thank you for your input & good wishes]
You are correct in pointing out that a single sensor cannot give "good" 3d results; they may be "accurate", but not unambiguous (all points on the surface of a prolate spheroid correspond to the same mouth-to-ear time-of-flight).
The array information from the individual cilia is - in the first place - exclusively a frequency-domain information, because the inner ear splits the incoming signal into its frequency components. The earshape can add a spatial dimension to this, because it can cause the ear to be sensitive in different directions for different frequencies. So it is not a question of "either or", ear shapes and frequency analysis by the cilia work together.
Radar uses electromagnetic waves, in-air sonar uses ultrasound. Ultrasound frequencies are by definition above the audible spectrum, so you don't hear them.
As a general comment, I would like to remind you that bats are an "existence proof" for the power of in-air sonar. These animals are active in 3-dimensional space, they are versatile, often predatory, and they can achieve all with biosonar as a sufficent far sense. As to your specific points:
1) sonar can be used both in active and passive mode. You are correct that range information is not easily obtained in the latter case, but range is not everything and you can learn a lot form listening to what is going on around you. Using an active sense and bringing your own energy sources can also be an advantage, if there are none around.
2) sonar can be used for many other things than building 3d models. You may for instance tell what kind of environment you are in, whether you are likely to collide with an obstacle, etc., all without a 3d map.
3) Ultrasound in air has a limited range due to absorption and spreading losses. However, when operating at a close range (say within 10-15 meters) you can get excellent signal-to-noise ratios (say 40 to 80 dB). If you don't move too fast, such a range should do.
4) The question is what you want to do. Sonar can easily outperform a camera, if there is no light, or if the air is filled with thick smoke, in murmky water, and so on. Other issues are the cost of processing, data storage, and so on. Your fance stereo processing isn't exactly cheap computationally, so far a small consumer electronics product, sonar may offer a much better deal, for instance.
First of all, it is still not fully clear if bats take special actions to avoid "jamming" each other. Ultrasound doesn't carry far (due to absorption) and ultrasonic emissions by bats (and the CIRCE head) are pretty directional, i.e., sound goes mostly into one direction. So the space that is "jammed" by a bat (or a robot) is really small. In the temporal domain, there are a lot of pauses, too. So one sonar system (bat or robot) really influences only a small volume for short time intervals, which should not hamper others all that much.
A few terrestrial animals other than bats use sonar. An example is the Oilbird (Steatornis caripensis) in South America as well as some swiftlets. Owls also use sound to track down their prey, but they do not produce "sonar pings" them themselves. Instead, the listen to sounds that the pre-produces.
You are right, receiving a signal is not enough. You need to receive the right signal and then process it in the right way. The question is where to process. Traditionally, simple receivers were used and sophisticated processing performed afterwards was then supposed to get all the information from their output. That never worked. What we learn from bats is that the processing has to start early on, i.e., what signal ("sonar ping") to use, how to spread its energy in the environment, where to point the ear in the environment and for which sounds from where to be sensitive. Remember that by virtue of the data processing lemma in information theory, you can only throw information away when processing sensor output. However, if you influence the process where information is created, you have control over what information comes into the system.
Producing an accurate 3d landscape representation is probably not the way to go either. If you design a robot, this is rarely what you want. A robot has to navigate around and to attain some goals, so it should extract from its sensors exactly the information that it needs to do that. 3d Map are unnecessary detours in most cases, because you still have to interprete the map then.