4 ms·

Because positioning doesn't work like that. It's possible that Rosetta could have carried a detector that could get the time of flight of Philae and so get a r

by joshvm 12y ago

Because positioning doesn't work like that. It's possible that Rosetta could have carried a detector that could get the time of flight of Philae and so get a rough sphere where it could be. However, over the distances that we're talking about, you'd probably need a powerful laser with a very low divergence in order to get any signal there/back. LIDAR will break a power budget easily. If you're thinking why couldn't we use some kind of GPS type positioning, remember that you need four satellites minimum for that (plus ground stations to correct things). There was some discussion about using multiple flybys of Rosetta, but it's not clear whether there'll be enough time to do this before the batteries die. We can locate things on Mars because we have excellent cartography of the planet and the landers were the right way up. We're able to figure out where things are based on the images they send back. Once we know where a lander is the first time, it's easier to figure out where it is afterwards based on this imagery.

13 12y ago
Why are four satellites needed for positioning, wouldn't three do the trick?
- joshvm 12y ago
  The first satellite gives you a sphere (a shell really) The second satellite gives you a circle of intersection between the spheres. The third satellite gives you two points where three spheres intersect. One of these is usually in space so it's discarded. The fourth one lets you correct for timing errors in the reciever. A microsecond uncertainty corresponds to around 300m (i.e. c * 1µs). You could get coarse information with three satellites, for instance if you already know your altitude, but four+ is preferable.
- tonyarkles 12y ago
  In Earth-based GNSS systems, it's for clock synchronization. You've got 4 unknowns (x,y,z,delta t) and 4 signals; if you only had 3 signals, you couldn't solve the system of equations.
jessriedel 12y ago
> Because positioning doesn't work like that. Yes it can. You don't need LIDAR when you the thing you are tracking can have a purpose-built transmitter attached to it. Obviously I'm not talking about GPS, I'm talking about some sort of EM radiation (e.g. the IR on your remote control) being emitted from Philae and being simply seen, as a bright dot in space, by Rosetta. Distance, obviously, you do with timing or with luminosity. Doubtlessly there is a list of many other things I haven't considered, at the top of which are nonintuitive weight, power, and reliability restriction. But your comment doesn't answer my question at all.
- joshvm 12y ago
  Frankly I find your response rather snarky, remember comments aren't just for your benefit. Hopefully this helps answer your question. You asked why it doesn't have a positioning system: the simple answer is the battery (Philae has only 130Wh or so), cost and whether it's necessary to do the science: it wasn't. If you measure distance by timing light, you're using a LIDAR system by definition. A laser isn't explicitly necessary, but because the power received is proportional to R^-4, using an isotropic emitter is wasteful. A pulsed LIDAR system with cm accuracy over 20km is doable, you'd need to time the light to within around 30ns. However we'd struggle to get that kind of accuracy on Earth, let alone orbiting a comet. Luminosity isn't good enough to measure distance (certainly not cm level), it depends on the efficiency of the emitter and the efficiency of the detector, both of which might have changed during flight. It also depends on the transmission medium, in this case it may well be dusty. It could potentially give you a location, but I'm sceptical Rosetta would see it above the background unless it was an extremely bright source. I think putting the transmitter on the lander would be the backward route. A corner cube reflector would probably be the most viable solution. We're assuming that Rosetta can overfly Philae and fire its laser in the right place, but I think that would ok. LOLA on the Lunar Recon. Orbiter flew as low as 30km and only got down to around 10cm resolution and required around 15-30W. My comment regarding GPS was that many people assume that you can get a 3D fix on an object with a single recieve/transmit pair (like in the movies) and it simply doesn't work like that. A directional antenna could work, perhaps.
  - 12y ago

by joshvm 12y ago

13 12y ago

Why are four satellites needed for positioning, wouldn't three do the trick?

joshvm 12y ago
The first satellite gives you a sphere (a shell really) The second satellite gives you a circle of intersection between the spheres. The third satellite gives you two points where three spheres intersect. One of these is usually in space so it's discarded. The fourth one lets you correct for timing errors in the reciever. A microsecond uncertainty corresponds to around 300m (i.e. c * 1µs). You could get coarse information with three satellites, for instance if you already know your altitude, but four+ is preferable.
tonyarkles 12y ago
In Earth-based GNSS systems, it's for clock synchronization. You've got 4 unknowns (x,y,z,delta t) and 4 signals; if you only had 3 signals, you couldn't solve the system of equations.

jessriedel 12y ago

> Because positioning doesn't work like that. Yes it can. You don't need LIDAR when you the thing you are tracking can have a purpose-built transmitter attached to it. Obviously I'm not talking about GPS, I'm talking about some sort of EM radiation (e.g. the IR on your remote control) being emitted from Philae and being simply seen, as a bright dot in space, by Rosetta. Distance, obviously, you do with timing or with luminosity. Doubtlessly there is a list of many other things I haven't considered, at the top of which are nonintuitive weight, power, and reliability restriction. But your comment doesn't answer my question at all.

joshvm 12y ago
Frankly I find your response rather snarky, remember comments aren't just for your benefit. Hopefully this helps answer your question. You asked why it doesn't have a positioning system: the simple answer is the battery (Philae has only 130Wh or so), cost and whether it's necessary to do the science: it wasn't. If you measure distance by timing light, you're using a LIDAR system by definition. A laser isn't explicitly necessary, but because the power received is proportional to R^-4, using an isotropic emitter is wasteful. A pulsed LIDAR system with cm accuracy over 20km is doable, you'd need to time the light to within around 30ns. However we'd struggle to get that kind of accuracy on Earth, let alone orbiting a comet. Luminosity isn't good enough to measure distance (certainly not cm level), it depends on the efficiency of the emitter and the efficiency of the detector, both of which might have changed during flight. It also depends on the transmission medium, in this case it may well be dusty. It could potentially give you a location, but I'm sceptical Rosetta would see it above the background unless it was an extremely bright source. I think putting the transmitter on the lander would be the backward route. A corner cube reflector would probably be the most viable solution. We're assuming that Rosetta can overfly Philae and fire its laser in the right place, but I think that would ok. LOLA on the Lunar Recon. Orbiter flew as low as 30km and only got down to around 10cm resolution and required around 15-30W. My comment regarding GPS was that many people assume that you can get a 3D fix on an object with a single recieve/transmit pair (like in the movies) and it simply doesn't work like that. A directional antenna could work, perhaps.
- 12y ago