4 ms·

We’ve had other discussions of these sort of quantum compass ideas, see https://news.ycombinator.com/item?id=40692333 https://news.ycombinator.com/item?id=4069

We’ve had other discussions of these sort of quantum compass ideas, see https://news.ycombinator.com/item?id=40692333 https://news.ycombinator.com/item?id=40692333 for example. I can never quite wrap my head around it or figure out if they are all the same idea, or what, and what exactly the idea is. In particular, the article soften describe something that “could replace GPS.” But then they start taking about compasses, and things that sound a lot like IMUs. Integrating accelerations is well known as a bad way of keeping track of position over longer timescales due to catastrophic cancellation. A more accurate IMU might let you do it for a little longer, but eventually you need an absolute reference point to correct your drift. But then, sometimes there are references to making very careful measurements of the change in the gravitational field. So maybe those could act as landmarks or something? I wish there was an article for people who, like, had a control theory or signal processing class, but it was too early in the morning and not very interesting.

gmrple 2y ago
Gravity map matching is an approach that can provide absolute references, the general problem there is that gravity maps are not good enough.
- bee_rider 2y ago
  They also vary over time or something, right? So even a more precise measurement device, to get better gravity maps, doesn’t solve the problem… maybe? This is all based on half-remembered other comments.
  - ricksunny 2y ago
    So I would like to learn more about gravity map matching. I've seen offhand references to it in context of placing an ICBM on target (can't find ref now) but that's it. I imagine it is, (or was for long enough to make info sparse) a classified piece of know-how.
    - rcxdude 2y ago
      For sufficiently accurate IMUs, like the ones found in ICBMs, you need a gravity map because otherwise the variations in gravity cause your position estimate to drift
      - ricksunny 2y ago
        That's helpful; so presumably for the same reason that gyrocompasses function (self-aligning vis-a-vis the local gravity vector). A part of me was subconsciously entertaining that a subsystem of the IMU / INS was actually being used to provide gravimetric navigation (producing a coarse global position update de novo rather than correct for an error-prone open-loop one from an a priori generated gravity map). a la https://ieeexplore.ieee.org/document/9059124 https://ieeexplore.ieee.org/document/9059124 or https://www.researchgate.net/publication/343278497_Analysis_of_structure_and_algorithm_features_of_new_type_strapdown_gravimetric_navigation_system https://www.researchgate.net/publication/343278497_Analysis_...
- kurthr 2y ago
  There's also the tides, and variations in the tides with 2nd and 3rd order effects on gravity. Absolute references based on variable references is ugly!
- rcxdude 2y ago
  Only when you're on the ground in the first place. Otherwise, you have a circular problem: past a certain point your IMU needs a gravity map to remain accurate, else the drift from variations in gravity is larger than the drift in your sensor. And you need to know where you are and how you're moving to measure gravity.
Arch485 2y ago
My understanding is that by using "the magic of quantum", accelerations can be measured with 0 error (for practical purposes), so it's an incredibly accurate dead reckoning positioning system. In that way, it's "GPS, but better" because you can't be jammed or lose signal. It would also work under water or under ground.
- horstbort 2y ago
  That's the gist of it, yes. The big advantage atom interferometers have over classical IMUs is that they do not need to be recalibrated every so often. They are thus often called "drift free". That still means that you need to compare your position with a reference after a certain time, but the position uncertainty is much lower and only determined by the error accumulating from integration, not due to the instrument itself degrading over time and introducing systematic errors. The reason here is not really "quantum magic" but rather the fact that the reference for the measurement is a laser beam which can be controlled with remarkable precision and accuracy.
actinium226 2y ago
From the article > Motion sensors in wearable devices are usually sensitive to accelerations in the range of [milli-gs]. Bigger, more expensive accelerometers ... on the order of [0.1 mill-gs] > More accurate than either of these mobile accelerometers are cold-atom interferometers. These currently have sensitivities of [nano-gs], with future potential for [femto-gs], Kodigala says.
Animats 2y ago
This used to be mostly a military problem. No longer. GPS is no longer reliable enough for civilian aerial navigation in much of eastern Europe and the Middle East.[1] "Most recently, for the one-month period from July 15 - August 15, 2024, a total of 41,000 flights experienced spoofing. Because modern aircraft have incorporated GPS into a large number of aircraft systems, the impact of a spoofed GPS signal has had severe and cascading effects. These include the FMS, Hybrid IRS, the aircraft clock, GPWS, Weather Radar, CPDLC, ADS-B and ADS-C, as well as numerous other systems." GPS is so low power that spoofing signals now take out a large area of service. [1] https://ops.group/blog/gps-spoofing-final-report/ https://ops.group/blog/gps-spoofing-final-report/
- MichaelZuo 2y ago
  There are many frequent flyers travelling on routes going through those regions, dozens of times each in the last year. And they haven’t reported any noticeable issues caused by GPS spoofing, at least not that I’ve heard of. So I’d be surprised if modern planes are designed such that the loss of GPS for an hour or two would cause issues severe enough to be noticed by even seasoned travellers.
  - aftbit 2y ago
    Not to be a jerk, but you would be wrong. GPS spoofing is a growing problem in commercial aviation. So far it has not led to any fatalities, but it has led to a significant number of system malfunctions which degrade the safety features available to pilots. One example that made the rounds on social media a few months back was a crew flying above 30,000 ft receiving "Terrain - Pull Up!" advisories. They ignored them, because there is no terrain on earth at 30k ft. If that had happened at a lower altitude, they would have likely had to do a terrain escape maneuver, which would definitely be felt by passengers, though it would likely not introduce a lot of additional safety risk. The real risk comes when pilots begin to doubt that their safety systems are working correctly and thus ignore true warnings. Or when a bungled GPS leads pilots to fly into military airspace and get shot down.
    - 2y ago
cryptonector 2y ago
Inertial navigation can be paired with GPS (and friends), terrain recognition, etc. to keep it accurate, and it can be used to detect jamming attacks on GPS. You can even use fairly low accuracy IMUs to detect jamming attacks on GPS, so it's certainly worth using. With enough accuracy the frequency at which an IMU needs to be recalibrated with GPS or other systems can be low enough that a ship need only do it when at port, or a plane need only do it when parked or at an airport gate. I.e., at places where differential GPS can be available and/or where location can be known statically since the location only changes due to continental drift (i.e., very slowly).

gmrple 2y ago

Gravity map matching is an approach that can provide absolute references, the general problem there is that gravity maps are not good enough.

bee_rider 2y ago
They also vary over time or something, right? So even a more precise measurement device, to get better gravity maps, doesn’t solve the problem… maybe? This is all based on half-remembered other comments.
- ricksunny 2y ago
  So I would like to learn more about gravity map matching. I've seen offhand references to it in context of placing an ICBM on target (can't find ref now) but that's it. I imagine it is, (or was for long enough to make info sparse) a classified piece of know-how.
  - rcxdude 2y ago
    For sufficiently accurate IMUs, like the ones found in ICBMs, you need a gravity map because otherwise the variations in gravity cause your position estimate to drift
    - ricksunny 2y ago
      That's helpful; so presumably for the same reason that gyrocompasses function (self-aligning vis-a-vis the local gravity vector). A part of me was subconsciously entertaining that a subsystem of the IMU / INS was actually being used to provide gravimetric navigation (producing a coarse global position update de novo rather than correct for an error-prone open-loop one from an a priori generated gravity map). a la https://ieeexplore.ieee.org/document/9059124 https://ieeexplore.ieee.org/document/9059124 or https://www.researchgate.net/publication/343278497_Analysis_of_structure_and_algorithm_features_of_new_type_strapdown_gravimetric_navigation_system https://www.researchgate.net/publication/343278497_Analysis_...
kurthr 2y ago
There's also the tides, and variations in the tides with 2nd and 3rd order effects on gravity. Absolute references based on variable references is ugly!
rcxdude 2y ago
Only when you're on the ground in the first place. Otherwise, you have a circular problem: past a certain point your IMU needs a gravity map to remain accurate, else the drift from variations in gravity is larger than the drift in your sensor. And you need to know where you are and how you're moving to measure gravity.

Arch485 2y ago

My understanding is that by using "the magic of quantum", accelerations can be measured with 0 error (for practical purposes), so it's an incredibly accurate dead reckoning positioning system. In that way, it's "GPS, but better" because you can't be jammed or lose signal. It would also work under water or under ground.

horstbort 2y ago
That's the gist of it, yes. The big advantage atom interferometers have over classical IMUs is that they do not need to be recalibrated every so often. They are thus often called "drift free". That still means that you need to compare your position with a reference after a certain time, but the position uncertainty is much lower and only determined by the error accumulating from integration, not due to the instrument itself degrading over time and introducing systematic errors. The reason here is not really "quantum magic" but rather the fact that the reference for the measurement is a laser beam which can be controlled with remarkable precision and accuracy.

actinium226 2y ago

From the article > Motion sensors in wearable devices are usually sensitive to accelerations in the range of [milli-gs]. Bigger, more expensive accelerometers ... on the order of [0.1 mill-gs] > More accurate than either of these mobile accelerometers are cold-atom interferometers. These currently have sensitivities of [nano-gs], with future potential for [femto-gs], Kodigala says.

Animats 2y ago

This used to be mostly a military problem. No longer. GPS is no longer reliable enough for civilian aerial navigation in much of eastern Europe and the Middle East.[1] "Most recently, for the one-month period from July 15 - August 15, 2024, a total of 41,000 flights experienced spoofing. Because modern aircraft have incorporated GPS into a large number of aircraft systems, the impact of a spoofed GPS signal has had severe and cascading effects. These include the FMS, Hybrid IRS, the aircraft clock, GPWS, Weather Radar, CPDLC, ADS-B and ADS-C, as well as numerous other systems." GPS is so low power that spoofing signals now take out a large area of service. [1] https://ops.group/blog/gps-spoofing-final-report/ https://ops.group/blog/gps-spoofing-final-report/

MichaelZuo 2y ago
There are many frequent flyers travelling on routes going through those regions, dozens of times each in the last year. And they haven’t reported any noticeable issues caused by GPS spoofing, at least not that I’ve heard of. So I’d be surprised if modern planes are designed such that the loss of GPS for an hour or two would cause issues severe enough to be noticed by even seasoned travellers.
- aftbit 2y ago
  Not to be a jerk, but you would be wrong. GPS spoofing is a growing problem in commercial aviation. So far it has not led to any fatalities, but it has led to a significant number of system malfunctions which degrade the safety features available to pilots. One example that made the rounds on social media a few months back was a crew flying above 30,000 ft receiving "Terrain - Pull Up!" advisories. They ignored them, because there is no terrain on earth at 30k ft. If that had happened at a lower altitude, they would have likely had to do a terrain escape maneuver, which would definitely be felt by passengers, though it would likely not introduce a lot of additional safety risk. The real risk comes when pilots begin to doubt that their safety systems are working correctly and thus ignore true warnings. Or when a bungled GPS leads pilots to fly into military airspace and get shot down.
  - 2y ago

cryptonector 2y ago

Inertial navigation can be paired with GPS (and friends), terrain recognition, etc. to keep it accurate, and it can be used to detect jamming attacks on GPS. You can even use fairly low accuracy IMUs to detect jamming attacks on GPS, so it's certainly worth using. With enough accuracy the frequency at which an IMU needs to be recalibrated with GPS or other systems can be low enough that a ship need only do it when at port, or a plane need only do it when parked or at an airport gate. I.e., at places where differential GPS can be available and/or where location can be known statically since the location only changes due to continental drift (i.e., very slowly).