Tracking spacecraft from Earth is an increasingly cumbersome issue as we continue to add new vehicles into the mix. The Deep Space Network can track a Voyager at the edge of the Solar System, but using round-trip times and the Doppler shift of the signal is a less than optimal solution for accurate tracking. What we’d like is a method that would allow the spacecraft to calculate its position on its own, taking precise readings from some system of celestial markers.
Pulsars have been in the mix in this thinking for some time. After all, these remnants of stars rotate at high speed and put out radiation beams that blink on and off at regular intervals. They’ve been called ‘celestial lighthouses’ because of this effect, and they’re usefully consistent, producing their pulses in intervals that vary from milliseconds to seconds. The easiest analogy is with the global positioning system, and in this recent article in IEEE Spectrum (thanks to Frank Smith for the pointer), that’s exactly how their use is described:
A craft heading into space would carry a detector that, similarly to a GPS receiver, would accept X-rays from multiple pulsars and use them to resolve its location. These detectors—called XNAV receivers—would sense X-ray photons in the pulsars’ sweeping light. For each of four or more pulsars, the receiver would collect multiple X-ray photons and build a “light curve.” The peak in each light curve would be tagged with a precise time. The timing of these peaks with respect to one another would change as you traveled through the solar system, drawing nearer to the source of some and farther from others. From this pattern of peaks, the spacecraft could calculate its position.
Image: The magnetic poles on a neutron star act like fountains, an escape valve for charged particles that get trapped in the star’s enormously strong magnetic field. As a neutron star spins, its polar fountains turn with it, like an interstellar lighthouse beam. From Earth, we see the beam as it quickly sweeps past us — there, gone, there, gone — many times a second. That looks like a pulse from here. Hence the name, “pulsar.” Credit: National Radio Astronomy Observatory.
Pushing the pulsar navigation idea forward is a table-top device known at NASA Goddard as the Goddard X-ray Navigation Laboratory Testbed (GXNLT), which has been developed to test out a navigation experiment that will be flown on the International Space Station as early as 2017. The ‘pulsar-on-a-table’ can mimic pulsar spin rates and model pulsar locations in the sky, simulating the environment that the upcoming ISS experiment will encounter. The X-ray photons it produces are detected and their arrival times processed by algorithms to extract orbital position.
These technologies are being validated for the Neutron-star Interior Composition Explorer/Station Explorer for X-ray Timing and Navigation Technology mission (NICER/SEXTANT), which will, in addition to demonstrating navigation by pulsars, study the interior composition of neutron stars (for more on NICER/SEXTANT, see this GSFC news release). The assumption is that like the 26-satellite GPS system, pulsar navigation will be able to use onboard software to calculate a position, but without the obvious, Earth-centric limitations of GPS signals.
Luke Winternitz, a co-developer of the Goddard testbed, thinks navigating with X-rays can open up deep space for autonomous navigation, and the test equipment is a vital step in shaking out the system:
“X-ray navigation has the potential to become an enabling technology for very deep space exploration and an important augmentation to NASA’s Deep Space Network. We had to have a way to test the technology. We have GPS constellation simulators that make our GPS receivers think they are in orbit; we needed something analogous for an XNAV receiver.”
So far the ground tests using the pulsar-on-a-table show that the system will be accurate within one kilometer in low-Earth orbit, but the goal is to reach accuracies in the hundreds of meters even in deep space. By the time we do get NICER/SEXTANT into place on the International Space Station, we should have a high degree of confidence that the system will work. Ultimately, basing navigation within the Solar System and beyond on Earth-based equipment will give way to autonomous navigation, a necessity for accuracy as well as practicality as we push outward.
Don’t forget that in addition to the spacial information provided by relative lightcurve phases, that Doppler-shift of each of the base pulsar frequencies give you relative velocity in several different directions, so that an overall velocity magnitude and direction can be determined.
I learned this the hard way, when in college I set out to capture the _optical_ lightcurve of the Crab Nebula pulsar. I failed to get an additive lightcurve because I neglected the Doppler shift caused by Earth’s orbit around the Sun, so my locally-generated synch signal pulses were not locked to the Doppler-shifted pulsar frequency. After correcting for this and locking into the proper frequency, it was easy to build up a lightcurve from photon arrival times added over many pulse periods.
My only concern with neutron (pulsar) navigation is the beam diameter, it will fade as we get further away from the earth but that should be made up by the fact that new ones will come into view. Other than that it is a stellar GPS for free.
I really don’t get it. There is some implicit fact missing in the explicit statement of the way this navigation works.
I know how regular the train of a pulsar’s pulses is, but is each pulse within them that regular? To get 10m accuracy from a pulsar rotating 30 times per second, you would have to know where you are within a millionth of a single pulse. Worse still, even that relies on perfect regularity, and they undergo episodes of sudden contraction (sorry, I don’t know what you astronomers call those). After each of these you would have to recalculate the curve for that particular pulsar, and it you were off by one part in a trillion, then your calculated position would drift out by a 3o metres each day.
That talk of knowing your ORBITAL position that closely must be a joke, since such accurate recalibrations seem impossible in an accelerating reference frame. At least I can see why you would need so many pulsars to even attempt this .
@Mike Lockmoore, Et al.
while this sounds all good, I can’t help wondering how stellar aberration will impact their keeping their course when relativistic speeds are used.
This system should also be known as GPS – Galactic Positioning System.
I wasn’t able to find much on this which I’m curious about, what are the limiting factors on the navigation accuracy of hundreds of meters? Size and power of the processing equipment?
I imagine for sub-hectometer accuracy will depend on radar/laser ranging and dead reckoning?
13 June 2013
** Contacts are listed below. **
Text & Image:
http://www.astron.nl/about-astron/press-public/news/radio-astronomers-use-precision-pulsar-positions-break-distance-recor
RADIO ASTRONOMERS USE PRECISION PULSAR POSITIONS TO BREAK DISTANCE RECORD
An international team of scientists led by astronomer Adam Deller (ASTRON) have used the Very Long Baseline Array (VLBA) to set a new distance accuracy record, pegging a pulsar called PSR J2222-0137 at 871.4 light-years from Earth. They did this by observing the object over a two-year period to detect its parallax, the slight shift in apparent position against background objects when viewed from opposite ends of Earth’s orbit around the Sun.
With an uncertainty less than four light-years, this distance measurement is 30 percent more accurate than that of the previous-best pulsar distance. The VLBA observations were even able to discern the orbital motion of the pulsar around its as-yet undetected companion object, despite this motion being no larger than a small coin observed at a tenth of the distance to the Moon.
The results of the research have been published in The Astrophysical Journal:
http://dx.doi.org/10.1088/0004-637X/770/2/145
By showing that PSR J2222-0137 is 15% closer than previous estimates, this impressive achievement can advance our understanding of the system. With the distance to the pulsar pinned down, proposed highly sensitive visible-light observations should determine the nature of the undetected companion. If no source can be found, the companion must be a neutron star, while a white-dwarf companion will show up as a faint optical source.
The accuracy of the new measurement promises to help in the quest to detect the elusive gravitational waves predicted by general relativity. By monitoring an array of pulsars across the Milky Way galaxy, scientists hope to measure the distortions of space-time caused by the passage of gravitational waves. Knowing the distances to these pulsars extremely precisely can improve the sensitivity of the technique to detect individual sources of gravitational waves. The VLBA is operated by the National Radio Astronomy Observatory (NRAO).
PIO Contact:
Femke Boekhorst
PR & Communication, ASTRON
+31 521 595 204
boekhorst@astron.nl
Science Contact:
Adam Deller
Astronomer, ASTRON
+31 521 595 100
deller@astron.nl
Caption to the image online (www.astron.nl): Illustrating trigonometric parallax: the VLBA can measure the slight apparent shift in the position of an object as seen from opposite sides of the Earth’s orbit. The size of this position shift is dependent on the distance of the object from Earth. Credit: Bill Saxton, NRAO/AUI/NSF.
You do not need to rely on individual pulses. Pulsar timing accuracy is as good or better as that of an atomic clock. Collect enough pulses, and you can time them collectively to sub-microsecond precision. That brings you into the 100 m range: Move 300 m towards the pulsar, and now all pulses will arrive 1 microsecond earlier, compared to your on-board clock. Move away 300 m, and they will arive a microsecond later. Measuring microseconds is child’s play, really. So is keeping long-term accurate time on-board. The luck factor is that pulsars are known to have the same long-term accuracy as our best clocks, or so I have read ….
I am not sure, but I think the detector does not need to be directional. If it did, it would be very expensive and not really feasible. You just take all X-rays, and identify the pulsar signals by their regularity and distinct frequencies. You know where they are, so you do not actually have to care about the direction from which the signals arrive. Ingenious, really, if you think about it.
You do not even have to have an on-board clock, because you can just compare the different pulsars against each other. The device would instead itself serve as an extremely accurate clock, saving you some hardware elsewhere.