# How accurate can near earth asteroids be tracked?

To get more statistical significance for the suspected flyby-anomaly (or to refute it), it would be desirable to track as many hyperbolic earth flybys to within 1mm/s as we can get. Can we track natural objects approaching/leaving earth with this accuracy by long-term observations?

Laser tracking of manmade objects may be easier. Think of a swarm of passive reflectors, spring-emitted way before perigee of an earth gravity-assist, so they pass with different distance and leave in different directions. But I fear that we will never see such a maneuver with a significant plane change, because missions rarely ever leave the ecliptic. With the observations so far, in- and outbound declination seems significant, especially going from high declination back to low would possibly give insight. Will the effect be reversed?

I doubt we will ever see a mission dedicated to this effect, so studying it with natural objects would be useful.

• @uhoh Yes, it sounds unreasonable if you remember that even tracking luna is done with a reflector. I was however not sure, what long-term integration of observations can do about this. – Andreas Nov 5 '16 at 14:52
• One of OSIRIS-REx's major tasks is to carefully characterize the 3D mass distribution and to carefully map the reflectivity and emissivity of Bennu, to improve on the estimations of the non-gravitational effects that will affect it's orbit. For the experiment you are proposing, the natural bodies will be much low mass - because high mass NEOs are (luckily) infrequent! Low mass means the uncertainties in the non-gravitational effects will be magnified, and may overshadow any hope to extract meaningful data. So ambiguous reflector + uncertain force – uhoh Nov 5 '16 at 15:06
• @uhoh Why wouldn't the anomaly be apparent for satellites in Earth's orbit? 1 mm/s is 31 km per year. – LocalFluff Nov 5 '16 at 15:20
• For good accuracy, radar is useful. You can get a very accurate axial velocity measurement with Doppler radar (1 mm/s doesn't seem out of the question). Radial velocity depends on being able to correlate multiple measurements into a single track. – Hobbes Nov 5 '16 at 16:06
• Here lpi.usra.edu/books/AsteroidsIII/pdf/3004.pdf are some results about 4179 Toutatis, a resolution as fine as 125 ns (19 m in range) and 8.3 mHz (0.15 mm/s in radial velocity) was achieved. – Uwe Nov 7 '16 at 15:47

• Note the ratio of quoted position accuracy in the Space.com article (10 m) to distance (324,600 km) is about $3\times 10^{-8}$, right in line with the NASA article. – Chris Nov 21 '16 at 15:23