Update on asteroid axis computer program
I modified the program, to fit with the ten spherical harmonics:
n=0,m=0
n=2,m=0
n=2,m=1 (sin & cos)
n=3,m=1 ( " )
n=2,m=2 (sin & cos)
n=3,m=2 ( " )
This amounts to replacing m=3 in the previous program, with m=1, but avoiding the forbidden n=1 harmonics.
The previous program found, basically, the round solutions; that is, those with the smallest sum of squares, of coefficients divided by the coefficient of the constant term. I made the modified program find, basically, the principal axis rotation solutions; that is, the smallest positive integral over the Gauss sphere, of "K inverse" times sin^2(latitude). This amounts to finding the solutions with the sharpest poles; i.e., the solutions most resembling a prolate spheroid rotating about its (stable) axis of least moment of inertia.
This modified program gave the same answer: the rotation axis of 129 Antigone is within a few degrees of the ecliptic pole. Again, there is prograde/retrograde ambiguity (i.e., N or S ecliptic poles fit about equally well). This answer is different from those in the literature, which cluster at two other axes. So, from this test with Antigone, I can't yet say that I have a program that will find the pole reliably from only two lightcurves.
