Murison and Munteanu (SEAP student) finished their joint work on the two-body distance problem, as the latter is going off to college. Munteanu finished working on implementing Sitarski's method (a method frequently used to numerically solve for minimum distances between two orbits). Tests show that our Newton-Raphson algorithm is at least five times faster than Sitarski's. Murison continued cleaning up loose ends for an AJ paper.

Murison started using a generalized Duffing equation (a classic differential equation for a driven oscillator with a cubic nonlinearity) for use as a simpler vehicle than the Lagrange Planetary equations (LPEs) yet useful for investigating Efroimsky's gauge theory in dynamics. Specifically, Murison is trying to analytically optimize the specific choice of gauge function for use in numerical integrations. Lessons learned will be applied to the more-complex LPE case.

Murison finished a nearly-final round of updates to his computer program that controls the USNO dFTS adaptive optics (AO) and focal plane imaging hardware. Additions and changes include:

• Memory optimization The transfer of image data from the frame-grabber card through the Sandwich program to the computer monitor screen has been optimized. The frame rate on Murison's desktop computer jumped from 11 fps to 18 fps.
• Subframing Murison implemented subframing throughout the image data path. The full image is updated every N frames, where N is selectable by the user (see the Sandwich homepage). In between, only small subframes are grabbed from the PIXCI video card. The subframes are centered on the current centroid position and are the same size as the inverse-image centroiding box. (The centroiding box only appears when the program is locked onto a star.) The frame rate on Murison's computer jumped to 40 fps.
• Impact on slow-poke AO computer After inserting a delay into the PIXCI card memory read loop, Murison was able to slow his computer down to 2 fps, which is an approximation to the lightning speed of the (decrepit) AO computer in Boston. Upon turning on subframing, that jumps to 8 fps. Thus, Murison thinks we should see a similar improvement on the Boston AO machine. Murison suspects 8-12 fps is going to be the limit on that machine.
• Improvements to the real-time centroid fluctuations plot It runs smoothly now, updating the plot once per second. (Data points are plotted at the full frame rate behind the scenes between plotting updates.)
• Memory leak The memory leak has been fixed.
• Various interface tweaks
• Help The Procedure and Toolbar sections of the help file have been updated to reflect the interface changes.

Murison reviewed a paper for Icarus.