Detailed Description - Descent Imaging The landing was accomplished with a series of 5 orbit correction maneuvers. The first maneuver, EMM1 began the decent from 35km circular orbit. The four remaining maneuvers, EMM2-5, thrusted in a direction that attempted to brake the fall of the spacecraft during the descent. The landing site was selected to allow good imaging of lit territory all the way down, while satisfying several operational constraints. These included keeping the high gain antenna locked onto the Earth for continuous high-rate playback, and keeping solar panel illumination within limits. This eliminated the possibility of a south polar landing. Landing site was selected to be -37lat 278lon. The majority of time during this period was spent either performing maneuvers, or slewing to the new maneuver positions. Mission design and navigation folks were able to design a set of maneuvers that allowed the camera boresight to be pointing down at the lit surface throughout much of the landing sequence. 25.2 Sequence Design ____________________________ MONOCHROME Descent Sequence | ---------------------------- Opnavs: ------ Following the EMM1 maneuver two 2x3 zig-zag mosaics were acquired and immediately played back. OPN_EMM1_DKD 32/1601 Final Descent Images: -------------------- See /eros/01036/descent_imagelist.gif for a full account of imaging and maneuver timing. ********************* The camera boresight was off the limb for the EMM2 maneuver position. The slew to the EMM3 eventually brought the boresight onto lit territory. From that point on we acquired images all the way down until contact; we imaged during all remaining burns and the s/c maneuvers to reposition to each new burn position. To reduce smear during these repositions, we built special scan patterns that slewed at a constant rate from the burn position to burn position; this was in lieu of the normal fast reposition which gets there faster, but would have introduced smear and would have required a long settling time. A special kind of playback routine was required to buffer the images in real-time and immediately send them to the ground. Normal process was to record images during a designated observation period then playback everything during designated playback period (no data acquisition during playbacks usually). Using the new scheme, the fastest we could play back a pair of images was a little less than 65 seconds. Therefore the final imaging sequence contained pairs of images spaced 65 seconds apart. Spacing between the two images in each pair was set to be 20 seconds. The reason for this was to maintain frame-to-frame overlap between at least the members of each pair during the faster slews between burn positions. If we had set the time delta between the two frames in each pair to be something like 32 sec, there would have been no overlap at all between images taken during some of these burn transition slews. This worked out well because we at least now have little two frame mosaics from those periods.