Continuing Analysis of Spacecraft Jitter in Lroc-nac

Introduction: The Lunar Reconnaissance Orbiter (LRO) has been flying in its nominal, or mapping, orbit of ~50 km above the surface of the Moon since September 15, 2009 [1]. The Lunar Reconnaissance Orbiter Camera Narrow Angle Camera (LROC-NAC) takes high resolution images using two identical line scan cameras (NACL and NACR) with pointing offset from each other by ~2.79o cross-track and ~0.082o along-track. This results in a measured average alongtrack offset of ~67-127 lines, depending on summing mode and scanning rate, and an overlap of the L-R footprints of 68 or 136 pixels in summed or nonsummed mode, respectively. The along-track offset was planned to provide a measure of pointing jitter at a higher frequency than is provided by the spacecraft Attitude Control System (ACS). The high spatial resolution (10 μrad IFOV, or up to 0.5 m pixel scale) [2] makes NAC images susceptible to spacecraft jitter that results in geometric distortions in line scan images. This was shown to occur and to have negative effects on Digital Terrain Model (DTM) quality during the Commissioning Phase of LRO [3,4]. Here we present the ongoing study of mapping orbit jitter, its effects on DTMs, and potential mitigation strategies. LRO is currently operating in the Science Phase of the mission and plans are to return to an orbit similar to that of the Commissioning Phase (fixed 27x216 km elliptical orbit) in the latter part of 2011. Measuring Jitter in LROC-NAC Images: The method for measuring jitter from non map-projected images remains much the same as in [3]. The overlapping portion of each NACL and NACR images is cropped out. The Integrated Software for Imagers and Spectrometers Version 3 (ISIS3) application coreg [5] is used to measure pixel offsets in the line (alongtrack) and sample (cross-track) directions in the overlapping area. Improved radiometric calibration for LROC-NAC incorporated into ISIS results in measurements of high confidence at the sub-pixel level. Pixel offsets are measured in images without any a priori pointing information (fig. 1). Mapping Phase vs. Commissioning Phase Jitter: Spacecraft jitter was lessened when the spacecraft went from the Commissioning orbit to the 50-km circular Primary Phase orbit. This reduction was demonstrable by the improved quality of DTMs using mapping orbit stereo images. The suspected cause of the improvement is the range of solar array disturbance frequencies during the Commissioning Phase included dominant solar array resonance modes that affect NAC stereo images, whereas in the mapping orbit the dominant resonance modes were outside of the range of disturbance frequencies. The solar arrays are stowed in the beta=90o position during beta>53o, which improves stability during optimal NAC stereo imaging conditions. However, jitter does still occur in the mapping orbit, causing noise and artifacts in DTMs (fig. 2), although for the most part these effects are typically 10x less than those seen in the Commissioning Phase orbit. Jitter effects in DTMs: In almost all cases, spacecraft jitter causes no obvious distortions in LROCNAC images. However, even imperceptible geometric distortions degrade the results of stereo matching software used in DTM production. These effects have been observed in some DTMs produced from NAC stereo images by all groups making DTMs, regardless of the method or software [4,6]. A significant number of stereo images acquired in the Commissioning Phase had a level of jitter that caused a distinctive ripple pattern in the DTM [3,4] (fig. 2, left). This pattern is parallel to the cross-track direction, indicating that the distortions responsible are predominantly in the sample, or spacecraft roll, direction [7]. Distortions in the line direction, or spacecraft pitch, cause y-parallax during stereo