M. Malin, Malin Space Science Systems, Inc.
Participation as a Facility Team Member of the Multispectral Imaging System (MIS) Facility Instrument Team on the Near Earth Asteroid Rendezvous (NEAR) mission is proposed. The investigation proposes to acquire and analyze images of the asteroid 433 Eros (an S-type asteroid), using the MIS Facility Instrument, in order to address problems associated with the geological and geophysical evolution of Eros, and related to questions concerning the nature and evolution of other small bodies in the solar system. These objectives will be met through observations of surface features during the approximately one year duration of the NEAR mission.
Research goals of this proposal include studies of the shape (large-scale relief and morphology) of Eros, of the general geology and geomorphology of the asteroid, and of small-scale morphology as determined by visual inspection of images and extraction of relief information using computer techniques. Degradation of relief and aggradation of impact ejecta and the nature of surface and near-surface transport of debris will be investigated, in an attempt to better understand the development of regolith on small bodies and its influence on the preserved cratering record. Structural features (grooves, ridges, etc.) will also be examined, and interpreted in concert with other Project Science Group members in terms of the internal configuration of the asteroid.
The proposed participation as a Team Member of the MIS Team (MIST) would include involvement in instrument design review, and in the development and implementation of instrument calibration requirements, in keeping with the investigator's present and past experience as Principal Investigator on the Mars Observer/Mars Global Surveyor Cameras and as a Team Member on the Comet Rendezvous/Asteroid Flyby (CRAF) Imaging Science Sub-System Team. Of specific focus during the instrument development phase of the project will be effective geometric, photometric, and spatial response calibration. Participation in mission design and planning is also anticipated, with the objective of defining operational procedures and techniques consistent with the low-cost, high-science return objectives of the operations phase of the mission. For the most part, existing hardware and software will be used and/or modified to address the operational and scientific objectives of this investigation; work will be conducted at both the Applied Physics Laboratory's observational control center and at the investigator's home institution.
During mission operations, the investigation will focus observations on preliminary body shape and the rotation axis and rate (Approach and 1000 km Orbit Phases), refined body properties and local geology (early Nominal Orbit Phase), and detailed topography and geomorphology (Nominal Orbit Phase). Using existing software, first-order processing of NEAR MIS images will be quite rapid--without relying of any special computer capabilities, the entire data set can be reprocessed from raw data in about three days. Other programs (existing and to be modified from extant software) will permit rapid development of shape and relief models. Archiving will occur as agreed upon by the PSG and Project.
The scope of the 6 year effort includes approximately 12.5 months of senior research personnel, about 20 months of research support personnel, and 18.4 months of administrative support. A phased approach will be implemented, with the investigator spending about 12 weeks on the project during the first year (for participation in instrument calibration and testing), about 6 weeks the second year (to participate in spacecraft/payload testing), and only about 2 weeks each year during the cruise to Eros. The investigator will devote approximately 40% of his time to the project during mission operations. Support personnel will devote about 1.5 person years to the project during the 18 month operations and data archiving period.
This investigation has two general scientific objectives: to determine the size, shape, and spin state of the Near-Earth asteroid 433 Eros, and to characterize the geology and morphology of the surface, with particular emphasis on the development and transport of fine debris. To accomplish these objectives, a third, technical objective of the investigation is to insure that the geometric, photometric, and spatial response performance of the primary instrument of interest, the Multispectral Imaging System (MIS), is well understood and documented.
The specific investigations to be conducted during the mission operations phase of the mission include the following:
1. Locate the spin axis position, orientation, and determine the spin rate;
2. Establish the shape of Eros using images systematically acquired during the Approach and 1000 km Orbit phases of the mission;
3. Enhance the shape model using images systematically acquired from the Nominal Rendezvous Orbit;
4. Track features between spacecraft maneuvers (optical navigation) as a secondary means of orbit determination to probe the mass and mass distribution of the asteroid;
5. Map geologic and geomorphic features to provide context to compositional mapping; and
6. Study slopes and surface morphology for clues to debris generation (by cratering and other possible processes) and its movement across the surface.
These objectives are directly related to three of the four the primary mission science objectives as noted on Pg. 3 of the Announcement of Opportunity:
Measure surface elemental and mineralogical composition with sufficient accuracy to enable comparisons with major meteorite types;
Characterize the morphology of the asteroid surface; and
Determine regolith properties and textures of the asteroid surface material
Measurement of the gross structure of the asteroid permits the bulk properties of Eros to be placed within the context of other solar system bodies (e.g., gross chemical composition, general relationship to meteorites), while measurement of the surface composition will examine its diversity and relationship to specific meteorite groups. Bulk properties may be seen to vary from position to position along a low orbit, providing clues to interior heterogeneity within the asteroid. These may be correlated with magnetic field phenomena. The shape can also provide clues to the asteroid's origin--the collisional history and, in some cases, the structure of the parent body, may be preserved in the present shape.
Observations of small-scale morphology and relief can be used to characterize the impact flux, which can be used for both relative,and, with some codicils, absolute age determinations. They can also be used to constrain models of surface processes in extremely low gravity environments, and provide insight into cratering mechanics and the relationships between target body, impacting body, and energy and momentum scaling.
The concept for this investigation is relatively simple: images of a small body (433 Eros) can be used to derive bulk shape properties, and combined with mass determinations derived from spacecraft tracking, then used to calculate the gross density of the asteroid. Higher resolution topography and tracking measurements, made closer to the asteroid, permit more detailed comparisons from place to place on the asteroid.
Geological studies of planetary surfaces proceed by the evaluation of classical photointerpretative criteria: planimetric configuration, topographic relief, albedo, color, texture, pattern, and association. Judicious application of prior experience (e.g., studies of Phobos, Deimos, Gaspra, and Ida) and, with caution, terrestrial analogies (e.g., formation of grooves by piping), permits the interpretation to be extended to process, material, and time relationships. Geomorphic maps illustrate the configuration of a surface and illuminate the specific processes that have acted on the surface. Geologic maps synthesize geomorphic, compositional, and stratigraphic information into an interpretative framework.
In the specific study of morphology as related to the development and evolution of the surficial debris layer, images of craters, lineaments, grooves, ridges, and other landforms are acquired, and subsequently examined, at a resolution that permits the scale of their development and potential variations to be characterized. Many of these features show pronounced differences if expressed in dense, hard rock or granular material. The search for these differences, and the scale at which they occur, is paramount to studies of asteroidal regoliths. For example, evidence suggests that these debris layers are thicker than would be expected from ballistic redistribution of impact crater ejecta, leading to the speculation that the near-surface asteroidal material is incompetent. Impact craters also provide information on timescales, although uncertainties in impacting body size/frequency distribution, impact velocity, and impactor composition often reduce this utility.
An important aspect of this investigation is the acquisition of calibration information. Both basic information (e.g., effective focal length, field distortion, etc.) and derived information (e.g., signal-to-noise ratio, absolute photometric response, etc.) are needed if the reduced spacecraft data are to have the desired accuracy and precision. Some of these measurements are relatively easy to perform; others require substantial effort (e.g., targets in sub-system thermal/vacuum testing). With the collection of data of the appropriate type, quality, and quantity, analysis and derivation of the desired factors is generally simple.
Collaborative studies will also be important to this investigation. It is anticipated that, owing to the small size of the total Facility Science Team, each Team Member will have primary responsibility for some area, and will collaborate with other team members in other areas. This investigator plans to work closely with those studying the photometric and multispectral attributes of the surface using the MIS, with investigators studying mineralogical composition using the Near-Infrared Spectrometer (NIS), and with the radio science and LIDAR investigators studying the bulk properties of the asteroid.
A variety of techniques exist for determining the shape of an irregular body. These include several types of photogrammetry, photoclinometry, limb reconstructions, and fused techniques (those that incorporate one or more of the other approaches). In addition, at least some of these techniques can be employed in either fully manual or fully automated modes, and in modes that are partially automated or partially manual. In this section, Viking observations of Phobos are used to illustrate some of these techniques.
Limb curves (and their close relatives, terminator lines) provide significant, though not necessarily complete, information about the shape of a body. With sufficient axial sampling, most of the features of the body can be reconstructed (the principal limitation being that the interiors of concave features such as craters never appear on the limb). Limb curve reconstruction may suffer from operational limitations if it is not possible to obtain a sufficient number of equally spaced, high resolution views around known, multiple axes, to eliminate all possible ambiguity (i.e., in cases where only one axis of rotation can be used, a small hill located between two larger hills might never appear on the limb).
Figure 1: Manual Definition of Limb and Terminator Shape
Figure 1 shows a technique for manual definition of limb and terminator positions in multiple images. The technique permits fitting an initial figure (generally a triaxial ellipsoid) to the object as observed in several images, and then deforming that figure by incorporation of features of known geometry (such as craters with circular planform and parabolic cross-section) or by manually moving grid points to specific locations. The particular software shown in Figure 1 can be used without prior information on pointing and other relevant information or, more effectively, used with the Navigation Ancillary Information Facility (NAIF) toolkit and SPICE (Spacecraft-Planet-Instrument-"Camera"-Event) navigation/attitude kernels for precise manipulation. In this latter case, limb/terminator fitting and other forms of photogrammetry become quite similar. Using this software, a dozen images can be fit in a few hours.
To produce a surface from the scattered points, an interpolation procedure must be used. A variety of spline-based and minimum-energy surface fitting routines have been implemented in existing software. As might be expected, some routines work well under some circumstances, while performing poorly in others. Also as might be expected, the quality of the surface is generally dependent on the number of points, as is the time needed for processing.
Figure 2 shows the "front end" of a set of programs used in stereogrammetry. Left and right images (in side-by-side format or portrayed as anaglyphs) are examined stereoscopically, and a movable cursor used to establish points "on" the virtual surface. Points are manually selected individually. This tool can be used to quickly generate topographic profiles (as shown in Figure 2), but its primary use is to edit points generated automatically by a combined edge/area correlation program.
Figure 2: Point Registration for Stereogrammetric Analysis
The point positions, derived either manually or automatically, are usually expressed as heights above a nominal ground plane, and a surface is interpolated by the methods mentioned above to produce a height grid or digital terrain model.
Photoclinometry, also known as shape-from-shading, attempts to invert the orientation of the surface at each pixel by using a shading model and knowledge of the illumination conditions. Traditional applications of this technique in planetary science have used line-based, integrative methods, which are highly sensitive to errors caused by mismatches between the shading model and the actual surface, imperfections in the imaging system, and albedo variations. At least some of these limitations can be overcome by the use of area-based techniques developed over the past fifteen years by the computer vision community. These techniques attempt to distribute error across the image by globally minimizing criteria like departure from integrability. Although the line-based methods usually require manual input, the area-based techniques are typically fully-automated (the implementation used to create the relief shown in Figure 3 was fully automated). Once the orientation of each surface patch is known, a height map can be built up by simple integration or more sophisticated iterative methods.
Figure 3: Visualization of Photoclinometrically-derived Relief
Figure 3 shows the results of applying a photoclinometry program to a portion of the Viking image of Phobos. It does not show the original data produced by the program, in which the shape of Phobos dominated the relief (as can be seen in the profile in Figure 2). In the upper left corner is a high-pass filtered version of the photoclinometry, where filtering was used as a surrogate for a shape model. The original image is shown at the same scale in the bottom center, and an oblique view of the image superimposed on the relief (with a vertical exaggeration of 4X) is seen in the upper right. Such illustrations are useful in visualizing spatial and topographic relationships, but quantitative values are more important in estimating volumes, relief, and potential energy.
The methods described above occupy different positions within a two-dimensional design space where one axis is the volume of information produced and the other is the accuracy of that information. Along the volume axis, photoclinometry produces a dense distribution of surface information, potentially determining the location of every imaged point on the body, while point photogrammetry only locates as many points as can be found to correspond on multiple images. One limb curve produces a distribution dense only in the plane perpendicular to the viewing direction (and even that is compromised by occultation effects); multiple limb curves are sparse about the axes of rotation, and interpolation is still required to construct a complete surface.
On the other axis, point photogrammetry is the most accurate and well-characterized method. Photoclinometry is easily confused by albedo characteristics, and is subject to a variety of pathological cases that can result in poor surface reconstruction, especially on a global scale. Typically, the error propagation characteristics of photoclinometry algorithms lead to good local results -- small features are well-reconstructed -- but error buildup leads to large global errors. Operational limitations in image coverage may cause limb-based methods to miss features or not recognize discontinuities.
The fact that these techniques have different strengths and weaknesses suggests that the best approach is to combine the results from all of them -- an approach called "sensor fusion" by the computer vision community. A coarse surface approximation will be derived from control points supplemented with a finer model supplied by manually-edited automated limb- and terminator-matching. A technique much like surface rendering of Computer-Aided Tomography (CAT) scans will be used to synthesize many limb models into a single model. This model will be compared with photometric models of comparable resolution, and the process iterated. High spatial resolution photometrically derived relief will be extracted after the shape models converge, with local constraints provided by high resolution stereogrammetry.
The following basic measurements should be made at the assembly level [Note: items marked with asterisks (*) should be checked for dependences on spectral filter)]:
All photometric measurements should be performed for all eight filters and at a variety of gain/offset/exposure values.
Figure 4: Illustration of Effect of Asymmetric Pixel Size
No special ground operations support requirements have been identified at this time. After review of relevant Project documents and discussions with MOS and GDS personnel, some potential requirements may be identified. These will be discussed with the MIS Team Leader and, if appropriate, brought to the Project for consideration.
Mission and operations planning software and communications will be critical to the success of the NEAR mission. Both manual and automated sequence generation should be supported. Interactive as well as batch processing may be needed. These types of planning tools are available to the investigator, and will be modified as appropriate to support his investigation. He expects to work closely with the MOS and GDS developers in determining what other software exists or will exist at the time of this mission, and what interaction would be most beneficial to the Project.
Upon receipt of raw instrument data, initial "quick-look" processing will be performed. This processing will include: random noise removal (despiking), first-order photometric calibration (pre-launch filter factors, exposure compensation, flat field, blemish removal), camera distortion removal (spatial resampling convolved with MTF restoration), and 12-to-8 bit rescaling (for display purposes). Cosmetic processing (contrast enhancement) may also be applied. Second-order processing might include application of in-flight radiometric calibration data during the photometric calibration step (after random noise removal) and additional spatial filtering and contrast enhancement. First-order processing takes approximately 4 seconds per frame on a single-processor SPARCStation 10 using existing software (80 frames can be processed in about 5 minutes, for a processing throughput of about 400 kbps); thus, first order processing can be accomplished and delivered on a daily basis.
Data analysis techniques (as described above) to be applied to the images includes procedures to extract the rotation axis and rate, measure the three-dimensional figure, and produce high-resolution relief information. Once shape information is available, processing will involve photometric function removal and photoclinometric production of topography. Processing associated with extraction of high resolution topography is considerably more time consuming--combined stereogrammetric and photoclinometric processing of a relief model of a 412 X 537 pixel area takes about 3 hours on a single-processor SPARCStation 10. Mosaics and other reduced forms of data will be assembled for analysis, and measurements made on both volatile and hardcopy versions (depending on the scale and area covered). Much of the morphologic and geologic analysis is subjective, using photointerpretative skills.
The investigator proposes to participate with the MIS Team Leader in maintaining the "best version" data base. The mission total data volume, calculated on the basis of information provided on Pg. 29-31 of the Announcement of Opportunity (~152 days at 2000 bps and ~213 at 6000 bps), is roughly 140 Gb. The MIS will probably account for about 70% of these data (96 Gb) Given the relatively fast processing applied to these data (400 kbps), the entire data set could be reprocessed from raw data in less than 3 days on one single-processor SPARCstation 10. Thus, "maintenance" may entail replacing all previous versions with newer versions at specific times, or piece-wise "besting" of the data.
The proposer will work closely with the Team Leader and other members of the PSG to define useful data formats for the storage and exchange of data between investigations, and will make both original and modified data files available to other investigators on mutually agreed upon bases.
The volume of data to be returned by NEAR will fill approximately 230 CD-ROMs. About 160 of these will be needed to contain the raw MIS data. Additional ancillary data will probably fill an additional 10-20 CDs. Production of two to three disks a week is a relatively simple task, and one that can be highly automated. The investigator will work closely with the PSG and the NEAR Project, in particular with engineers with the GDS, in defining this process.
Principal Investigator: Michael C. Malin
Malin, M. C., 1992, Mass movements on Venus: Preliminary results from Magellan Cycle I observations, J. Geophys. Res. 97 (E10), 16337-16352.
Malin, M. C., Danielson, G. E., Ingersoll, A. P., Masursky, H., Veverka, J., Ravine, M. A., and Soulanille, T. A., 1992, The Mars Observer Camera, J. Geophys. Res. 97(E5) 7699-7718.
Malin, M. C., Danielson, G. E., Ravine, M. A., and Soulanille, T. A., 1991, Design and Development of the Mars Observer Camera, Int. J. Imaging Sys. Tech. 3, 76-91.
Phillips, R. J., Grimm, R. E., and Malin, M. C., 1991, Hot-spot evolution and the global tectonics of Venus, Science 252, 651-658.
McEwen, A. S. and Malin, M. C., 1989, Dynamics of sediment gravity flows: Lahars, avalanche, pyroclastic flows, and blast surge of Mount St. Helens, J. Volcan. Geotherm. Res. 37, 205-231.
Kelley, A. D., Malin, M. C., and Nielson, G. M., 1988, Terrain simulation using a model of stream erosion: Computer Graphics 22(4), 263-268.
Malin, M., 1986, Rates of geomorphic modification in ice-free areas, southern Victoria Land, Antarctica: Antarctic Journal of the United States 20(5), p. 18-21.
Malin, M. C., 1986, Density of martian north polar layered deposits: Implications for composition: Geophys. Res. Lett. 13 (5), p. 444-447.