SEARCH

SEARCH BY CITATION

Keywords:

  • RAT;
  • grinding;
  • IDD;
  • payload;
  • rocks;
  • instrument

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Performance Requirements
  5. 3. Operational Considerations
  6. 4. Electromechanical Description of the Rock Abrasion Tool
  7. 5. Data Products
  8. 6. RAT as a Science Instrument
  9. 7. Conclusions
  10. References
  11. Supporting Information

[1] The Rock Abrasion Tool (RAT) is an integral part of the Athena Science payload. Serving primarily as the geologist's rock hammer, the RAT will expose fresh surfaces of Martian rocks to other instruments on the payload. The RAT also brushes dust and debris from an excavated hole or unaltered rocks. To accomplish these tasks autonomously, the RAT, a sophisticated 3-axis precision-controlled device, was designed. Data products derived from RAT telemetry enable the RAT to also act as an important rock physical properties science instrument. The returned RAT grinding and penetration rate data will be inverted and compared to a rock library on the Earth. The design is also very compact and lightweight: the RAT is contained within a cylinder 128 mm long and 85 mm in diameter and has a mass of 687 g.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Performance Requirements
  5. 3. Operational Considerations
  6. 4. Electromechanical Description of the Rock Abrasion Tool
  7. 5. Data Products
  8. 6. RAT as a Science Instrument
  9. 7. Conclusions
  10. References
  11. Supporting Information

[2] The main objective of the Athena science investigation [Squyres et al., 2003] is to explore two landing sites on the surface of Mars, and to assess the extent to which past aqueous activity at those sites may have yielded habitable environments. A central goal for the Athena payload therefore is to read the geologic record at the landing sites, making compositional and textural measurements of rocks that will reveal evidence about the environments in which the rocks formed.

[3] As is true for many geologic investigations, the search for this evidence may be hindered by subsequent modification of the rocks. Chemical weathering can greatly alter the composition of the near surface of a rock, masking evidence of how the rock originally formed. A variety of other processes, including deposition of aerosols, can coat rock surfaces. Such difficulties are why geologists on Earth carry rock hammers: to allow them to break open weathered or coated rocks and expose fresh surfaces for study. In the case of Mars, the rocks on the surface may have an oxidation or weathering rind. In addition, the surface has a covering of naturally occurring fines that makes all rocks appear similar when observed.

[4] The Rock Abrasion Tool (RAT) is a robotic tool designed to brush debris from rocks and remove the surface rind, providing a fresh interior surface for the instruments to observe. The automated equivalent of a rock hammer and brush, the RAT also generates various useful data products that will help infer rock properties from known rocks on Earth. Thus the RAT has the profile of a scientific instrument in addition to its primary function of enabling science observations.

2. Performance Requirements

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Performance Requirements
  5. 3. Operational Considerations
  6. 4. Electromechanical Description of the Rock Abrasion Tool
  7. 5. Data Products
  8. 6. RAT as a Science Instrument
  9. 7. Conclusions
  10. References
  11. Supporting Information

[5] Some performance requirements for the RAT were derived in a straightforward manner from other characteristics of the Athena payload. The diameter of the area abraded by the RAT must be large enough to allow study by the three in situ instruments on the payload: the Alpha Particle X-ray Spectrometer (APXS) [Rieder et al., 2003], the Mössbauer Spectrometer [Klingelhöfer et al., 2003], and the Microscopic Imager [Herkenhoff et al., 2003]. The Mössbauer Spectrometer has a roughly circular field of view about 15 mm in diameter, and the APXS has a circular field of view 37 mm in diameter. The largest field of view of the three instruments is that of the Microscopic Imager (MI), a square 30 mm on a side (42 mm on the diagonal). All of these instruments, along with the RAT, are placed on rocks using the payload's Instrument Deployment Device (IDD) [Squyres et al., 2003]. The abrasion diameter of the RAT also accounts for the positional uncertainty of the IDD, which is several millimeters. Using these considerations, we arrived at a requirement for the RAT to remove rock from an area 45 mm in diameter.

[6] Other requirements deal with production of a clean, unaltered surface for the instruments to view. Cuttings generated by the grinding process must be removed thoroughly enough that the petrographic texture of the underlying rock is revealed to the Microscopic Imager. The grinding process must not be so aggressive as to cause changes in the mineralogy or elemental chemistry of the underlying rock that are detectable by the Mössbauer Spectrometer or the APXS. Erosion of the grinding heads must not leave behind any residue detectable by either instrument.

[7] Some performance characteristics involve a compromise between what is desired and what is practical. Grinding depth is an example. Very little is known about the depth to which rocks on Mars might be weathered. While it clearly would be desirable to grind arbitrarily deep, practical engineering considerations require that the maximum depth be bounded. A mix of terrestrial experience, engineering restrictions, and geologic judgment led to a grinding depth requirement of 5 mm.

[8] Selection of the penetration rate and rock hardness requirements involved similar considerations. The hardness of rocks at the MER landing sites is unknown. On the basis of Martian rock compositions measured for SNC meteorites and inferred from spacecraft data, dense basalt was chosen as the hardest material into which the RAT must be able to penetrate. Modeling of rover operations profiles suggested that two hours was a desirable length of time that the RAT development should target for a 5 mm penetration. However, initial exploratory uses of the RAT showed that penetration of very strong rocks could extend this operational time to 3 or 4 hours.

[9] Another important design consideration involved the interaction between the IDD and the RAT. The RAT is shown in Figure 1 attached to the end of the IDD. The IDD is required to place the RAT normal to the surface of a rock with an adequate preload, and to maintain it fixed in that state as grinding takes place. All further actuation needed for grinding and brushing are solely functions of the RAT.

image

Figure 1. The RAT, mounted on the 5 axis IDD, is driven up to a target rock for the purpose of excavating a clean 45 mm diameter and 5 mm deep hole. The fresh rock revealed will be accessible by the other IDD mounted payload instrument.

Download figure to PowerPoint

3. Operational Considerations

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Performance Requirements
  5. 3. Operational Considerations
  6. 4. Electromechanical Description of the Rock Abrasion Tool
  7. 5. Data Products
  8. 6. RAT as a Science Instrument
  9. 7. Conclusions
  10. References
  11. Supporting Information

[10] The primary use of the RAT will be to remove the outer surface materials from a rock, exposing the rock's interior to study by all of the Athena instruments. Use of the RAT requires that the intended target be within the work volume of the IDD. It is also an operation that consumes significant time and energy. For both these reasons, the RAT will be used primarily on high-priority scientific targets that have been identified using Pancam and Mini-TES, the two remote sensing instruments on the payload. In some instances the Microscopic Imager, APXS and Mössbauer may be used first on the unmodified surface, and then used again after abrasion by the RAT so that the unmodified and abraded surfaces may be compared. In other instances it may be adequate to apply the RAT immediately, making only subsurface measurements with the instruments. In either case, the typical approach will be to use a single application of the RAT to remove a commanded amount of material from the target.

[11] Once the RAT has been used on a target, every instrument on the payload may view the abraded region. As noted above, all of the in situ instruments have fields of view smaller than the abraded region. Imaged from a typical distance of ∼2 m, the abraded region will be about 80–90 Pancam pixels in diameter, providing good visible/near infra-red color data for the rock interior. The most poorly matched post-abrasion measurement will be with Mini-TES, since the Mini-TES beam diameter at close range is almost twice the diameter of the region abraded by the RAT [Christensen et al., 2003]. One approach to mitigate this issue will be to raster the Mini-TES beam across the abraded region in steps much smaller than the beam diameter, over-sampling the scene and allowing use of spatial deconvolution techniques on the Mini-TES data.

[12] The RAT grinding heads that make actual contact with the target rocks are very small, leveraging the modest preload capability of the IDD. Strong rocks will significantly wear these grinding heads, so use of the RAT must be treated as a consumable resource. An effort will be made to grind only as far into each rock as it is necessary to reach unaltered material. For this reason, time may be taken early in the mission to use the RAT in several small depth increments, with compositional measurements after each increment. This technique could help establish the extent to which typical coatings or weathering extends, and also would be valuable in searching for scientifically interesting compositional gradients through a weathered zone.

[13] Not every rock can be investigated using the RAT, and RAT targets must be chosen carefully. It is best although not necessary that a rock surface should have a fairly planar facet that is large enough to admit the RAT with sufficient margins. The facet must be positioned and oriented such that the IDD can provide an adequate preload, and the rock must be large enough that this preload will not cause the target rock to move. Some care also must be taken to assure that the RAT is positioned so that cuttings generated by the grinding process will not be ejected toward sensitive surfaces like the optics of the forward facing pair of Hazard Avoidance cameras (Hazcams) [Maki et al., 2003].

[14] While the RAT is primarily intended for use in rock, it could also be useful for performing excavations into highly consolidated soil units like duricrust. The rover's wheels may be used for this purpose as well [Arvidson et al., 2003]. And as previously mentioned, the RAT can be used simply for removal of naturally occurring fines from high-priority rock surfaces if weathering is not a problem but coating by fines is. Since fines removal by the RAT only minimally consumes system resources, it may also be employed on an opportunity basis.

[15] As noted above, the RAT will be useful to study the physical properties of rocks, via investigation of penetration rates and motor currents. Therefore the RAT could be used on some rocks primarily to test their mechanical strength properties.

4. Electromechanical Description of the Rock Abrasion Tool

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Performance Requirements
  5. 3. Operational Considerations
  6. 4. Electromechanical Description of the Rock Abrasion Tool
  7. 5. Data Products
  8. 6. RAT as a Science Instrument
  9. 7. Conclusions
  10. References
  11. Supporting Information

4.1. Excavation Description

[16] The depth to which the RAT grinds the 45 mm diameter area is selectable within the range of the RAT's Z axis and the topography of the rock surface. The typical commanded depth is 5 mm, although as much as 15 mm is possible depending on the planar nature of the rock surface and the angular placement of the RAT to the rock surface. There is also a small rotating brush and a larger brush that are used to clean the rock surface, both in conjunction with grinding and independent of grinding. In this way, a rock to be investigated can be studied with the suite of instruments before and after the surface has been excavated.

4.2. Dimensions, Mass

[17] The RAT flight unit shown in Figure 2a is contained within a cylinder 85 mm in diameter and 128 mm long, and has a mass of 687 grams. This size and mass is about the same size and mass as each of the other instrument heads located at the end of the IDD.

image

Figure 2. (a) The RAT shown is a 3 axis torque controlled robotic mechanism. In addition to providing fresh rock surfaces for other payload instruments, the telemetry from the RAT will be inverted and compared to terrestrial specimens to determine the compressive strength of Martian rocks. (b) As seen from below, key elements of the RAT are the two stainless steel brushes for cleaning the excavation, the diamond epoxy cutting heads and the four cylindrical Sm2Co17 magnetic properties experiment magnets.

Download figure to PowerPoint

4.3. RAT Axes, Actuators, and Power Profile

[18] The RAT's three axes of motion are referred to as Rotate, Revolve, and Z. The Rotate axis describes the rotation of the grinding wheel about its center. The Revolve axis describes the revolution of the grinding wheel about the RAT's central axis. The Z axis describes the linear motion of the grinding wheel, allowing it to plunge into a rock. The RAT uses three small DC brushed motors with magnetic encoders to effect grinding and consumes between 8 and 11 watts of power during its operation.

[19] The RAT is used as a stand-alone device in that it does not require any input or movement of the IDD once placed onto a rock surface to facilitate grinding to any desired depth. The IDD must preload the RAT against the rock within a range of 10 and 100 newtons.

[20] The three RAT motors each serve a distinct function and operate in different modes. The RAT grinds a hole in the rock by rotating a grinding wheel at high speed (about 3000 RPM). The grinding wheel is 23.37 mm in diameter that has been cut so that it is 6.35 mm wide, and therefore resembles a “paddle wheel” but is still referred to as a grinding “wheel.” A diamond-impregnated resin is mounted on the outer region of the grinding wheel. As the grinding wheel rotates quickly about its center, it also revolves about the center of the RAT at very slow speeds (0–2 RPM), that is under closed-loop control. This motion then creates a circular grinding area that is 45 mm in diameter. The lower half of the RAT that houses the grinding wheel is then advanced into the rock in small incremental steps (0.05 mm per revolution for dense basalts) until the desired grinding depth is achieved. This incremental step is adjustable through software parameter changes. For friable surface rinds and conglomerate targets, a smaller step will reduce the probability of loosening chunks that could jam the grinding mechanism.

[21] The Rotate motor is used to rotate the two shafts at the bottom of the unit where the grinding wheel and the Rotate brush are attached. These shafts are equally displaced from the center of the housing and centerline of the RAT by 11.11 mm; separated from each other by a distance of 22.22 mm. This motor operates at the bus voltage of the rover while the current drawn by the motor is monitored and used as feedback for controlling both the Rotate and Revolve motors.

[22] The Revolve motor causes the housing that supports the two Rotate shafts to revolve about a common center point so that the grinding wheel and Rotate brush sweep through a 45 mm diameter circle. The primary reason this is done rather than using a single 45 mm diameter wheel that rotates about the center of the RAT is that the speed of the grinding material at the center of the wheel will effectively be zero and will therefore prevent grinding at the center, halting any advancement of the wheel into the rock. Secondly, removal of the cuttings would be problematic with such an approach. There are speed constants for the Revolve motor applied to the grinding algorithm equations. The speed at which the Revolve housing turns is inversely proportional to the amount of current the Rotate motor is drawing. Thus the Revolve housing will turn at its maximum speed when there is no load on the Rotate motor. The maximum speed is also a set point that is selectable for each grinding task. As the current draw from the Rotate motor increases, the Revolve motor slows down so as to not cut into the rock too aggressively which could cause stability problems. When the Rotate motor draws the same amount of current as defined by the current set point, the Revolve motor will stop advancing the grinding wheel across the rock until the current draw drops and the rock material is worn away. The Revolve motor then starts to again advance the grinding wheel.

[23] The Z axis moves the lower section of the RAT in a linear fashion. This movement is used to engage the grinding wheel with the rock and allow a user-selectable excavation depth to be achieved. This axis has a very high resolution of motion so that a very small amount of material can be removed with each pass of the grinding wheel as it makes a full revolution about the center of the RAT.

4.4. RAT Brushes

[24] Figure 2b shows two small brushes on the RAT, the Rotate and Revolve brushes. Another brush, the Tooth brush, is mounted on the forearm of the IDD. The two brushes on the RAT work together to remove both naturally occurring fines from a rock surface and cuttings from within the hole that is produced. The Rotate brush removes cuttings from the abraded area as they are generated and deposits the cuttings on the perimeter of the hole. As the abrasion continues, the walls of the hole form and cuttings collect. The Revolve brush ensures that the cuttings that collect do not obstruct the Rotate brush as it continues to clean. With every revolution of the grinding wheel about the center of the RAT, the Revolve brush pushes material on the perimeter away from the edge of the hole so that the Rotate brush can eject the cuttings onto a relatively debris-free perimeter. This dual-brush action produces a sufficiently clean surface to accommodate high-resolution observations by the Microscopic Imager.

[25] The Tooth brush mounted on the IDD is used to clean the grinding heads. The cutting elements in the grinding head may become covered in rock cuttings or the space between cutting elements could become clogged with rock dust, thereby reducing the cutting action of the grinding heads. To clean the grinding heads, the RAT is positioned to a point within reach of the brush bristles. The Rotate and Revolve axes move the RAT grinding wheel through the brush bristles to provide the cleaning action necessary.

4.5. Contact With the Target

[26] When the RAT is placed onto a feature of interest, whether it is a rock, duricrust, blocky soil, or a layer of dust, the IDD preloads the RAT onto the feature. If the preload should drop below 5 N as a result of rover or rock movement, the RAT will abort the operation. This event will be uploaded along with all other RAT grinding data.

5. Data Products

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Performance Requirements
  5. 3. Operational Considerations
  6. 4. Electromechanical Description of the Rock Abrasion Tool
  7. 5. Data Products
  8. 6. RAT as a Science Instrument
  9. 7. Conclusions
  10. References
  11. Supporting Information

[27] During all phases of RAT operations (i.e., calibration, seek-scan, grinding, diagnostic or brush station), the RAT application software stores sensor values and information about the RAT machine state. RAT data products will be comprised of time-ordered data records that include motor positions, motor currents, contact switch states, temperature sensor values, bus voltage, software/algorithm states, error bit fields, etc. Data records will be written to the detailed data product at a nominal rate of 2-Hz. Anomaly reports will include data stored at 8-Hz.

6. RAT as a Science Instrument

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Performance Requirements
  5. 3. Operational Considerations
  6. 4. Electromechanical Description of the Rock Abrasion Tool
  7. 5. Data Products
  8. 6. RAT as a Science Instrument
  9. 7. Conclusions
  10. References
  11. Supporting Information

6.1. Introduction

[28] Recent studies from Mars Global Surveyor and Mars Odyssey suggest that Mars may be dominated by two types of igneous rocks: basalts and andesites. Additional hematite deposits, aeolian features such as dunes, and impact crater materials collectively indicate that Mars has a diversity of rock types. Once the Mars Exploration Rovers (MER) set down on the surface of Mars early in 2004, the primary tools for investigating the mechanical properties of the varied rock types will be the RAT and the rover wheels.

[29] Stress on a rock is created when a force is applied to it. Since rocks are composed of a variety of minerals, the way in which a rock reacts to an applied stress can be characteristic, or even diagnostic, of the minerals that compose the rock. The physical property of a mineral largely depends on the strength of the chemical bonds and the amount of impurities in the crystal lattice. One method used by geologists to distinguish different rock types is “rock hardness.” Rock hardness denotes the cohesiveness of a rock and rocks are tested for hardness by loading or stressing a rock core until brittle failure occurs. The fracture strength of a rock is defined as the maximum stress necessary to initiate rock failure. By measuring the fracture strength, the cohesiveness and density (rock hardness) of a rock can be identified. Table 1 shows that rock types vary in hardness from very weak in the case of weathered sedimentary rock to very strong, such as diktytaxitic-textured, porphyritic basalt, respectively. In most rocks, the main factors controlling rock hardness are porosity, grain size, and shape. All three of these factors result from the chemical bonding associated with the mineral grain-to-grain contact. Generally on Earth, the harder the rock is, the higher the surface area of mineral grain-to-grain contact. By decreasing the grain size and porosity in rocks, the surface area of grain contacts and hardness increases. For example, this is especially true for fine-grained igneous rocks such as porphyrictic basalt, but not true of highly vesicular cinder.

Table 1. Classification of Rock Hardness From Attewell and Farmer [1976]
Strength ClassificationStrength Range, MpaRock Type
Very Weak10–20Weathered and weakly compacted sedimentary rocks
Weak20–40Weakly cemented sedimentary rocks, schists
Medium40–80Competent sedimentary rocks; low-density coarse-grained igneous rocks
Strong80–160Competent igneous rocks; some metamorphic rocks and fine-grained sandstones
Very Strong160–320Dense sedimentary rocks such as quartzites and basalts

[30] Sedimentary rocks generally have a range of porosities depending on the rock makeup (e.g., amount and composition of the clasts of specific shapes, matrix, and cementing agent). As a general rule, the greatest percentage of sedimentary rock types with generally high porosities and low rock hardnesses are fluvial, colluvial, and aeolian. These rocks can be differentiated from fine-grained and lower porosity igneous rocks of much greater harnesses using the RAT. Although metamorphic rocks have not yet been identified on Mars, the presence of enormous mountain ranges such as Thaumasia highlands and Coprates rise [Dohm et al., 2001] and gigantic impact crater basins that are partly surrounded by mountainous topography may suggest their occurrence. For medium to high-grade metamorphic rocks on Earth, rock hardness is generally lower due to the development and orientations of mineral grains; these rocks may be difficult to distinguish from sedimentary rocks. For low-grade metamorphic rocks where foliation and fracturing does not develop and compaction dominates, the porosity of the rock will be reduced and the hardness increased.

6.2. Compositional and Depositional History of the Target Rock

[31] The RAT's main function on Mars is to remove the exposed outer surface of rocks. In performing its main function, the RAT also provides secondary data such as motor current and position. From the motor currents, torque can be directly calculated, and subsequently torque can be empirically correlated to the density and hardness of the rock. Combining these parameters, along with a close-up view from the Microscopic Imager of the fresh surface, one should be able to assess the compositional and depositional histories of the rock.

6.3. Insight to Aerosol Coatings

[32] RAT torque and penetration rate telemetry, combined with views from Microscopic Imager, Pancam and Hazcams will provide important new insights into the extent to which rocks become coated with dust and other aeolian deposits. The data will also be used to evaluate the extent to which potential coatings become indurated in much the same way that coatings form on Earth, including classical high silica-bearing varnish that forms on basalt flows in Hawaii [e.g., Farr and Adams, 1984].

6.4. Extent of the Surface Oxidant

[33] An important avenue of investigation with the RAT will be to use the natural and exposed surfaces and relevant measurements to explore the extent to which weathering, particularly oxidation, extends beneath rock surfaces. Similarly, an excellent use of the RAT will be to leverage its capabilities to remove loose surface material from duricrust and cemented soil [Binder et al., 1977; Mutch et al., 1976; Moore et al., 1987] to expose the underlying indurated soil. In addition to exposing an area of duricrust, the RAT may be used to grind the duricrust as if it were rock. This experiment will be accomplished with the same duricrust target as planned for trenching operations using the rover wheels [Moore et al., 1999; Arvidson et al., 2003] and will focus on a campaign designed to test the competing hypotheses that duricrust forms by evaporation of soil water, leaving behind sulfates and other salts, or perhaps as a direct deposition as volcanic ash [Bishop, 2001].

6.5. Rock Strength

[34] The RAT's mechanical data will include time series of motor currents and penetration distances measured during the RAT operation. These data will be used to estimate physical properties of the targets. The flight data will be compared to a comprehensive database of similar measurements conducted during grinding operations in the laboratory on a large number of rock targets. The measurements will be accomplished with an engineering model of the RAT, identical in design to the flight versions of the system.

[35] In the time period leading up to surface operations and after the mission is completed, the laboratory measurements will be correlated with rock strength parameters, particularly cohesive strength and angle of internal friction, under the assumption that rock failure during grinding occurs when the Mohr-Coulomb failure envelope is exceeded. The intent is to use the laboratory data as a transfer function to retrieve basic rock properties from the RAT telemetry. To that end a wide range of rock types has been collected and will be used to establish the database. The targets include hard crystalline igneous and metamorphic rocks with varying textures and degrees of induration (basalts to rhyolites, serpentinites to orthoquartzites), together with a large array of sedimentary rocks (various sandstones and shales, carbonates), and materials that are likely to emulate the properties of Martian duricrust (caliche, slightly to highly indurated tuffs with silica and sulfate cement and sinter-induced lithification).

6.6. Topography

[36] The topography of the rock removed through grinding can also be reconstructed. By looking at data such as the current drawn by the Rotate motor, the time it takes to make each revolution of the Revolve axis and the position of the Revolve axis during periods of high and low current draw of the Rotate axis, we can extract an approximate image of the topography of the rock that has been removed. The topographical data will be low resolution because the geometry of the grinding wheel is a 23.37 mm long rotating cutter and not a single point tool that would also require a radial axis in order to provide a high quality map of the removed material. Nevertheless, we can build a topographical model by analyzing each layer of material that is cut away to form a 3D image of the surface. For a dense basalt, the thickness of this layer is expected to be 0.05 mm. A topographical map of the material removed will also play an important role in determining the strength of the rock. If the amount of material removed is known, we can determine how much energy per unit volume of material was required. This ratio can then be compared to data within the rock library to see which type or types of rock tested more closely represents the excavated Mars rock.

6.7. Strength and Direction of Sediment Driving Winds

[37] Another use of the RAT is to take advantage of that fact that grinding and brushing activities will produce piles of fine-grained debris that will be out of equilibrium with the atmosphere, both dynamically and chemically. That is, the surfaces may be subjected to erosion by wind, providing an indication of the strength and direction of sediment driving winds encountered while the rovers are in the vicinity of the targets. Monitoring of the deposits will be accomplished using the Pancam to search for evidence of erosion in a manner similar to what was done during the Viking Lander Missions [Arvidson et al., 1983], and for changes in spectral properties after the exposure of these fine grained materials to ambient conditions, perhaps via hydration or oxidation. Pancam and Mini-TES will thus be used to monitor these surfaces while the rovers are in the vicinity of the targets.

[38] The brush can also be used to sweep soil-like surfaces to expose underlying outcrops or more indurated materials. This operation may be attempted, particularly during any extended missions. It is expected that other uses of the RAT will develop during the course of mission operations as the system is used and its performance and capabilities are evaluated.

6.8. Magnetic Properties Experiment

[39] The strong crustal magnetic anomalies on Mars, discovered by Mars Global Surveyor in 1997 [Acuña et al., 1999], demonstrate that significant amounts of a strongly ferromagnetic mineral are present in certain Martian rocks. Such a ferromagnetic phase may also be present in rocks at the landing sites of the Mars Exploration Rovers, although the proposed landing sites are not located in regions where strong magnetic anomalies have been detected. The presence of titanomagnetite in most SNC meteorites supports the hypothesis that Martian surface rocks do contain moderate amounts of a strongly magnetic phase.

[40] In order to detect and characterize such magnetic phases, four permanent magnets (of different strengths) are mounted in the lower part of the RAT housing. These magnets are shown on the bottom of the RAT in Figure 2b. The magnets are designed to attract magnetic dust particles liberated during grinding [Madsen et al., 2003]. Two of the RAT magnets are identical. The maximum value of the magnetic field B is 0.28 T and the maximum value of the magnetic field gradient ∇∣B∣ is 350 T m−1. The remaining 2 magnets are weaker: For magnet “2”, B = 0.1 T and ∇∣B∣ = 120 T m−1, while the maximum values for magnet “3” are B = 0.07 T and ∇∣B∣ = 80 T m−1. When the RAT is operating the active surfaces of these magnets will be about 4 mm above the rock surface.

[41] Figure 3 shows a cross section of the three types of RAT magnets. It is seen how the above mentioned values of magnetic field and field gradient are realized by an appropriate internal geometry of the RAT magnets. Each of the units contains an axially magnetized Sm2Co17 magnet of cylindrical geometry.

image

Figure 3. The four RAT magnets are simple aluminum structures each holding one axially magnetized Sm2Co17 magnet of cylindrical geometry. The sketch shows the geometry of the 3 different types of RAT magnet.

Download figure to PowerPoint

[42] Data sets returned from this experiment on Mars will consist of Pancam images of the RAT with dust on the RAT magnets. From such images, information on spectral and magnetic properties of the dust on the RAT magnets will be retrieved. RAT dust properties extracted from images from Mars can be translated into rock properties by comparison with simulation experiments in the laboratory (C. S. Binau et al., Simulation of the rock abrasion tool magnet experiment on the Mars Exploration Rovers, submitted to Earth and Planetary Science Letters, 2003).

[43] Figure 4 shows the result of one such simulation experiment, obtained in a simulated RAT setup grinding a basaltic rock from Kjalarnes, Iceland. After grinding, the 4 magnets were imaged in a way that simulates imaging by Pancam during the mission (with respect to the focal distance, f-number, resolution and defocus). As the image in Figure 4 shows, the amount of dust adhering to magnets 1, 2 and 3 is consistent with the relative strengths of these magnets. From the relative reflectance and diameter of the 4 dust patterns, information on the mineralogy and magnetic properties of the abraded rock can be derived. Additional information from Mössbauer and Alpha-Particle-X-Ray spectroscopy will help to identify the magnetic phases present in the rock.

image

Figure 4. Images of dust on the 4 different strength RAT magnets after grinding a basaltic rock from Kjarlarnes, Iceland.

Download figure to PowerPoint

7. Conclusions

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Performance Requirements
  5. 3. Operational Considerations
  6. 4. Electromechanical Description of the Rock Abrasion Tool
  7. 5. Data Products
  8. 6. RAT as a Science Instrument
  9. 7. Conclusions
  10. References
  11. Supporting Information

[44] To provide the Athena payload with the equivalent of a geologist's rock hammer, an altogether new robotic device was developed, the Rock Abrasion Tool. To reach its primary goal of providing access to fresh rock surfaces for the in situ instruments, the RAT was endowed with precision controllable actuators. The actuator and control profile of the RAT augments its geologist's hammer and rock brushing functions as data products generated by the RAT will be inverted and compared to RAT grindings on terrestrial rock specimens. The RAT is primarily a tool in service of the science payload but it also serves as a science instrument providing its own insight to the density and strength of rocks on Mars.

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Performance Requirements
  5. 3. Operational Considerations
  6. 4. Electromechanical Description of the Rock Abrasion Tool
  7. 5. Data Products
  8. 6. RAT as a Science Instrument
  9. 7. Conclusions
  10. References
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Performance Requirements
  5. 3. Operational Considerations
  6. 4. Electromechanical Description of the Rock Abrasion Tool
  7. 5. Data Products
  8. 6. RAT as a Science Instrument
  9. 7. Conclusions
  10. References
  11. Supporting Information
FilenameFormatSizeDescription
jgre1706-sup-0001-tab01.txtplain text document0KTab-delimited Table 1.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.