The Design and Engineering of Curiosity
Page 30
The pixel scale of a MAHLI image can be derived from the working distance (instrument team convention) using the following formula:17
The MAHLI team has derived an empirical relationship between motor count and working distance, based upon measuring objects of known size and distance on both Earth and Mars:18
in which
w = working distance (instrument team convention)
m = motor count
a = 0.576786
b = –11.8479
c = 2.80153 × 10–3
d = –2.266488 × 10–7
e = 6.26666 × 10–12.
Table 7.8 is a lookup table relating motor count to working distance, pixel scale, and image size. The last three columns provide a guide to the size of sediment grains that can be resolved in images taken at different distances, according to the classic Wentworth (1922) sediment classification scheme. It takes at least 2 pixels across an object to be able to detect it, so sand grains are resolvable from MAHLI standoff distances under about 1 meter, while silt grains are only resolvable at the very closest standoff distances, and then only the coarsest silt grains that have good contrast against the background. As a rule of thumb, if you can see individual grains in a sedimentary rock in a Mastcam image, you’re looking at a conglomerate; if you can see individual grains in a MAHLI image but not in a Mastcam image of the same target, it’s a sandstone; and if you can’t detect grains even with MAHLI, it’s a siltstone or mudstone.Table 7.8. MAHLI image dimensions and pixel scale with respect to motor count. Lines in bold represent very common motor counts for MAHLI, near standard target standoff distances of 25, 5, and 2 centimeters.
Motor count (cover closed)
Motor count (cover open)
working distance from MAHLI front window (cm)
toolframe or standoff distance (cm)
pixel scale (μm)
Width of 1600-pixel image (cm)
Depth of field (near, mm)
Depth of field (far, mm)
width of 10mm scale bar (pixels)
Max diameter of a granule (4mm) (pixels)
Max diameter of a grain of sand (2mm) (pixels)
Max diameter of a grain of silt (62.5 μm) (pixels)
4475
12600
231
229
819
131
-891
4531
12
4.9
2.4
<1
4375
12700
84.0
82.1
302
48.4
-155
261
33
13
6.6
<1
4275
12800
50.7
48.8
185
29.7
-60.9
83.2
54
22
11
<1
4175
12900
35.9
34.0
133
21.3
-31.7
39.4
75
30
15
<1
4075
13000
27.6
25.7
104
16.6
-19.1
22.4
96
38
19
<1
4061
13014
26.7
24.8
101
16.1
-18.0
20.9
99
40
20
<1
3975
13100
22.2
20.3
85.1
13.6
-12.7
14.2
117
47
23
<1
3875
13200
18.5
16.6
71.9
11.5
-8.9
9.7
139
56
28
<1
3775
13300
15.7
13.8
62.1
9.94
-6.6
7.0
161
64
32
1
3675
13400
13.5
11.6
54.6
8.73
-5.1
5.2
183
73
37
1.1
3575
13500
11.9
10.0
48.6
7.78
-4.0
4.0
206
82
41
1.3
3475
13600
10.5
8.6
43.8
7.00
-3.2
3.2
229
91
46
1.4
3375
13700
9.3
7.4
39.7
6.36
-2.7
2.6
252
101
50
1.6
3275
13800
8.4
6.5
36.4
5.82
-2.2
2.2
275
110
55
1.7
3175
13900
7.5
5.6
33.5
5.36
-1.9
1.8
299
119
60
1.9
3077
13998
6.9
5.0
31.0
4.97
-1.6
1.6
322
129
64
2.0
3075
14000
6.8
4.9
31.0
4.96
-1.6
1.6
323
129
65
2.0
2975
14100
6.2
4.3
28.8
4.61
-1.4
1.4
347
139
69
2.2
2875
14200
5.7
3.8
26.9
4.31
-1.2
1.2
371
148
74
2.3
2775
14300
5.2
3.3
25.3
4.04
-1.1
1.1
396
158
79
2.5
2675
14400
4.8
2.9
23.8
3.81
-1.0
1.0
420
168
84
2.6
2575
14500
4.4
2.5
22.5
3.59
-0.9
0.9
445
178
89
2.8
2475
14600
4.1
2.2
21.3
3.41
-0.8
0.8
470
188
94
2.9
2411
14664
3.9
2.0
20.6
3.29
-0.8
0.8
486
194
97
3.0
2375
14700
3.8
1.9
20.2
3.24
-0.7
0.7
495
198
99
3.1
2275
14800
3.5
1.6
19.3
3.08
-0.7
0.7
519
208
104
3.2
2175
14900
3.3
1.4
18.4
2.94
-0.6
0.6
544
218
109
3.4
2075
15000
3.0
1.1
17.6
2.82
-0.6
0.6
568
227
114
3.6
1975
15100
2.8
0.9
16.9
2.70
-0.5
0.6
593
237
119
3.7
1875
15200
2.6
0.7
16.2
2.59
-0.5
0.5
617
247
123
3.9
7.4.2.5 Calibration target
MAHLI’s calibration target is attached to the robotic arm shoulder azimuth actuator (Figure 7.13). A preflight photograph of the calibration target is shown in Figure 7.14. The calibration target contains red, green, blue, and gray color swatches made from the same material used in the Mastcam calibration target, leftover materials from the Mars Exploration Rover Pancam calibration target. There is also a fluorescent chip made of a material called SpectraFluor Red that glows red (at a wavelength of 626 nanometers) when illuminated with the MAHLI ultraviolet LEDs (365 nanometers). An opal glass bar target has a chart modeled on the US Air Force 1951 Resolution Test Chart, designed to monitor camera focus and resolution performance over time. The calibration target is mounted vertically on the rover, which was intended to discourage dust settling and keep it relatively clean. Unfortunately, the calibration target was coated with a thin film of dust thrown up during landing, but the calibration target functions adequately for its primary purpose of checking that there is no drift in camera focus. The calibration target has been imaged on sols 34, 165, 179, 322, 411, 591, 825, 989, 1091, 1157, 1340, 1519, 1632, and 1696. MAHLI has imaged the Mastcam calibration target on sols 544, 707, and 1028. MAHLI also images the Mastcam and ChemCam calibration targets in nearly every self-portrait.
Figure 7.13. Photo of the rover taken during assembly at JPL, showing the location of the calibration target. NASA/JPL-Caltech release PIA14289. Insets: two images of the calibration target, taken before departing Earth using a DSLR camera (left) and after landing on Mars, by MAHLI (right). Image 0034MH0000460010100041E01. NASA/JPL-Caltech/MSSS/Emily Lakdawalla.
Figure 7.14. MAHLI calibration target, taken during spacecraft assembly. Inset images are from Mars, images 0034MH0000440010100031C00 and 0034MH0000450010100C00. NASA/JPL-Caltech/MSSS/Emily Lakdawalla.
There are several quirky elements in the MAHLI calibration target. A cartoon of “Joe the Martian” is meant as a thank-you to the public for the opportunity to conduct the MAHLI investigation and as an invitation for children to follow the Curiosity mission. The Greek letters “γδβγ” are printed within the “0” of the “1.0” text on the bar target. A 1909 United States penny is embedded in the bottom of the calibration target. It is intended as an homage to field geologists’ practice of placing a coin or other small object on a rock outcrop to provide scale before taking a photo, and the MAHLI team often includes a picture of it in public releases of MAHLI images (Figure 7.15). 1909 was the first year that the Lincoln cent was issued, and would have been a century before the year of Curiosity’s launch; unfortunately, the launch delay to 2011 obscured the significance of the date on the 1909 coin. The coin is 19 millimeters in diameter.
Figure 7.15. A mosaic of nine MAHLI images on a conglomerate target taken at the Darwin waypoint, sol 400. The penny is 19 millimeters in diameter. NASA/JPL-Caltech/MSSS image release PIA17362.
7.4.2.6 Bad pixels and blemishes
MAHLI has a few dozen dark specks in all images. These have been constant throughout the mission, and are caused by microscopic particles on the detector. The MAHLI team is more concerned about the possibility that operating the camera close to the surface will invite dust to settle on the optics when the cover is open, especially when operating near freshly drilled rocks and their piles of fine drill tailings. They try to take flat-field images of the sky about once every 180 days, in part to watch for new dust particles affecting their images. One dust particle was detected on the sapphire window in a Mastcam image of MAHLI taken on sol 617.
Like the Mastcams, MAHLI is susceptible to shutter smear (see section 7.2.1.5). In fact, MAHLI experiences shutter smear at longer exposures than Mastcam and MARDI because it takes longer for MAHLI to read out its images. However, it doesn’t affect science data and is mostly only noticeable in self-portrait photos, when white surfaces of the rover can appear smeared. Hot pixels often cause streaks running down images due to shutter smear. MAHLI landed with a number of hot pixels, and new ones have appeared during the course of the mission; some heal, but others have persisted. While these are cosmetically annoying, they don’t affect the quality of the data for science.
7.4.3 Using MAHLI
Because using MAHLI almost always requires using the arm, there are fewer opportunities for MAHLI science than for Mastcam science. A substantial portion of the MAHLI data set is engineering support imaging and rover self-portraits. Various types of MAHLI imaging also provide context and support for drilling, scooping, sample dumping, and APXS activities, helping to build a multi-instrument data set. When MAHLI can reach them, it’s often pointed at ChemCam shot points. Occasionally, MAHLI captures large mosaics, allowing detailed study of the sizes, shapes, colors, and distributions of grains within a rock for sedimentology studies.
7.4.3.1 MAHLI nested target imaging
A major use of MAHLI is to capture sets of nested images of targets that are usually also APXS targets, including drill sites. Usually, MAHLI takes a single context image from a standoff distance of 25 centimeters, achieving a scale of about 100 microns per pixel. Then it moves to a standoff distance of 5 centimeters and takes a z-stack at about 31 microns per pixel, often repeating the observation from a slightly different position for stereo imaging purposes. This image scale is the same as that of MI images taken by the Mars Exploration Rovers, making it simple to compare close-up data from Spirit or Opportunity APXS and MI with Curiosity APXS and MAHLI. For some observations, MAHLI acquires an observation from a standoff distance of only 1 or 2 centimeters, with a “best” resolution of 16 to 21 microns per pixel (Figure 7.16). To capture these, the MAHLI team asks the rover planners to get MAHLI as close as possible to the target, which varies depending upon the target’s topography and reachability. Sometimes MAHLI will do these observations at night, so that the LEDs can illuminate the target with light of a well-known intensity, allowing the team to directly compare the color of one target to another imaged at another location. Sometimes MAHLI will take nested images of targets both before and after brushing.
Figure 7.16. MAHLI images of target “Mojave” taken at night on sol 809 under LED illumination at three different working distances, after the target had been brushed. Images 0809MH0004440010300853C00, 0809MH0004450010300857C00, and 0809
MH0004460010300905C00. NASA/JPL-Caltech/MSSS/Emily Lakdawalla.
7.4.3.2 Mosaics
It takes close cooperation between the rover planners and MAHLI team to produce a MAHLI mosaic, so there have not been many, but they are scientifically productive on sedimentary targets with varying grain size (e.g. Figure 7.16). It’s also fun and educational to use MAHLI to image a vertical rock face from a low angle not accessible by Mastcam; these “dog’s eye” views are often expanded into mosaics. A list of MAHLI mosaics is in Table 7.9.Table 7.9. Sols and names of MAHLI mosaics as of sol 1648.
Bradbury Group
Pahrump Hills
North of the Dunes
South of the Dunes
66 Rocknest Scoop 1 Trough
67 Rocknest Scoop 2 Trough
84 Self Portrait
85 Self Portrait (stereo)
154 Persillion
158 Tindir
168 JK/YKB Drill Candidate Site
177 Self Portrait
230 John Kllein Hole and Cuttings
270 McGrath
283 Cumberland Drill Site
291 Narrows_3
292 Narrows_3
303 Point Lake
322 Ailik_RP
324 Fleming
387 Ruker
398 vein_mosaic
400 vein_mosaic
400 conglomerate mosaic_left
400 conglomerate mosaic_right
442 Cooperstown
487 Cumberland Dump Pile
550 Bungle Bungle
583 Square_Top
584 Square_Top Dogseye
585 right of Square_Top
585 Square_Top
585 Rock face right of Square_Top