Figure 2.31. Artist’s concept of the rover pulling to the end of its bridle as the Mars Lander Engines fire to maintain a steady rate of descent. NASA/JPL-Caltech release PIA14839.
As the rover descended on its cables, it also deployed its landing gear. Pyros fired to separate the rear bogies from the rover body 0.7 seconds after the rover separated. The bogies fell, pulling downward on the bent rockers, and locking them into their final, straight positions. After the rover reached the end of the bridle, another pair of pyros fired to separate both rockers.53 Finally, just before touchdown, one more pyro fired to release the differential restraint; waiting until the very last moment kept the wheels as coplanar as possible for touchdown, and would allow the landing gear to passively accommodate any surface roughness.54 One thing the landing gear could not handle however, would be the presence of a rock more than 66 centimeters tall positioned to spear the rover’s belly pan. HiRISE images had shown few such rocks in the landing ellipse, but bad luck could win, and MSL had no active terrain hazard avoidance capability.
Throughout, the descent stage should have continued to drop at a slow rate of 0.75 meters per second. It should then have taken 15.67 seconds for the rover’s wheels to touch the ground. However, the actual time was 17.9 seconds, far longer than estimated. That is, the rover actually descended slower than planned, at only 0.6 meters per second at the moment of touchdown. Moreover, the rover was still drifting horizontally at more than 0.1 meters per second at the moment of touchdown, more than twice as fast as expected.55 This slower-than-expected descent, right at the moment of touchdown, was a very serious error. Rob Manning explains:
We were to discover after MSL had landed on Mars that we had missed a crucial item. The long list of variable parameters had not included one that should be obvious: gravity. In the simulations, the EDL team used a fixed value for gravity that was rather generic for that part of Mars. We failed to take into account that the shape of the surrounding terrain and hills might affect the actual gravity, and because we didn’t try other values, we didn’t notice just how sensitive the landing was to being slightly off with the value the team had chosen. The value for Mars gravity used in the simulation turned out to be slightly too high – very slightly, only 0.1 percent – but significant enough that MSL’s slowest-ever landing was even slower than we expected.56
Had the value for gravity been off by 0.1% in the other direction, the maximum design touchdown velocity could have been exceeded, potentially damaging the mobility system.57 Fortunately, the error was in a safe direction, and the rover touched down on its wheels very gently at 05:17:57, or 431 seconds after entering the Martian atmosphere. At that moment, the rover computer stopped the descent of the descent stage, and gave command of the descent stage to the descent stage thruster system computer.58 The rover commanded pyros within the bridle exit guides on the rover’s top deck to fire guillotine-like blades that cut through the three bridle cables and the umbilical. Spring-loaded spools within the bridle exit guides retracted the cut ends of the cables attached to the rover, and a tensioned cable that had unwound with the last few meters of the umbilical lifted the cut ends of the umbilical and bridle cables dangling from the descent stage. The Curiosity rover was all by itself on the surface of Mars – but wasn’t yet out of danger.
The descent stage hovered for 0.7 seconds. To avoid dragging rocket exhaust across the rover, it needed to depart the rover either forward or backward, not sideways. Because the rover was landing to the north of the eventual science target, the descent stage had been commanded to depart whichever of those two directions was the more northerly, taking it away from the likely drive direction.59 The rover knew it had landed facing east-southeast, so the descent stage pitched backward and then burned the four canted engines at full throttle for 6 seconds, sending the descent stage on a long parabolic arc away from the rover, to a crash landing 650 meters away about 20 seconds later.60 Throughout powered descent, it had burned 270.4 kilograms of fuel, leaving 119 kilograms of usable hydrazine in the tanks during the crash.
Back on Earth, engineers were waiting for three distinct signals to confirm that the landing had been successful and that the rover and descent stage were safely separated. Jody Davis announced the first at 05:31:45 UTC, when she noticed that the Mars Lander Engines had throttled down to half their former power, indicating that the descent stage was no longer supporting the weight of the rover: “Tango Delta nominal.” Several seconds of quiet followed that comment, because the landing would not be over safely until the descent stage had disconnected and flown safely away.
David Way announced the second positive landing signal when he noticed that the Rover Inertial Measurement Unit was no longer reporting a changing position: “RIMU stable.” The rover was therefore not being dragged by a connection to the descent stage, nor was it sliding down a slope, or tumbling off a cliff. The third announcement came from EDL communications engineer Brian Schratz, who was monitoring the strength of the UHF radio signal between rover and orbiter, which would vary (or worse, disappear) if the descent stage dragged the rover off the ground, or landed atop the rover. Eight seconds after landing, he announced “UHF strong.”61
The last two announcements collided with each other over the microphones. Adam Steltzner walked over to Allen Chen while pointing to Schratz, asking him to repeat himself; “UHF strong,” Schratz said again. Steltzner tapped Chen on the shoulder and gave him a thumbs up signal. “Touchdown confirmed,” Chen said. “Time to see where our Curiosity will take us.” The room erupted.
2.4 CURIOSITY ON MARS
It had all gone precisely according to script. Curiosity’s landing had been targeted at 4.5965°S and 137.4019°E. The actual landing location was 4.5895°S, 137.4417°E. Curiosity had arrived only 2385 meters away from its intended target, slightly downrange and to the north of the center of the landing ellipse. In computer simulations of the landing, only 24% of simulated landings got closer to the target.62
Curiosity remained in contact with Mars Reconnaissance Orbiter and Odyssey for another six minutes after landing. That was long enough for Odyssey to receive the first data that Curiosity returned from the surface of Mars, and dutifully relay the images onward to Earth. As the dust settled, Curiosity snapped photos with its belly-mounted Hazcams, giving it a fish-eye view of the ground immediately around the rover. Months prior, the mission had offered the science team a choice: receive the rear Hazcam image first, or the front one first? The mission had assumed that the scientists would want to see the forward view first, because the view of Mars would be less obscured by hardware. The science team replied that the first image is not about science; it’s about seeing wheels on the dirt. They requested that the rover’s first image show a wheel in contact with the ground.63
So the first image to arrive on the computer monitors of the landing engineers, two minutes after landing, was a tiny 64-pixel-square thumbnail from the rear Hazcam that was nevertheless big enough to show the horizon, the sky very brightly lit by the afternoon Sun, and in the shadows a wheel clearly sat on the surface. “We are wheels down on Mars,” an engineer stated into the microphone. The celebration on Earth for that first photo was even louder than that for the successful landing (Figure 2.32). By the time Odyssey set below the horizon, it had returned a 256-pixel-square version of the same image, as well as a view from the front Hazcam (Figure 2.33). The images were mottled with dust, some of it still swirling in the air, some of it stuck to the lens caps on the Hazcams.
Figure 2.32. MSL team members in the Mission Support Area celebrate after the successful landing and return of the first tiny Hazcam image, which is barely visible on the screen in the background. NASA photographer Bill Ingalls stood on a table and poked his camera above a similar monitor to catch the team’s reaction in this photo.
Figure 2.33. MSL’s first views of its landing site. Top: Rear Hazcam (RLA_397502188EDR_D0010000AUT_04096M1), taken at 5:18:39, less than a minute after landing. Bottom: Front Hazcam (FL
A_397502305EDR_D0010000AUT_04096M1), taken at 5:20:37, about 3 minutes after landing. NASA/JPL-Caltech photos.
Curiosity lost contact with both Mars Reconnaissance Orbiter and Mars Odyssey at about the same time, at 05:23:53, as both spacecraft set below the horizon. Contact with Odyssey was lost earlier than expected because the spacecraft had gone slightly long, causing Odyssey to set behind the peak of the mountain at the center of the crater. Already Curiosity was on its own, on the far side of Mars, out of contact with Earth.
Two hours later, Odyssey passed above the horizon to the west of the landing site. In the intervening time, Curiosity had stored additional Hazcam images, taken both before and after releasing their lens caps. A close look at the new rear Hazcam image revealed something astonishing: a feature visible on the horizon in the image taken immediately after landing was no longer visible in an image taken an hour later. The smudge on the horizon in the first photo returned from Mars was later determined to be the plume of dust rising from the impact site of the descent stage, 650 meters away (Figure 2.34).
The cruise stage, aeroshell, and descent stage had all done their work admirably. The rover, on Mars, still had the brains of an interplanetary spacecraft. The next major task for the mission was to teach the spacecraft to become a Mars rover.
Figure 2.34. Cropped sections from two rear Hazcam images from landing day. Left: RLA_397502188EDR_D0010000AUT_04096M1, taken at 5:18:39, less than a minute after landing, includes a lumpy plume on the horizon, in the right direction to be the impact plume from the descent stage; the air appears to be cloudy with dust thrown up by the landing rockets. Right: the same region from RLA_397504876EDR_F0010000AUT_04096M1, taken about an hour later at 6:03:26, contains no such plume. Bright dots near the image center are internal reflections within the camera caused by the bright Sun being in the camera field of view. NASA/JPL-Caltech.
2.5 EPILOGUE: VIEWS OF THE CRUISE HARDWARE
The day after the landing, Mars Reconnaissance Orbiter HiRISE imaged the landing site again, catching all of the hardware on the ground (Figure 2.35). The rover was visible as a box on the surface, the descent rocket blast zone surrounding it like butterfly wings. The lighter-colored impingement zones of the four canted descent rockets looked like lighter dots on the wings (Figure 2.36). The crash sites of the heat shield, descent stage, and parachute were arrayed around the rover. The descent stage was marked with an extended spray of ejecta more than 100 meters long. Engineers suspect that its remaining fuel may have detonated on impact, blasting the spacecraft to pieces.
Figure 2.35. HiRISE image of the MSL landing site, sol 1 (August 7, 2012). The impact sites of the backshell, descent stage, and parachute are to the left of the blast zone that marks the rover, uprange; the heat shield is downrange, to the right. HiRISE image ESP_028269_1755. NASA/JPL-Caltech/UA.
Since the landing, HiRISE has imaged the landing site regularly while monitoring the rover traverse, seeing the parachute blowing around over time. Post-landing HiRISE images of landing hardware are listed in Table 2.3.
Figure 2.36. Detail view of MSL landing hardware on the surface on sol 1. All scale bars are 20 meters long. Upper left: descent stage impact site. Upper right: rover. Lower left: backshell and parachute. Lower right: heat shield. HiRISE image ESP_028269_1755. NASA/JPL-Caltech/UA.
Table 2.3. HiRISE images of landing hardware. Emission angle is a measure of how directly overhead the orbiter was at the moment the image was taken. Lower emission angle is more directly overhead and produces a higher-resolution, less-distorted image. HiRISE images are grayscale except for a narrow color strip at the center; whether the hardware is visible in the grayscale only (“gray”) or color parts of the image is indicated.
Image
Date
Sol
Emission angle (degrees)
Descent stage
Backshell
Landing site
Heat shield
ESP_028269_1755
7 Aug 2012
1
45
gray
gray
gray
gray
ESP_028335_1755
12 Aug 2012
6
30
gray
gray
color
gray
ESP_028401_1755
17 Aug 2012
11
10
gray
gray
color
gray
ESP_028612_1755
2 Sep 2012
27
9
color
color
color
gray
ESP_028678_1755
8 Sep 2012
32
17
gray
gray
gray
gray
ESP_029746_1755
30 Nov 2012
113
3
gray
gray
–
–
ESP_029957_1755
16 Dec 2012
129
2
color
color
gray
–
ESP_030168_1755
2 Jan 2013
145
17
gray
gray
gray
color
ESP_030313_1755
13 Jan 2013
157
4
gray
gray
color
gray
ESP_034572_1755
11 Dec 2013
479
10
gray
gray
gray
gray
ESP_035350_1755
10 Feb 2014
538
8
gray
gray
gray
gray
ESP_036128_1755
11 Apr 2014
597
3
gray
gray
gray
–
ESP_037117_1755
27 Jun 2014
672
14
gray
gray
color
–
ESP_040269_1755
28 Feb 2015
911
2.6
gray
gray
gray
–
ESP_048774_1755
21 Dec 2016
1556
2.7
gray
gray
–
–
REFERENCES
Abilleira F (2013) 2011 Mars Science Laboratory trajectory reconstruction and performance from launch through landing. Paper presented to the 23rd AAS/AIAA Spaceflight Mechanics Meeting, 10–14 Feb 2013, Kauai, Hawaii, USA
Abilleira F and Shidner J (2012) Entry, descent, and landing communications for the 2011 Mars Science Laboratory. Paper presented to the AIAA Guidance, Navigation, and Control Conference, 13–16 Aug 2012, Minneapolis, Minnesota, USA
Baker R et al (2014) Mars Science Laboratory Descent-Stage Integrated Propulsion Subsystem: Development and flight performance. Journal of Spacecraft and Rockets 51:4, DOI: 10.2514/1.A32788
Beck R et al (2010) The evolution of the Mars Science Laboratory heatshield (part III). Presentation to the 7th International Planetary Probe Workshop, Barcelona, Spain, 16 Jun 2010.
Bhandari P et al (2011) Mars Science Laboratory Launch Pad Thermal Control. Paper presented to the 41st International Conference on Environmental Systems, 17–21 Jul 2011, Portland, Oregon, USA
Bose D et al (2013) Initial assessment of Mars Science Laboratory heatshield instrumentation and flight data. Paper presented to the 51st AIAA Aerospace Sciences Meeting, 7–10 Jan 2013, Grapevine, Texas, USA, DOI: 10.2514/6.2013-908, DOI: 10.2514/6.2013-908
Bose D et al (2014) Reconstruction of
aerothermal environment and heat shield response of Mars Science Laboratory. Journal of Spacecraft and Rockets 51:4, DOI: 10.2514/1.A32783
Chang K (2012) Simulated Space ‘Terror’ Offers NASA an Online Following. The New York Times 11 Jul 2012, p. A14
Chen A et al (2014) Reconstruction of atmospheric properties from Mars Science Laboratory entry, descent, and landing. Journal of Spacecraft and Rockets 51:4, DOI: 10.2514/1.A32708
Chen C and Pollard B (2014) Radar terminal descent sensor performance during Mars Science Laboratory landing. Journal of Spacecraft and Rockets 51:4, DOI: 10.2514/1.A32641
Cruz J et al (2014) Reconstruction of the Mars Science Laboratory Parachute Performance. Journal of Spacecraft and Rockets 51:4, DOI: 10.2514/1.A32816
Edquist K et al (2009) Aerothermodynamic design of the Mars Science Laboratory heatshield. Paper presented to the 41st AIAA Thermophysics Conference, 22–25 Jun 2009, San Antonio, Texas, USA, DOI: 10.2514/6.2009-4075
Edquist K et al (2009) Aerothermodynamic design of the Mars Science Laboratory backshell and parachute cone. Paper presented to the 41st AIAA Thermophysics Conference, 22–25 Jun 2009, San Antonio, Texas, USA, DOI: 10.2514/6.2009-4078
Gallon J (2012) Verification and validation testing of the Bridle and Umbilical Device for Mars Science Laboratory. Paper presented to the 2012 IEEE Aerospace conference, 3–10 Mar 2012, Big Sky, Montana, USA, DOI: 10.1109/AERO.2012.6187289
Hoffman P et al (2007) Preliminary design of the Cruise, Entry, Descent, and Landing Mechanical Subsystem for MSL. Paper presented at the 2007 IEEE Aerospace Conference, 3–10 Mar 2007, Big Sky, Montana, USA, DOI: 10.1109/AERO.2007.352826
The Design and Engineering of Curiosity Page 12