Curiosity

by Rod Pyle

  A man named Tony Spear was Pathfinder's project manager, and Matt Golombek was the principal investigator. Spear assembled a small team of young and energetic engineers and scientists within the lab to design and build the spacecraft in a skunkworks-type environment. The overall budget, including flight and operations, was to be capped at under $300 million. When this is compared to the design-and-build price tag of half that, it shows you just how expensive it is to launch and operate these missions—in this case, it doubled the cost.

  Pathfinder would be comprised of two components: a lander and a rover. The lander, looking like a small, metal pyramid, would set down on Mars. Then the sides would drop, and a tiny rover, no larger than a microwave oven, would drive off and explore the locale. They named it Sojourner. It would only range about thirty feet from the base station and had to use the lander's radio to talk to Earth, but this was a groundbreaking mission, and getting anything onto the surface of Mars would be a major accomplishment after two decades of no Mars missions. The team took to the task with passion.

  Many years later, I sat with Rob Manning, who had been a pivotal character in the Pathfinder saga. When we talk Mars, there is a gleam in his eye and he gets so excited about space exploration that he can barely get the words out—but he does exclaim that this or that bit of technology “is just fantastic!” quite a lot. I sympathize—I don't possess half the knowledge he does, but I too get impassioned about the subject, and it's something I've been taught by long public exposure to limit in myself—some people find it slightly frightening. But with Rob, it's a charming combination, as witnessed by the number of JPL people who stopped by our table as we dined. We had visited a favorite eatery not far from the lab, and at least five groups of people wandered over to check on something project oriented, ask him a question, or just say hello. Each was greeted with exuberance. It's as if he is permanently just out of Sunday school, in an exhilarating way.

  Fig. 6.1. “JUST FANTASTIC!”: Rob Manning was the chief engineer of MSL until he moved to his current assignment as chief engineer of a new, high-speed EDL (entry, descent, and landing) research program for future Mars missions. Rob is the kind of guy who gets other people excited about Mars. He does frequent speaking events, and his enthusiasm is incredibly contagious. He led the design team for Pathfinder and has worked on every Mars rover mission to date. Of Pathfinder's bouncing arrival, he remembers that it had been “tough to pass the laugh test.” Image from NASA/JPL-Caltech.

  I took him back to the beginning phases of Pathfinder in the early 1990s to learn more about the machine that would ultimately provide many core technologies for Curiosity as well as the Mars Exploration Rovers, Spirit and Opportunity. As it turned out, it was the fastest and by far the simplest mission he ever worked on…partly because NASA's expectations were low.

  “The great thing about Mars Pathfinder is that we had a single page of requirements, the government just said ‘land on Mars during the 1996 launch opportunity, deliver the rover and send back some pictures. Good luck! Here's some money to do some science if you have the time.’ This kind of mission design had really not been heard of since the 1960s. It was delightfully freeing.” The memory brings a smile to his face.

  “Basically the mission model was to prove that NASA can do things cheaply, efficiently, and effectively. Also to demonstrate that there are efficiencies that we can achieve using the faster, better, cheaper approach which had not been really invented yet. In fact, nobody had ever really agreed what that meant….” He chuckles.

  The first decision was to have a team working together, all in one spot, at JPL. “We thought that was very important. We couldn't completely achieve that because there were space [as in office space] problems at the lab at the time. But we got the majority of the core team members all in one spot.” This allowed for close coordination, rapid communication, and a delightful lack of paperwork. “There really was very little documentation,” he says with a smirk, “there just wasn't time!”

  Perhaps the largest challenge, as it would be with Curiosity, was how to deliver the spacecraft to Mars. Viking, with its vast funding, had carried along large tanks of rocket fuel not just for the lander but also to allow it to enter orbit around Mars. When it arrived in the neighborhood, it fired powerful braking thrusters to slow it enough to be captured by Mars's gravity. The twins orbited the planet for weeks, while at JPL scientists fretted over how nasty the landing zones looked in the improved pictures. But they picked a spot remarkably close to the original one, and landed, falling out of the sky at mere orbital speeds.

  Pathfinder was an entirely different animal. It was smaller and far lighter, true, but with their tiny budget (about one-twentieth Viking's when inflation is accounted for), they would be unable to pause in Martian orbit—they simply did not have enough rocket power to stop there. So they targeted the probe to fly directly into Mars's atmosphere. It was a lot like shooting an enormous hunting rifle at where Mars should be in eight months—with no allowance for error.

  This meant that Pathfinder would slam into the Martian air at interplanetary-transit speeds of about 12,000 mph, way faster than Mars orbital speeds. It also meant that the heat shield needed to work harder, and they had more energy to get rid of on the way down. Otherwise—splat, no more Pathfinder.

  To accomplish this, they chose…beach balls.

  Rob explains the original idea: “When we hit the ground, you don't want to store energy and rebound. So the original idea was to have vented airbags like the ones in cars that would deflate right away, so all this energy is dissipated and is more like a sandbag.” In short, once the lander had slowed to manageable speeds, they would inflate airbags—looking very much like a collection of three-foot beach balls—to absorb the impact of landing, and they would immediately exhaust the air out the sides upon impact. He continued: “But then we realized we couldn't get the vents to work quickly enough…so we all agreed that we would be bouncing instead.” They would scrub off the remaining landing energy—its speed—by bouncing until the spacecraft rolled to a stop. Then the airbags would be deflated.

  While the project generally ran somewhat under the radar, they did have to submit to the usual NASA design review. This is where the best minds on the project develop their sales pitch, then meet with a group of engineers, scientists, and others from the field to sell their big idea. On the panel for Pathfinder were some Viking people (that was good), an assortment of current NASA folks, and one somewhat-curmudgeonly Cadwell Johnson (maybe not so good).

  Johnson was an interesting choice. While he was nearly a god in the aerospace community, he did not even have a college degree. He had earned his stripes before the modern era of NASA where it is generally recognized that a master's degree is a point of entry, and PhDs are common. But when Johnson came into the space program, at the dawn of NASA in the 1950s, they just needed his pencil. He was one of the best and brightest draftsmen that Max Faget, designer of America's manned spacecraft from Mercury through the space shuttle, had ever met. Johnson was hired on the spot and became a legend of the space race. And now he was staring down a somewhat-nervous JPL team.

  “So here we were are, having this review, and Cadwell is listening impatiently. Richard Cook, the flight-operations engineer, was talking and about mission design and defending the airbag architecture.” Manning smiles at a memory that brought little joy at the time. “Cadwell says, ‘Listen buster, don't tell me how to land on another planet! This is a stupid idea that's never gonna get off the ground!’ He was very negative. And the Viking people were rolling their eyes. They're rolling their eyes and they are saying ‘You're gonna do what?’” He laughs his infectious laugh. “At that point, we had told them how high we were bouncing—we're talking fifty to seventy-five feet above the surface of Mars. We were testing it to be able to bounce one hundred feet. They just thought we were nuts.”

  As we all know, it turned out to be a sound design, but the review was a confidence rattler. Nonetheless, it wa
s eventually approved and they moved forward. The lander and rover were designed and built at JPL by a small and dedicated crew of engineers and technicians. The parachute was a headache, though—all parachutes are headaches when used to land on Mars because fabric does not behave like metal or plastics. “It kind of does its own thing,” as Manning says. But the real challenge remained the airbags. Now that they had sold the landing system, they had to make it work.

  “NASA had said to us to keep it simple. Don't overdo it, don't overspend….” In short, stay within the boundaries of the Discovery Program. “Keep it simple—that was our mantra.” Actually, there was another, even more urgent imperative—don't screw up. “The other mantra was to test, test, test,” Manning recalls. “We did not skimp on testing. Sometimes the test articles were low budget, but we didn't skimp on test itself. We had to convince them, and ourselves, that we knew how to get [this landing system] to work.”

  Testing would be a key ingredient in the success of all the rovers, from Sojourner to Curiosity. This remains the only way to run a successful program.

  There is a lot of jargon in aerospace engineering, and one part of it is around the readiness and flightworthiness of a given design or system. It's called the DRL for Design Readiness Level. The design is assigned a number to indicate its reliability ranging from 1 to 9. “I would say that airbags were probably at a 3,” Manning continues. “They're supposed to be a 6. We did a lot of things for the first time that were later adopted by other projects, technologically speaking. But we weren't stopped from being innovative, because it was considered part of our objectives.”

  They tested and they tested. The bags ripped and tore. They inflated late and early. The engineers rolled them down inclines at JPL and dropped them on the desert floor from a helicopter. It was all somewhat ad hoc, though thoroughly professional. Just…small. But over time they wrestled the airbag system into compliance.

  The rover itself was an innovative design. The small box was topped with a solar panel that provided just enough current to allow the machine to work in various lighting conditions. It was mounted flat atop the little machine, however, and one good coating of the ubiquitous Martian dust would reduce its efficiency something fierce. But it was the only way it would fit. Sojourner sported six wheels, each about four inches in diameter with little aluminum spikes jutting out of them for traction. The wheels were mounted to an ingenious suspension system called “rocker-bogie,” a jointed set of levers on each side that allowed maximum ground contact when driving over obstacles. This suspension system worked so well that it has been used on all Mars rovers since.

  Of course, the lander and rover themselves also had to be tested. While the lander was fairly straightforward, with its landing bags and unfolding sides, the rover needed to be driven. And on something approximating Mars. And…on the cheap.

  “In the center of the JPL campus is a building called the SFOF—the Space Flight Operations Facility. Upstairs is one giant room that extends the length of the whole building, from corner to corner. That giant room was committed to Mars Pathfinder, so I had sandboxes installed in there.” There is a playful, possibly mischievous glint in Manning's eye as he relates the story. “We had to do a lot of arm-twisting to let me bring in thousands of pounds of sand for the test area. In the end, I bought playground sand from Monrovia [a sleepy suburb fifteen minutes east of Pasadena, where things tend to be less expensive]. We washed it first so it wouldn't be too dusty, but we still had to seal off the room to keep the sand inside. So we had our own little sandbox up there as well as the electronics, the lander, and airbags. All in one place where the team could work together. It was great fun.” I can just see the administrators gleefully enjoying the idea of a few tons of sand next to their überexpensive operations center…. Not.

  And in one more trip to Monrovia he visited a Home Depot store and bought a leaf blower. “I had to use that to inflate the airbags for the early tests.” He laughs.

  The lander and rover went through tough testing at the lab—mostly successfully. This included thermal testing: the little machines were roasted, then frozen repeatedly in JPL's thermal chamber. They were shaken to make sure that they could survive the savage rocket ride into orbit. Then they were subjected to a thorough cleaning to avoid contaminating Mars with earthly bacteria. It was a grueling process, and all done on a tight budget.

  The entire assembly left Earth on December 4, 1996, aboard a Delta II rocket—the only US rocket powerful enough to do the job on their budget. Pathfinder transited the cold dark between Earth and Mars for seven months, then slammed into the Martian atmosphere on July 4, 1997. The heat shield, the first US design to be subjected to an atmosphere other than Earth's since Viking's twenty-one years earlier, held. The parachute deployed and didn't tear. The braking rockets, a simple, solid-fueled design not unlike skyrockets, slowed the lander at the last minute as intended. And then came the moment of truth—at the proper moment another small rocket engine fired, inflating the seventeen-foot-high pyramid of Vectran airbags ten seconds before Pathfinder reached the surface….

  And the assembly bounced, and bounced. Then it bounced some more. After hitting the ground at 45 mph, faster than any spacecraft had ever encountered a planetary surface and survived, it bounced at least fifteen times in all. Then it rolled to a stop. It was 3:00 a.m. local Martian time.

  After time for checkouts, preparations, and a bit of celebration in mission control, the triangular pedals lowered from the base of the landing stage. As they did so, they righted the machine to the proper orientation. There it sat for the remainder of its first cold night on Mars, marking the Fourth of July holiday and the twenty-first anniversary of the landing of Viking 1. America's first roving machine was on Mars.

  Manning recalls landing day: “I was the flight director during the landing, which means that I knew when everything was supposed to happen. We had a transmission delay of about eleven minutes [due to the distance between Earth and Mars], so by the time we heard anything, it had either worked or it had not. Of course, we had practiced this extensively with our simulation setup [next to the sandbox]. We put a simulated eleven-minute delay into the signals between what was going on in the test area and the control room. So even though we could talk from one room to another in seconds, it was like adding 125 million miles of distance. We had practiced so much that when we actually landed, it felt fake!”

  The next day, the little rover detached itself from its restraints and, slowly and daintily, crawled down the ramp. It managed to avoid snagging on the deflated airbags, which was another area of critical concern, and rolled the first two of six wheels onto Mars.

  Pathfinder had landed in an area called Ares Vallis, a region chosen using twenty-year-old Viking orbital imagery. MER's and Curiosity's landing sites would be picked with far more precision due to the ultra-high-resolution orbital imaging that came later. Luckily, Ares Vallis turned out to be a reasonably good combination of a nice, flat, and safe surface that was still interesting enough to titillate the geologists. The region had been the recipient of vast water flows in the distant past, promising a fine variety of rocks and soils to examine.

  The Sojourner rover spent almost three months driving around its small backyard. It never ranged much beyond thirty feet from the lander, as its radio transmitter had a limited range. In addition, they wanted to keep an eye on the little rover from the lander's cameras. Its first target, a rock affectionately dubbed “Barnacle Bill,” was a scant fifteen inches from where the rover first drove onto the planet's surface. Many more specimens would follow, with such names as Yogi and Wedge. There was even a nice little collection of targets they called Rock Garden.

  While this was occurring, the lander took its own photos, the first color panoramics of Mars since 1976. The images were stunning.

  Everything seemed to be going swimmingly for the first six weeks, and then something happened that would become a regular occurrence as modern computers (albeit creaky ones by
today's standards) met the red planet. The lander decided to attempt a reboot. It didn't ask, nor did it warn anyone. It just decided to. It took a day to fix the problem. This is one downside of modern, reprogrammable flight computers—they have occasional hiccups. Subsequent rovers have suffered similar challenges.

  Upon regaining communications, the controllers noticed that the rover was not moving. It had stopped, jammed up against the aforementioned rock named Wedge. The tilt sensor had told it to stop before it hurt itself, and there it sat, awaiting instructions from home. After an intense conference, they radioed instructions Marsward and the machine freed itself, continuing its slow trek. Progress was measured in inches or a few feet per day.

  Sojourner investigated more rocks with its instruments, basic versions of the much more sophisticated units that flew on MER and Curiosity. Its APXS—for Alpha Proton X-ray Spectrometer—visited many rock targets, sniffing out their composition with its limited sensing power. It could take days to evaluate one rock, but they were learning all the while. The rover's three little cameras, two on the front and one on the back, guided it and provided ground-truth images that were relayed to the lander and sent home along with the lander's own images, tracking the rover's progress. They tried out various experiments with the drive system—locking one side and turning with the other, or locking five wheels and allowing the sixth the dig a trench to observe the soil underneath. The lessons learned would guide rover design for a decade.