JPL was handed the task of engineering the rovers, landers and EDL systems, as well as planning the mission operations. Pete Theisinger, a JPL veteran whose experience went back to the Mariner program of 1967, would be the project manager. He assembled a team of the finest available JPL engineers to work on the MER program. Squyers would work with Theisinger's engineers to realize the Athena scientific payload that would go on the rovers. The rovers would be larger than Sojourner and bear little resemblance to the microrover, but it was believed that the Pathfinder EDL system and lander could be used with little modification, though the mass of the eventual rover would prove otherwise. The problems that the MER program would encounter could not be foreseen, but no additional time could be given to the program to resolve them. The rovers had to be launched in a given window of time only several weeks long, or they would have to wait another twenty-six months before trying again - an unlikely possibility. Most daunting was the fact that three years was an alarmingly short time span to engineer, build, test, validate, deliver and launch two sophisticated and complex spacecraft on two different missions.
"The primary resource constraint on MER was time,'' Mishkin recalled. "Basically, the rule is that the planets don't wait. You can only launch to Mars at certain times. The MER mission started up with less time than we would have liked. There was a transition from Pathfinder to MER, actually from a relatively small team to a much larger team, but even so, the team had to work even harder than the Pathfinder team had due to the schedule constraints. The resources were there but the time pressure was extreme and it did require many people to work sustained periods of many long hours.''
In the face of this challenge, JPL did have some things working in its favor regarding the MER. One of them was the Athena payload. This scientific payload had actually been developed for an earlier Mars mission, Mars Surveyor 2001. As a package for the Surveyor '01 rover mission, it included an Alpha Particle X-Ray
Spectrometer (APXS), a Microscopic Imager, a Mini-Corer, a Mini-Thermal Emission Spectrometer (Mini-TES) and a Mossbauer Spectrometer. The rover was to be outfitted with Panoramic Cameras (Pancam) mounted on a mast at the front. All these technologies had matured to the hardware stage before the Surveyor 2001 program was cancelled, but this development maturity of the Athena payload was one reason NASA had selected Squyers' proposal, and went a long way toward making the Mars Exploration Rover missions possible. For the new package, the Mini-Corer would be replaced by a Rock Abrasion Tool mounted on a robotic arm. Also mounted on the robotic arm would be the Microscopic Imager, the Alpha Particle X-ray Spectrometer and the Mossbauer Spectrometer.
The MER rover program had something else working in its favor. Since 1999, JPL had been conducting tests in the Mojave Desert with the Field Integrated Design and Operations testbed, known as FIDO. This vehicle was originally designed and built to test the systems needed in a rover as part of the Mars Sample Return mission. Even as a testbed, FIDO's capabilities were quite advanced.
"FIDO's advanced technology includes the ability to navigate over distances on its own and avoid natural obstacles without receiving directions from a controller,'' Dr. Eric Baumgartner stated in an April 1999 JPL press release. "The rover also uses a robot arm to manipulate science instruments and it has a new mini-corer or drill to extract and cache rock samples. There are also several camera systems onboard that allow the rover to collect science and navigation images by remote-control.''
"FIDO is about six times the size of Sojourner and is far more capable of performing jobs without frequent human help,'' Dr. Paul S. Schnenker, who was in charge of FIDO's development at JPL, stated in the press release. "FIDO navigates continuously using on-board computer vision and autonomous controls, and has similar capabilities for eye-to-hand coordination of its robotic science arm and mast. The rover has six wheels that are all independently steered and can drive forward or backward, allowing FIDO to turn or back up with the use of its rear-mounted cameras.''
When the Mars Sample Return mission was cancelled, FIDO was called upon to contribute to the development of the Mars Exploration Rovers. Many of the technologies developed for FIDO would find their way into the MER rovers, but there were still critical issues of size and weight. There was a MER preliminary design review in October 2000, and it was clear that the new rover could not fit on the lander as designed for Pathfinder. The lander would have to get bigger, even as the design for the rover was evolving. The main elements of the rover's evolving design, apart from the Athena scientific package, were the rover's Warm Electronics Box, the wheels and rocker-bogie suspension, and the all-important solar arrays.
The solar arrays were a critical item, because the rover needed all the electrical power it could get. The larger the arrays, the more power the rover would have. However, there were limits to the size of the arrays, both to fit inside the lander's tri-pyramid envelope and to keep the rover's weight down as much as possible. Mechanical Lead for the Mars Exploration Rover was Randy Lindemann and of all the systems he was responsible for on the MER, the solar arrays would be the most problematic - in terms of struggling to find a suitable design that provided the needed power while fitting into such a compact envelope. Lindemann and his engineers spent months working out the packaging problem of laying out the photovoltaic cells in groups called strings, arranging them on the rover's panels, and experimenting with different designs for the solar array panels to unfold from the rover. The finished design met the rover's power requirements and packaging constraints and the opened panels, measuring a total of 1.3 square meters, gave the MER rover a swept-wing look.
The rover's core structure employed composite honeycomb panels with the interior insulated with silica aerogel, the same amazing material used inside Sojourner. The Warm Electronics Box would keep the rover's vital electrical and electronic components warm through the use of eight radioisotope heater units, electrical heaters and heat given off by the components themselves. The rover's computer would employ a Rad 6000 32-bit micro-processor, a radiation-hardened version of the commercial product. The Rad 6000 was capable of twenty million instructions per second, with the rover's computer running the VxWorks operating system. The computer had 128 megabytes of memory (RAM), but was backed up by 256 megabytes of flash memory.
The two cognizant engineers responsible for The MER rovers' mobility system were Chris Voorhees and Brian Harrington. This system included the wheels, drive
and steering actuators and rocker-bogie suspension. The sophistication and capability of the MER rover mobility system was as challenging to engineer as the solar panels. The components were designed in concert with each other and with elements of the lander (and even the airbag system), and they evolved as the rest of the rover and lander evolved. The wheels were an example of this evolutionary process.
"We started with the Sojourner wheels as a base to work from," Voorhees said. "Because of many different engineering demands on the wheels, the wheels for our new rovers didn't mature until late in the game. A big challenge was to be able to get enough traction to get through soil and over rocks but also to be benign enough to get off the lander without getting entangled in the deflated airbags.''
Like virtually every other part in the MER rover, the complex wheel was designed with 3-D modeling software and evaluated with finite element analysis software in order to make the wheel as light as possible while being strong and flexible. The wheel's hub featured spiral flexures which gave the wheel necessary shock-absorbing capability. The flexures were filled with Solimide, an open-cell foam that retained its flexibility even in the temperatures encountered on Mars. This was done to protect the drive and steering actuators inside the wheel. The twenty-six-centimeter-diameter wheels were curved along their entire circumferential surface to maintain uniform contact with the Martian soil throughout its total steering geometry. Each wheel was not an assembly, but was machined entirely from a single, solid billet of aluminum. The wheel design went through numerous iterations that changed as a result of testing and subsequent improvement.
The rocker-bogie suspension system was designed so that when one side of the rover's suspension moved upward upon reaching a large rock or obstruction, the suspension on the other side of the rover moved downward to keep the rover level. The two front wheels, drives and steering actuators were designed to pivot toward the center of the rover and stow, while the middle pair and rear pair of wheels were retractable to permit the rover to crouch on the lander for the lowest possible height and most compact envelope. The front and rear wheels also all had individual steering actuators and all six wheels had individual drives.
Rising above the rover's equipment deck was the Pancam Mast Assembly (PMA). This articulated mast supported both the Pancam and the Mini-TES. The Pancam included two digital cameras mounted within the camera bar, using Charged Coupled Devices (CCDs), with each camera having a small multi-spectral filter wheel. The Pancam would be capable of providing multi-spectral, stereoscopic and panoramic images, while the Mini-TES would provide remote sensing of soil and rock mineralogy. The optics of each camera used three-element symmetrical lenses, spaced thirty centimeters apart and 1.5 meters above the planet's surface. The optics and filters were protected from the Martian environment by a sapphire window in front of the optics barrel. The camera bar could pivot plus or minus ninety degrees in elevation, and the PMA could rotate 180 degrees in either direction for 360 degrees of coverage. The rover's navigation cameras (Navcams) were also mounted on the camera bar. The PMA would be erected from the stowed horizontal position to its vertical operating position by a deployment actuator at its base.
On the rover equipment deck behind the PMA were the High-Gain Antenna, the Low-Gain Antenna and the UHF Antenna. Both the High- and Low-Gain Antennas were capable of sending and receiving data utilizing the Deep Space Network on Earth. The High-Gain Antenna had the advantage of being able to precisely point to antennas on Earth and send data at much higher transmission rates. Through the UHF Antenna, the rovers could also uplink and downlink information initially with the orbiting satellite Mars Odyssey and, eventually, the Mars Global Surveyor. This approach was preferable because data could be transmitted to and from the orbiters faster, and the orbiters could remain in communication with Earth longer than the rovers.
The lander and EDL systems had size and weight issues of their own. As originally proposed by engineers at JPL in April 2000, the "build-to-print" concept was to employ the Mars Pathfinder lander and EDL system for the Mars Exploration Rover, with perhaps fifty per cent of the launched dry mass inheriting detail design from the Pathfinder program. It didn't turn out that way. As the rover grew, the lander grew and virtually every component in the EDL system increased in size and weight. The cruise stage actually ended up weighing less than its initial mass estimate. The first tests of the MER airbag system with a dummy lander and rover were performed in the United States' largest vacuum chamber at NASA's Plum Brook facility at the Glenn Research Center in Sandusky, Ohio. The airbag system was made up of four, six-lobed tetrahedral airbag segments, each one mated to an exterior lander surface. The airbags used an abrasion layer to prevent tearing of the inflated bags from inconveniently placed rocks on the sloped surface it would impact. This design had worked well with Mars Pathfinder, but the first test was a catastrophic failure, ripping through the abrasion layer and puncturing the inner bladder of the bags, surprising the test engineers.
JPL worked with the manufacturer of the airbag system, ILC Dover of Frederica, Delaware, over the next two-and-a-half years to redesign and strengthen the MER airbag system and confirmed the effectiveness of the changes with more tests (including deflation and retraction under the lander and its petals). JPL engineers also discovered that the greatest likelihood of airbag tearing took place when a lobe struck one of the sharp test rocks at an oblique angle when simulating being carried along by Martian wind during impact. Some method had to be developed to eliminate the horizontal velocity of the spacecraft with inflated airbags during the final seconds of descent, and JPL engineers devised a Descent Image Motion Estimation Subsystem (DIMES) to work with the Traverse Impulse Rocket Subsystem (TIRS) that would eliminate lateral movement during firing of the retro-rocket system. This would permit the lander to inflate its airbag system, cut the bridle and drop to the planet's surface and bounce more than fifteen times before coming to a stop.
The EDL parachute system development went through similar heart-stopping moments for the JPL team. Mars Pathfinder had employed a disc-gap-band parachute that featured a hemispherical main segment with an air gap and a band of parachute material designed to add stability during descent. The engineers knew the original parachute was too small to do the job, so the size of the parachute was increased. However, there were limits to what the size the parachute could be in order to fit into its cylindrical storage canister aboard the lander. The first parachute test took place at the National Guard Gunnery Range in Boise, Idaho in April 2002. To replicate the shock to the parachute system opening in the Martian atmosphere while traveling at hundreds of kilometers per hour, the weight of the dummy payload had to be increased to 3,600 kilograms. The first test resulted in a torn canopy and shredded band. A second test the following day produced the same result. With launch day now just over a year away, three new parachute designs were developed and were tested several months later at NASA's Ames Research Center wind tunnel. The first test parachute exhibited "squidding" where the chute failed to open completely. The chute was redesigned, then redesigned again before tests were finally successful.
In August 2002, JPL conducted new field tests with a FIDO that had received modifications to permit command, control and communications as planned for the MER missions. JPL issued a news release that described the scope and success of the field tests.
"The test rover has received and executed daily commands via satellite communications between JPL and the remote desert field site,'' Dr. Ed Tunstel stated in the news release. "Each day, they have sent images and science data to JPL that reveal properties of the desert geology.''
The pictures taken by the MER rovers Spirit and Opportunity were digitally aligned to form a nearly seamless mosaic final image. This partial panoramic image was taken by Opportunity's navigation camera on sol 34. It shows the MER's lander in the center of a crater on Meridiani Planum. Note the circular impressions on the walls and floor of the craft from the inflated airbags. ((NASA/JPL)
The pieces of the MER puzzle were fitting into place. As with every program of such complexity, however, problems continued to arise and continued to be solved almost up to the time the spacecraft were on the launch pad in Florida. This critical phase was known as Assembly, Test and Launch Operations (ATLO). The Mars Exploration Rovers, landers, and cruise stages were assembled in JPL's famed Spacecraft Assembly Laboratory. The spacecraft's aeroshell, including the backshell and heat shield, would be supplied by Lockheed Martin Astronautics Company of Denver, Colorado. ATLO had been underway since 25 February 2002, and was under tremendous schedule pressure. Many JPL personnel secretly doubted that the spacecraft would get to Florida in time for the launch window in June 2003. To succeed sometimes required a twenty-four hour, seven-day-a-week multi-shift work schedule. The individual rovers were not yet named, instead being referred to as MER-A and MER-B. The first rover to complete its final tests was MER-B, and it left the Jet Propulsion Laboratory in the early morning hours of 22 February 2003, in a special cargo truck accompanied by other convoy vehicles, headed for Cape Canaveral Air Force Station in Florida. MER-A followed several weeks later.
The MER rovers were delivered to the Payload Hazardous Servicing Facility at the Cape. There, they would spend the remaining months of ATLO undergoing further testing, integration with their landers, more testing, closing up each lander and securing them inside their aeroshells, and assembling the aeroshells to the cruise stages, before finally mounting the payload to the solid rocket motor third stage and then securing them inside the payload fairing of the Delta II rocket. Problems continued to plague the rovers and other equipment up to the final days before launch, but they were all overcome thanks to overwhelming dedication by all the members of the team.
NASA had held a contest to name the rovers, and the winning entry came from nine-year-old Sofi Collis from Arizona with a poem that included Opportunity and Spirit. On 10 June 2003, Spirit (MER-A) was launched at 17:59 UTC destined for Gusev Crater, named after the Russian astronomer Matvei Gusev. Located fifteen degrees south of Mars' equator, Gusev Crater was a shallow impact crater measuring roughly 150 kilometers in diameter. Opportunity's landing site was the Meridiani Planum halfway around the planet from Gusev. The MER team had succeeded in reaching the goal of launching the spacecraft on time, but at great personal sacrifice. Between the summer of 2002 and the launch of both rovers in mid-2003, the JPL team had been given only one two-day weekend off.
Was this article helpful?