Launching Science: Science Opportunities Provided by NASA's Constellation System (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4 Human and Robotic Servicing of Future Space Science Missions The capability of the Orion spacecraft to travel beyond low Earth orbit together with recent developments in robotics creates new opportunities to service future large spacecraft, primarily observatories. The ability to service spacecraft enables scientists to repair missions and to add science capabilities to existing missions, even changing their scientific focus; to address new questions and changing scientific emphases; to avoid the 15- to 20-year wait- ing period for new missions; and to take advantage of advances in technology. Thus, telescopes would no longer be obsolete by the time they were launched, and making adjustments to existing missions would be cheaper than launching a new mission to replace outdated science and technology capabilities. The current NASA administrator has offered a general endorsement of making future observatories serviceable. When NASA first began operating the space shuttle, the agency had ambitious plans to use it to service numer- ous science spacecraft in orbit. The most notable science spacecraft regularly serviced by humans has been the Hubble Space Telescope (HST). Several Hubble servicing missions resulted in a much more capable instrument than was originally launched. Hubble serves as a testament to the potential to upgrade and improve scientific instru- ments placed in space. As science spacecraft, particularly observatories, are placed beyond low Earth orbit, new methods will be required to service them. The Constellation System has potential use for spacecraft servicing. Despite the fact that NASA currently conducts regular construction and repair of the International Space Sta- tion (ISS) and at the time of the writing of this report was preparing for a fifth and final servicing mission to the Hubble Space Telescope, the concept of regular servicing of spacecraft has not been a part of NASA’s future plan- ning for numerous reasons. Both the Constellation System and new robotic capabilities can change that situation. However, the agency needs to take these capabilities into account and not simply to assume that they will develop automatically. This means not only actively considering the future servicing of spacecraft, but also considering new and novel approaches to servicing, such as the creation of way stations, or “servicing nodes,” in space. Designing spacecraft for human servicing increases cost. However, with careful planning, it should be pos- sible to develop methods to enable the most likely servicing without adversely increasing costs. Because robotic servicing is only a recent development, it is unclear how designing spacecraft for robotic servicing would affect costs. The committee has determined that further study is required.   Orion is limited in its capabilities for servicing spacecraft. For example, it lacks an airlock and has insufficient extra payload capability for carrying equipment and replacement instruments to a servicing location. In response to its request for information, the committee received a proposal for the creation of a servicing node for future observatories. The committee also received briefings about the opportunities of servic- ing for several possible telescope designs. 83 

84 LAUNCHING SCIENCE Recommendation:  NASA should study the benefits of designing spacecraft intended to operate around Earth or the Moon, or at the libration points for human and robotic servicing. HUMAN SERVICING OF SPACECRAFT NASA first began developing extravehicular activity (EVA) experience with the Gemini program in the mid-1960s. The early goals for these EVAs were to gain basic understanding of human operations in space and to develop better spacesuit designs for the purpose of eventually conducting EVAs on the lunar surface and for simple maneuvering and maintenance tasks in space. By the latter part of the Apollo program, astronauts conducted EVAs during the return from the Moon in order to retrieve film canisters from the Apollo Service Module’s Scientific Instrument Module. Although the idea of having humans repair and upgrade spacecraft in orbit had existed for a long time (and in fact were a staple of 1950s science fiction movies), it was not until the first Skylab mission in 1973 that the use- fulness of having humans perform spacewalks to repair a spacecraft was actually demonstrated. Skylab, launched in May 1973, sustained severe damage during its ascent, including damage to its micrometeoroid/sun shade and one of its solar panels. The launch of the first piloted mission to Skylab was delayed in order to develop repair techniques for the ISS. The astronauts were eventually launched into orbit with a special repair kit. During a series of difficult and largely unrehearsed EVAs, they successfully made substantial repairs, including the deployment of a parasol sunshade that cooled interior temperatures enough so that Skylab was habitable. Subsequent Skylab missions installed the twin-pole sunshade, repaired a malfunctioning antenna, and replaced film in the solar obser- vatory. The Skylab experience successfully demonstrated that humans could conduct complex, challenging, and unplanned repairs to spacecraft in low Earth orbit. From relatively early in its development, the space shuttle was designed to be capable of rendezvous with and capture of orbiting spacecraft, and the possibility of using shuttle crews to repair and upgrade spacecraft evolved out of these plans. The shuttle design evolved to include a robotic arm (the Robotic Manipulator System, or RMS, more commonly referred to as the Canadarm) and also equipment inside the payload bay to secure and support spacecraft as well as astronauts during EVA repair and maintenance missions. Many of the tools and techniques for servicing spacecraft were initially developed early in the shuttle program as an effort to demonstrate the shuttle’s ability to provide services that expendable launch vehicles could not. During the course of these operations, NASA learned a great deal about the development of EVA tools and techniques as well as about the importance of designing spacecraft for servicing. In November 1984, the space shuttle Discovery (STS-51A) performed the first retrieval and return to Earth of a satellite for repair and relaunch. Two astronauts wearing jet-propelled manned maneuvering units retrieved two malfunctioning satellites, Palapa-B2 and Westar-VI, which were in improper orbits due to kick motor malfunc- tion. A “stinger” was used to capture Palapa-B2 and Westar-VI. This mission was the last time that the Manned Maneuvering Unit, a backpack that allowed an astronaut to maneuver untethered from the shuttle, was used. It was also the last time that astronauts performed EVAs without a tethering system (besides the 1994 Simplified Aid for EVA Rescue [SAFER] test on STS-64). The ability to improvise has proven to be a key asset in servicing satellites, as demonstrated with the rescue of Palapa-B2 and also with the attempted servicing of Leased Satellite (LEASAT) 3. LEASAT 3 was deployed in April 1985, and it immediately became apparent that its engines were not firing. Engineers theorized that an unreleased hook on the outside of the spacecraft was preventing the engines from working properly and advised the astronauts to push the hook down. No spacewalks had been planned for the mission, and there was the danger that upon releasing the hook the engines would accidentally immediately fire and injure anyone nearby. Further complicating matters was the fact that the satellite was rotating and could not be easily grabbed by the shuttle’s RMS. Four days after its deployment, the astronauts tried to repair LEASAT 3 by building a “fly swatter” (Figure 4.1)—created on the shuttle using a plastic document cover attached to the shuttle’s RMS. An astronaut was then able to carefully extend the arm toward LEASAT 3 and snag the hook in the “fly swatter.” Unfortu-   See http://www.nasa.gov/mission_pages/shuttle/shuttlemissions/archives/sts-51A.html. 

HUMAN AND ROBOTIC SERVICING OF FUTURE SPACE SCIENCE MISSIONS 85 FIGURE 4.1  On space shuttle mission STS-51D, April 1985, an improvised “fly swatter” is used by the Canadarm to try to activate the Leased Satellite (LEASAT) 3 satellite. SOURCE: Courtesy of NASA. nately, snagging the hook at least twice did not fix the satellite’s problem, and engineers on the ground realized that something else must have been malfunctioning. The satellite was successfully repaired when it was rewired on a later shuttle mission. In May 1992, during the maiden flight of the space shuttle Endeavour during STS-49, NASA conducted the only three-person EVA in order to capture the INTELSAT VI communications satellite. The initial plan was for an astronaut to attach a “capture bar” to the satellite so that it could be grabbed with the RMS. However, after the attempt to attach the capture bar failed, the shuttle crew developed a backup plan. Mounted on footrests, the three astronauts simultaneously grabbed the satellite to stop its rotation (Figure 4.2). After capture, the satellite was fitted with a new perigee kick motor and subsequently released back into orbit, where the new motor successfully fired the spacecraft into a geosynchronous orbit. One of the primary lessons from these early years of spacecraft servicing was the requirement for spacecraft to be designed for capture. When the spacecraft did not have a grapple fixture, it was difficult and time-consuming for astronauts to capture the spacecraft and to move it to a location where it could be serviced. Before the space shuttle even began flying, NASA recognized that in addition to tools and techniques for servicing spacecraft, it would be easier to service spacecraft if they were designed for such servicing. This meant not only a grapple fixture, but also requirements that instruments and important systems be easily accessed by astronauts in their bulky spacesuits. However, NASA abandoned an effort to require that all future NASA Earth- orbiting spacecraft be modularly designed in order to facilitate astronaut servicing. One mission that was designed for servicing and benefited greatly from it was the Solar Maximum Mission 

86 LAUNCHING SCIENCE FIGURE 4.2  STS-49 crew members complete successful capture of the INTELSAT VI into Endeavour’s payload bay. Left to right, mission specialists Richard J. Hieb, Thomas D. Akers, and Pierre J. Thuot, positioned on the remote manipulator system, have handholds on the satellite and prepare to attach the capture bar (tethered to Hieb). SOURCE: Courtesy of NASA. (SMM), a joint endeavor of NASA Goddard Space Flight Center, the Jet Propulsion Laboratory, and NASA Marshall Space Flight Center, launched on February 14, 1980, to study the Sun during the period of greatest solar activity during the solar cycle (Figure 4.3). The SMM experienced a malfunction in 1981 that would have cut its mission short. However, the SMM had been fitted with a shuttle “grapple feature” in anticipation of unforeseen malfunctions, and this enabled a shuttle to use its robotic arm to capture the satellite for repair. On April 6, 1984, a servicing mission by space shuttle Challenger successfully repaired it. An astronaut using a Manned Maneuver- ing Unit approached the slowly spinning satellite and slowed its rotation, enabling the shuttle arm to capture it 

HUMAN AND ROBOTIC SERVICING OF FUTURE SPACE SCIENCE MISSIONS 87 FIGURE 4.3  Mission specialist George Nelson approaches the damaged Solar Maximum Mission in a capture attempt during the April 6, 1984, servicing mission. SOURCE: Courtesy of NASA. and place it in the payload bay where it could be repaired. This was the first capture with the arm of the shuttle, performed during the first on-orbit servicing mission in history. The Palapa, LEASAT, and SMM repair missions all demonstrated the ability of astronauts to adapt to unfore- seen challenges, and SMM demonstrated the value of designing a spacecraft for modularity and repair. Both of these attributes were employed extensively for the best-known examples of spacecraft repair and upgrade—the four completed Hubble Space Telescope servicing missions (plus one planned) conducted over a 15-year period between 1993 and 2008. (See Box 4.1.) Designing spacecraft for planned human servicing increases costs. Spacecraft have to be designed to include handholds and foot-restraint sockets. They have to provide access to all the interfaces. They must also eliminate potential hazards such as sharp edges and dangerous fuels such as hydrazine. Furthermore, ground simulators must be produced and procedures developed. (See Figure 4.1.3 in Box 4.1.) However, as the history of human servic- ing demonstrates, astronauts have been able to perform repairs to spacecraft and spacecraft systems that were not 

88 LAUNCHING SCIENCE BOX 4.1 The Hubble Experience There have been four servicing missions to the Hubble Space Telescope, each bringing new technologi- cal advances to the telescope that have enabled some of the best-known and most breathtaking science in the field of astronomy. After the Hubble was launched in 1990 with an improperly focused mirror, the first planned servicing mission assumed significantly greater importance than it had had before. Servicing Mission 1 (SM1, STS-61) in December 1993 (Figure 4.1.1) enabled the installa- tion of the Wide Field and Planetary Camera 2, Cor- rective Optics Space Telescope Axial Replacement (COSTAR), gyros, and solar arrays. The new scientific instruments led to the determination of the age of the universe, the observation of the birth and death of stars, and the observation of the formation of planets and stars. The Imaging Spectrograph, Near Infrared Camera, and fine-guidance sensor were installed dur- ing Servicing Mission 2 (SM2, STS-82) in February 1997. From these new scientific instruments came proof of black holes, the observation of Supernova 1As, the determination that the universe is accelerat- ing, observations of the effects of dark matter and dark energy, and the Hubble Deep Field image. In De- cember 1999 (SM3A, STS-103), the gyros, advanced computer, and fine-guidance sensor were installed to improve Hubble operations. Finally, in March 2002 (SM3B, STS-109), the Advanced Camera, solar arrays, power control unit, and Near Infrared Camera and Multi-Object Spec- trometer (NICMOS) were installed on Hubble (Figure 4.1.2). Since their installation, Hubble has seen into the Hubble ultradeep field, seen the evolution of galaxies, confirmed the dark energy principle, created the first direct mapping of dark matter, made the first detection of an organic molecule in the atmosphere of a Jupiter- like planet in the Milky Way Galaxy, and detected over 500 extremely old proto galaxies formed just after the big bang. FIGURE 4.1.1  After replacing the Hubble Space Tele- The Hubble servicing missions have been so scope’s solar panels, astronaut Kathy Thornton releases successful because of the methodical approach adopt- one of the old solar panels during the first Hubble Space ed by the engineers, astronauts, and project managers Telescope Servicing Mission in 1993. SOURCE: Courtesy of NASA. involved. While originally planned as a 15-year mission, by 2001 servicing Hubble had extended its lifetime to 20 years. Mission planners have credited the follow- ing factors for the success of Hubble servicing: a focus on “EVA friendliness” in the telescope design, and the extensive preplanning involved in the servicing missions themselves. Servicing Hubble requires the space shuttle’s robotic arm to grapple Hubble and to place it on and latch it to a special carrier platform that can rotate and pivot the telescope in the shuttle payload bay. 

HUMAN AND ROBOTIC SERVICING OF FUTURE SPACE SCIENCE MISSIONS 89 FIGURE 4.1.2  The Hubble Space Telescope in orbit. The spacecraft has been repaired and upgraded repeatedly during nearly two decades in orbit. SOURCE: Courtesy of NASA. After the Columbia accident, the NASA administrator decided to cancel a planned servicing mission to Hubble. After a significant public outcry, NASA announced that it would attempt to service Hubble robotically, using a dedicated spacecraft and manipulator system that would be developed in a relatively short period of time and launched before Hubble degraded too significantly to be repaired. However, Hubble was not designed for robotic servicing, and such an operation had never been demonstrated before. Under congressional pressure, NASA sponsored a National Research Council (NRC) study on whether a robotic or a human servicing mission would accomplish the goals of a servicing mission most successfully.1 After thorough deliberation, the committee concluded that the state of robotic technology was not advanced enough to complete the mission and successfully overcome unanticipated risks that are bound to occur on any given mission. Many of the findings in that NRC report focused on the fact that this complex work in space had never been done before by a robot. Since much of the technology had not yet been proven, more time would have been required to test the technology and, in some cases, develop it from scratch. Preparing for a robotic repair mission would have exceeded the planned 39-month development schedule for the servicing mission. Besides this risk to the development schedule, other risks included successfully launching the vehicle to the desired orbit and conducting a successful rendezvous, proximity operations and grappling by the robot with the telescope, and successfully completing the necessary combination of complex autonomous and robotic activi- ties. The final risk lay in the inflexibility of a robotic mission to address unforeseen Hubble equipment failures that might occur before the mission was completed. To plan for such failures would have required even more development time, and it would have added complexity to the already difficult robotic mission concept. continued 

90 LAUNCHING SCIENCE BOX 4.1  Continued By contrast, the NRC committee that wrote this report on options for extending the life of the Hubble concluded that it was much simpler to do a human servicing mission, especially since all four previous human servicing missions to Hubble had fully met their objectives. Furthermore, that committee found that “space shuttle crews, in conjunction with their ground-based mission control teams, have consistently de- veloped innovative procedures and techniques to bring about desired mission success when encountering unplanned for or unexpected contingencies on-orbit.”2 The same NRC committee discussed the findings of the Columbia Accident Investigation Board’s (CAIB’s) report and NASA’s reaction to the findings (which was full acceptance and the institution of even more stringent guidelines on shuttle missions), and it found that meeting both the CAIB and NASA requirements for a shuttle servicing mission to Hubble was possible, and that such a mission would be no riskier than one to the International Space Station. On October 31, 2006, NASA administrator Michael Griffin decided that Hubble would receive a final servicing mission before the retirement of the space shuttle. Dr. Griffin’s decision was based not only on the findings of the NRC report, but on his own knowledge of the robotic servicing option prior to his becoming NASA administrator. The final Hubble servicing mission, scheduled for October 2008, is to be performed by the crew of space shuttle Atlantis (STS-125). On this mission, astronauts will install the Cosmic Origins Spectrograph, Wide Field Camera 3, fine-guidance sensor, aft shroud cooling system batteries, and gyros; repair the Space Telescope Imaging Spectrograph and the Advanced Camera for Surveys; and add the Soft Capture Mechanism. One of the planned repairs to Hubble includes equipment and techniques originally developed for the robotic servicing mission and involves accessing a part of Hubble that was never designed for ser- vicing. (See Figure 4.1.3.) With these new upgrades to the telescope’s capabilities, scientists will be able to track galaxy evolu- tion, learn more about the formation of stars, measure the composition and structure of baryonic matter, study the characteristics of quasars, and investigate the nature of cold dark matter. These additions to the Hubble Space Telescope will illuminate both the large-scale structure of the universe and the progressive changes in chemical composition of matter as the universe has grown older. FIGURE 4.1.3  Astronaut practices for the final Hubble Servicing Mission. He is using a device that was originally developed for the robotic ser- vicing mission to prevent screws from floating free. This device will be used for a part of Hubble that was never intended for servicing. SOURCE: Courtesy of NASA. 1National Research Council, Assessment of Options for Extending the Life of the Hubble Space Telescope: Final Report, The National Academies Press, Washington, D.C., 2005. 2Ibid., p. 91. 

HUMAN AND ROBOTIC SERVICING OF FUTURE SPACE SCIENCE MISSIONS 91 designed for such servicing. This indicates that it should be possible to develop some servicing capabilities into a spacecraft without adverse impact on design or cost. In addition, in recent years a new servicing option has also emerged—robotic servicing. ROBOTIC SERVICING OF SPACECRAFT Robotic servicing is a capability that has yet to be fully proven, although substantial advances have been made in recent years. After NASA canceled the shuttle servicing mission for Hubble following the Columbia accident, the agency sought to develop a robotic servicing mission. This project was eventually canceled for several reasons, including its complexity and the short development time available. It would have been one of the most complex spacecraft ever developed, and there was insufficient time to build and fly it before the expected demise of Hubble. One of the major factors contributing to the spacecraft’s complexity was that although Hubble was designed to be serviced by humans, it was not designed to be serviced robotically. The Hubble servicing spacecraft would have had to rendezvous with what is known as a noncooperative target, increasing the complexity level. In addition, several repair tasks scheduled for Hubble were never even intended for humans, let alone robots. The challenge for robotic servicing can be divided into two primary areas: (1) automated rendezvous and dock- ing and (2) the exchange of equipment and fuel. In the past several years, the United States and its European ISS partners have demonstrated significant advances in these areas. In 2007 the Orbital Express mission demonstrated the ability to automatically rendezvous and dock with another spacecraft and also to transfer fuel and replace vari- ous equipment such as batteries and electronic units. In 2008, the European Space Agency’s (ESA’s) Automated Transfer Vehicle (ATV) successfully delivered the Jules Verne cargo spacecraft to the ISS after performing an automated approach and docking. SUCCESSFUL ROBOTIC SERVICING MISSIONS Orbital Express, a mission managed by the Defense Advanced Research Projects Agency, was launched on March 8, 2007, with the goal of demonstrating the technical feasibility of autonomous, on-orbit satellite servic- ing. The spacecraft’s two vehicles were the Autonomous Space Transfer and Robotic Orbiter (ASTRO), which was the servicer, and the Next Generation Satellite/Commodity Spacecraft (NEXTSat/CSC), which was the client (Figure 4.4). These two vehicles had the following goals: interface with a satellite, transfer propellant in orbit between a serviceable satellite and the servicing satellite, and transfer a battery and computer. In addition to success- fully completing these requirements, the vehicles also demonstrated autonomous operations such as rendezvous, operation in close proximity to the target satellite, and capture of the target satellite. Four levels of supervised autonomy were demonstrated, showing vehicle operation in increasingly challenging environments. ESA’s ATV is another example of successful autonomous robotic technology. The ATV is a highly intelligent spacecraft that has an onboard high-precision navigation system that automatically guides ATV on a rendezvous trajectory with the ISS and docks it with the Russian service module Zvedzda. It is a resupply vehicle that also at regular intervals boosts ISS into higher orbit to offset the drag effects of oxygen molecules above Earth’s atmo- sphere. There are five planned ATV launches from 2008 until approximately 2018. The Jules Verne ATV launched on March 9, 2008, and docked with the ISS on April 3, 2008, providing fuel, water, oxygen, food, spare parts, and clothing. In the course of its launch and docking, Jules Verne demonstrated relative navigation with the ISS, its capability to execute the Collision Avoidance Maneuver, successful use of the relative Global Positioning System, close proximity maneuvering and control, and contingency maneuvers. The ATV can remain attached as a pres- surized and integral part of the station for up to 6 months, at which point it will de-orbit and burn up in Earth’s atmosphere along with up to 14,000 pounds of waste from the ISS. The interior of the ATV consists of the service module and the pressurized integrated cargo carrier. Thus, the ATV combines full automatic capabilities with human spacecraft safety requirements. Although no crews will be launched in the ATV, once it is docked with the ISS, astronauts will be able to use its interior like any other habitable part of the ISS.   See ESA’s ATV Web site at http://www.esa.int/esaMI/ATV/index.html. 

92 LAUNCHING SCIENCE FIGURE 4.4  Illustration of Orbital Express’s two vehicles: the Autonomous Space Transfer and Robotic Orbiter, or ASTRO (servicer) (left), and the Next Generation Satellite/Commodity Spacecraft, or NEXTSat/CSC (client) (right). SOURCE: Cour- tesy of the Defense Advanced Research Projects Agency. FUTURE SERVICING CAPABILITIES The successes of Orbital Express and ESA’s ATV demonstrate how far technology has progressed and hint at a future when the robotic servicing of space science missions will not only be possible but also a practical and cost-effective solution to servicing spacecraft in orbit. If NASA investigates the value of on-orbit servicing using autonomous robotics, more advanced concepts can be tested and implemented. The basic requirements for servicing include the following: some degree of autonomous rendezvous capabil- ity, a capture mechanism, some kind of manipulator or arm to grapple with specific elements on the satellite, and finally, the satellite itself must be designed so that it can be serviced. The Hubble Space Telescope was designed to be serviced, but the James Webb Space Telescope, because it is being sent farther out, to the Sun-Earth L2 point, is not. To be serviceable, a satellite needs systems that are designed to be easily removed and replaced and needs to have an interface design that works with current or future tools. High-fidelity physical and electrical simulators on the ground are important for designing new instruments—allowing “fit checking” of devices to be installed and the development of contingency procedures. Many of the items repaired or replaced on Hubble were not designed to be serviced, but access to replicas of the spacecraft both in the underwater training facility and in a ground facility at the NASA Goddard Space Flight Center (see Figure 4.5) made it possible for engineers to 

HUMAN AND ROBOTIC SERVICING OF FUTURE SPACE SCIENCE MISSIONS 93 FIGURE 4.5  Hubble Space Telescope high-fidelity simulators in the Goddard Space Flight Center clean room along with other flight items. At back center is the axial equipment bay where Corrective Optics Space Telescope Axial Replacement (COSTAR) and the other axial instruments were fit-checked. At back right is the equipment bay with real electrical and computer compo- nents for testing and checking out electronics boxes. The ability for astronauts to train on the ground with hardware similar to that in space is important for human servicing missions. SOURCE: Courtesy of NASA. develop repair procedures and, in some cases, specialized tools that were used successfully to repair or replace failed or aging components. FUTURE HUMAN AND ROBOTIC SERVICING OPTIONS The Orion spacecraft, although capable of traveling beyond low Earth orbit, has a number of limitations with respect to acting as a servicing mission. Orion has limited capability for change in velocity (delta-v) and would require additional propulsion to leave low Earth orbit. The spacecraft is not equipped with an airlock, which would therefore require that the entire cabin be depressurized or that an airlock be provided for the crew. In addition, Orion has limited extra mass and volume capacity for carrying equipment for a servicing mission, such as a robotic manipulator system, toolkits, and any equipment to be installed. Any plan to use the Orion for a servicing mission will have to address these limitations. Options include launching a second spacecraft to carry servicing equipment, designing spacecraft to include many of the necessary tools and even an airlock, and launching a dedicated servic- ing spacecraft to which Orion would rendezvous and dock before journeying to the spacecraft to be serviced. Several of the mission concepts evaluated in this report would operate at the libration points a significant distance from Earth (see Figure 2.1b in Chapter 2). Several telescopes currently operate at these locations, and others, such as the James Webb Space Telescope, are scheduled to be placed there. Although these sites are attrac- 

94 LAUNCHING SCIENCE tive for operating heliophysics and astronomy observatories, they are less than ideal for servicing purposes owing to their distance and communications lag times and for other reasons. One possibility is to move the spacecraft closer to Earth for servicing purposes and then move it back to its operating location. Several proposals exist for establishing a servicing node or way station at a closer location that could be visited either by humans in an Orion spacecraft or by a robotic servicing spacecraft. Transferring the observatory from one Lagrangian point to another requires very little change in velocity and subsequently only small amounts of fuel. The James Webb Space Telescope will operate at the Sun-Earth L2 point, which maintains the orientation of the spacecraft with respect to the Sun and Earth and is an attractive location for astronomy missions because it is thermally stable (sources of heat and scattered light are in the same general direction), it has continuous com- munication capability with Earth, it can easily make use of solar power, and it has very low debris and dust. The James Webb Space Telescope was not designed for human or robotic servicing. Its delicate thermal shield could collapse if exposed to a spacecraft’s thruster jets, and rocket exhaust could condense on the telescope’s mirror. The instruments were also not designed for access. A servicing mission at the Earth-Moon L1 (or L2) point would allow easy access to the Sun-Earth L2 location. A servicing station at this location would be able to make use of the same advantages afforded by the Sun-Earth L2 location, but it would also provide easy access to lunar-capable architecture (approximately 4-day transfer time) and build on the lunar navigation-and-communication network. By contrast, low Earth orbit makes it difficult to service spacecraft because propulsive loads would be too large for the lightweight support structures of space telescopes. Additionally, these telescopes would be exposed to damage from the contaminants and debris in low Earth orbit. To get humans to the Earth-Moon L1 point, the Constellation System will be required. Other observatories (e.g., lunar surface missions, Single Aperture Far Infrared Telescope, Advanced Technol- ogy Large-Aperture Space Telescope) may also be operated at the Sun-Earth L2 and Earth-Moon L1 locations in the future. It is possible to have a servicing station for spacecraft at Earth-Moon L1, as illustrated in Figure 4.6. Such a servicing station would have a servicing node, which would remain in orbit and would require some additional avionics and propulsion capabilities beyond those found in a simple airlock. In addition to servicing spacecraft out at Earth-Moon L1, it could also possibly be used to aid lunar surface exploration. This servicing node could enable two types of robotic servicing: a robot operated by astronauts and a robot operated autonomously from Earth. Ultimately, a servicing node at the Earth-Moon L1 or L2 point would make it possible to construct large astronomical observatories that surpass even Ares V single-launch capabilities, as shown in Figure 4.7. It could also be used to facilitate lunar exploration goals, and in the far term, an Earth-Moon L2 point servicing mission could provide a stepping stone between lunar missions and Mars missions. It could be used as a test site for issues such as duration in space, distance from Earth, communication delays, and supply issues. However, these missions might prove to be expensive, and it would be reasonable to expect both NASA’s Science Mission Directorate and the Exploration Systems Mission Directorate to share these costs. ADVANTAGES OF SERVICING MISSIONS Although the advent of the Orion spacecraft and the Constellation System’s other capabilities reopens the possibility of human servicing of spacecraft, it is clear that this possibility will create operational complexities greater than those that NASA has faced in the past. Large spacecraft deployed at distant locations may have to be moved nearer to Earth for servicing, and human servicing missions will require the development of major new pieces of equipment, such as airlocks, which are not currently planned by NASA. The recent development of robotic servicing options opens up additional exciting possibilities for future spacecraft servicing. It is possible to conceive of a robotic servicing node at the Earth-Moon L1 point similar to that discussed earlier. A robotic spacecraft could be designed to rendezvous with an observatory and to repair and   The L1 and L2 points are unstable, so a small amount of fuel is required to keep the spacecraft from drifting away. Harley A. Thronson, Daniel F. Lester, Adam Dissel, Jason Budinoff, and John Stevens, “On the Way to the Moon: Using NASA’s Constellation Architecture to Achieve Multiple Science Goals in Space,” presentation to the Committee on Science Opportunities Enabled by NASA’s Constellation System, June 9, 2008. 

HUMAN AND ROBOTIC SERVICING OF FUTURE SPACE SCIENCE MISSIONS 95 FIGURE 4.6  Concept for in-space servicing throughout the Earth-Moon system. This figure shows the stack after Earth-orbit rendezvous of the Orion with the Centaur and servicing node. The Centaur is ejected before rendezvous with the observatory. NOTE: EVA, extravehicular activity; LIDS, Low-Impact Docking System. SOURCE: Courtesy of NASA. upgrade it. Another option could be a combination of human and robotic servicing, using robotic spacecraft for basic upgrading and other tasks and using human servicing for more complex tasks. For all of these robotic and human servicing capabilities to be realized, servicing capabilities must be con- sidered as the telescopes and spacecraft themselves are built. This type of adjustment might involve additional costs when constructing the spacecraft. However, making spacecraft serviceable will provide numerous savings and benefits in the long run. Components with long lifetime ratings are more expensive than those with shorter ones. Servicing can mitigate this issue (since components need not have such long lifespans), and replacing com- ponents with new technology can increase capability and capacity. The ability to service a spacecraft also enables the correction of various mission-design problems and failures and the ability to increase the longevity of critical systems by transferring fuel and cryogenics. Additionally, while NASA’s space science budget has been stable in recent years, cost overruns and setbacks (e.g., the Columbia accident and its return to flight costs and effect on the shuttle launch schedule) have caused budgets for the various divisions in the Science Mission Directorate to become even more restrictive. Because of the expense of large space observatories, launching fewer platforms and gaining the ability to operate them longer may be a cheaper way to continue to do new science, by fitting new capabilities on already-orbiting spacecraft. Finding:  The Constellation System and advanced robotic servicing technology make possible the servicing and in-space assembly of large spacecraft. Finding:  Designing spacecraft components for accessibility is essential for in-space servicing and is also advantageous for preflight integration and testing. 

96 LAUNCHING SCIENCE FIGURE 4.7  An illustration of the concept of operations for a servicing node at the Earth-Moon L1 point. The facility to be Figure 4.7.eps serviced, in this case an astronomical observatory, travels by means of low-energy (i.e., velocity) transfer to the Earth-Moon L1 (or L2) location, where it is met by the Orion and servicing node module, which features an airlock. This system can re- Bitmap image - Low resolution main onsite for 2 to 3 weeks and perhaps could carry out other tasks. On completion of the servicing mission, the Orion crew module returns to Earth and the servicing node may remain at one of the libration points for further duties. NOTE: SM, service module; ΔV, delta-v; TEI, trans Earth injection. SOURCE: Courtesy of NASA. Human servicing missions have already enabled fantastic advances in space science through the Solar Maxi- mum Mission and the Hubble Space Telescope. In the future, these advances can only improve as technology and science continue to progress at increasing rates. Future servicing missions may continue to be human servicing missions, or a combination of human and robotic capabilities, but as robotic technology advances, robotic autono- mous missions to scientific spacecraft will be possible. ESA is already using autonomous robotics to service the ISS, and Orbital Express has proven that some of the technology already works. As robotic technology continues to improve, new servicing capabilities will become possible, creating opportunities for even greater science that better keeps up with evolving space technologies.