AbstractThe STARDUST Discovery mission will collect samples of cometary coma and interstellar dust and return them to Earth. Five years after launch in February 1999, coma dust will be captured by impact into ultra-low-density silica aerogel during a 6 km/s flyby of Comet Wild 2. The returned samples will be investigated at laboratories where the most critical information on these primitive materials is retained. The Jet Propulsion Laboratory provides project management with Lockheed Martin Astronautics as the spacecraft industrial partner. STARDUST management is aggressively pursuing cost control through the use of Total Quality Management principles, specifically operating in a Project Engineering and Integration Team that "flattens" the traditional hierarchical structure by including all project elements from the beginning, in a concurrent engineering framework focusing on evolving Integrated Mission Capability.

1. INTRODUCTION
2. CONCURRENT ENGINEERING FOR MISSION SUCCESS
3. STARDUST SCIENCE GOALS
4. STARDUST MISSION DESIGN
5. FLIGHT SYSTEM DESIGN
6. PROJECT STATUS SUMMARY
7. CONCLUSIONS

The Discovery Program is about a new way to continue the legacy of the Mariners, Voyager, Magellan, and Galileo in deep space exploration. Discovery is changing the way NASA does business. It is a central element in a complete culture change for planetary exploration and space science. Discovery's goal is to achieve results faster, better, and cheaper. It will be more effective, do more with less-specifically, carry out planetary flight missions with highly constrained total cost.

STARDUST was selected from a pool of 28 proposals in 1994 [3]. It becomes the fourth mission in the series: Near Earth Asteroid Rendezvous, Mars Pathfinder, and Lunar Prospector.

Historically, planetary missions evolved to large, complex platforms with up to 14 scientific experiments and price-tags of up to $2 billion. These missions endeavored to do remote-sensing and in-situ investigations on extremely stringent diets of power, mass, and volume. The struggles in the scientific community to be one of their cramped "passengers" were difficult and frustrating. With their high price-tags, such missions are clearly on the path toward extinction.

STARDUST proposes to exactly reverse the paradigm. It is a sample return mission whose fundamental premise is to bring the essence of the solar system, material from a comet, home! With samples back on Earth, literally hundreds of experimenters can participate. They can apply existing instruments-with relatively unlimited power, mass, and volume constraints-operational in the finest labs and universities. This will allow participation in solar system exploration by a broad community. And the opportunity is offered at a Discovery price of less than 10% of the traditional approach!

STARDUST plans the first return of material from a solar-system body since the Apollo and Luna sample-return missions of the 1970s and, more importantly, the first ever from beyond the Earth-Moon system. As such it becomes a model for planning follow-on sample-return missions to other planetary bodies. The simplicity and compactness of the Sample Return Capsule (SRC) should be very attractive to follow-on applications. Figure 1 shows the STARDUST spacecraft in its sampling configuration.

Figure 1. The STARDUST Spacecraft (Two Views)

The major features of the STARDUST flight system are (1) the SRC, about a meter in diameter, shown open like a clamshell with the dust collector grid deployed into the dust stream above the so-called "Whipple shield"; (2) the shield consists of two plates with Nexel™ curtains between to stop the high-speed particles from impacting sensitive spacecraft elements; (3) solar-arrays; and (4) the Cometary and Interstellar Dust Analyzer (CIDA) to be provided by Germany. The flight system also carries an upgraded Voyager camera to provide optical navigation capability. The plan is to also use this camera for imaging the nucleus of the comet to a resolution an order-of-magnitude better than Giotto imaged Halley.

2. CONCURRENT ENGINEERING FOR MISSION SUCCESS

The STARDUST Project operates on the Total Quality Management (TQM) principle of concurrent engineering, which saves time, improves communications, and reduces costs by bringing all elements together early. The objective is to mitigate the traditional culture of serial processing of project elements.

Concurrent engineering means that producibility, testability, and operability are fully considered in all phases. But how is this done? More importantly, how is it done efficiently? The key to success is modern computer and communications technologies.

The STARDUST team has engineered a common, collaborative server approach to provide a central repository of all communication products. The products include drawings, presentations, memos, formal documents, spreadsheets, etc. This collaborative server environment brings visual access to all project players at their office workstations or in a conference room setting. With full, flexible video access to the products, it remained to ensure audio communications with equal user access and flexibility. This has been achieved by leasing a dedicated "meet-me" teleconference line. This audio dial-in access is available from virtually any telephone. Up to about 50 participants can access any meeting from sites anywhere. In addition, with a portable computer containing downloaded products from the server, both visual and audio involvement is achieved from any telephone. The impact has been to allow team members just-in-time involvement in meetings while working in their offices. This "virtual meeting room" environment has been put in place to save travel (even from office to conference room), allow last minute and real-time changes to products such as documents, presentations, spreadsheets, etc., and facilitate "spur-of-the-moment" meetings "in the server." Thus widely-separated team members at Lockheed Martin Astronautics (LMA) in Denver and the Jet Propulsion Laboratory (JPL) in Pasadena can meet virtually and quickly on issuesand have the full repository of Project data and communication products available on screen.

By placing a priority on communications with this cost-saving, user-friendly technique, the STARDUST team is able to focus all elements

Figure 2. Integrated Team Management Structure

in a concurrent engineering group called the Project Engineering and Integration Team (PEIT). The PEIT meets weekly as a central forum for discussion and actions. It is led by the Project Engineer and comprises the engineering leaders or cognizants of the key project elements. Here, cross-system issues are brought front-and-center. End-to-end information issues are being dealt with early in the development cycle such that the tradition of spacecraft first, and testability and operability second, is being broken. The PEIT focus is on teamwork. This involves everybody in the big-picture fundamental mission success metric, return of cometary and interstellar material.

The PEIT also involves, at the appropriate level, business functions, safety, and public outreach/education. It provides the central control forum for the principal investigator (PI) and project manager, providing an arena where performance and accountability in each element of the project is highly visible. Figure 2 depicts the concurrent engineering and management structure that "wraps around" the major project elements by operating in the flexible "virtual meeting" communications environment.

Key to the successful achievement of the mission goals within the cost cap are the controls and processes applied by the PI and his management team. The key development process is the use of a central mission test environment, the Stardust Mission Test System (SMTS), to bring all the elements together early in an end-to-end environment. The idea of keeping the focus at a mission level is to faithfully serve the ethic derived from the famed Lockheed "Skunkworks" to reduce risks by "testing it like you plan to fly it."

Figure 3 shows the integrating role of the SMTS in getting at the mission interfaces early and iterating in phases of increasing interface complexity. This process, pioneered on the Mars Pathfinder mission, is aimed at discovering interface issues early, fixing them, and allowing a seamless, smooth transition in to the Assembly, Test, and Launch Operations (ATLO) portion of the project schedule.

Figure 3. Mission Testbed as an Evolutionary Integration Environment

Specifically, STARDUST is exploiting capabilities provided by the Spacecraft Technology Center and Spacecraft Test Laboratories at LMA and the Flight System Testbed at JPL. Each of three product-accountable managers will deliver STARDUST-specific hardware and software at different levels of maturity to this mission integration environment.

The SMTS plan includes engineering development units, software, mission models, dynamic models, and, where possible, flight articles. This will provide the PEIT with real data, not paper, to validate the designs; control algorithms; and interfaces in a mission/system context as they are produced by the design and fabrication elements at different phases.

The primary ethic of the STARDUST Project is to develop the target low-risk Integrated Mission Capability (IMC), e.g. the complete set of integrated and tested hardware, software, operational procedures, analysis procedures, and facilities to attain the fundamental Mission Success criterion, and operate it successfully within the baseline budget established with NASA. The IMC is the product of the manage-to-budget culture central to the project's continued viability.

3.1 Science Instruments

STARDUST carries only two dedicated science investigations: the aerogel dust collector and the CIDA. All other science data is obtained from engineering functions that are required for the operation of the spacecraft. These engineering instruments are the navigation camera and the Whipple-shield flux monitors. Dynamic science is obtained without special hardware.

3.1.1 Aerogel Dust CollectorThe dust collector will simply expose blocks of underdense, microporous silica aerogel and other low-density media to the sample flux. The collector will consist of modular aluminum cells housing 1- to 2-cm thick aerogel blocks. The cells will form a two-sided, grid-shaped array that will deploy from the SRC.

One side of the array, as indicated in Figure 4, will collect comet particles and the opposite side, interstellar dust particles. The useful collecting area will be about 1000 cm² for each target sample. The density of impacts will be low and this dual use will cause no problem with sample discrimination. The aerogel array will have an average density of 0.02 g/cm³. The collector will be totally inert and have only to be exposed and recovered.

Figure 4. Sample Return Capsule with Collector Deployed

Extensive experience exists in both laboratory and space flights with aerogel for collecting hypervelocity particles [12, 13]. More than 2.4 m² of silica aerogel capture cells have been flown and recovered on Shuttle flights, Spacehab II, and Eureca.

3.1.2 Comet and Interstellar Dust Analyzer The CIDA is essentially the same instrument that flew on Giotto and the two Vega spacecraft, obtaining unique data on chemical composition of individual particulates in Halley's coma. It consists of an inlet, a target, an ion extractor, a time-of-flight (TOF) mass spectrometer (MS), and an ion detector. The inlet is baffled to prevent sunlight from entering the instrument and raising the background noise in the detector. The target is 50 cm² of corrugated silver or other heavy metal. A light flash which accompanies the initial impact sets the zero for the TOF measurement. Electrostatic grids extract positive or negative ions from the impact microplasma. These ions move down the bent-tube TOF MS where an electrostatic reflector focuses ions of similar energies onto the ion detector. Measuring arrival time determines the mass of the ions. This instrument is sensitive over a range of 1 to 150 atomic mass units. Even sub-µm-sized particles will produce observable signals and compositional profiles.

3.1.3 Navigation CameraThe camera optics for STARDUST are spare Voyager wide-angle units. Plans also include a single Voyager eight-position filter wheel and thermal housing and a 1024 1024 charge-coupled device detector with 12-µm pixels. This will give 6-mperpixel resolution at 100 km.

The Voyager camera provides a "bargain" price capability with a proven heritage. Although other candidate cameras were evaluated, none provided a more competitive cost/benefit ratio. Selection was entirely in keeping with the "capability-driven" approach to achieve a faster-cheaper solution. It avoids the situation where "better is the enemy of 'good enough'."

Shutter speed is 5 ms. Some image motion compensation is planned by moving the imaging scan mirror. This will improve the resolution to perhaps 5 m, an order of magnitude better than Giotto. Less dust opacity and the lower flyby speed (6 km/s versus 70 km/s) promise better and more comprehensive imaging of the nucleus.

3.1.4 Whipple Shield Dust Flux MonitorDust impact sensors will be part of the STARDUST payload to measure the dust flux along the trajectory through the coma. The dust impact sensors serve three functions: (1) to monitor the dust environment for spacecraft health and interpretation of any spacecraft anomalies, (2) to measure the mass distribution, spatial distribution, and total fluence of the dust, and (3) to provide the context for the collected dust sample.

The dust flux monitor will employ sheets of electrically polarized polyvinylidene fluoride (PVDF). A high-velocity particle (>1 km/s) impacting the PVDF film causes very rapid local depolarization in the foil volume destroyed by the particle. This results in a fast (nanosecond-range) current pulse which is fed to the electronics. The total charge carried by the current pulse is determined by the mass and velocity of the impacting particle and can be calibrated in the laboratory for the Wild 2 flyby velocity of 6 km/s.

The STARDUST instrument will have two sensor units. Sensor 1 has a 28-mthick circular PVDF film with sensitive area 200 cm². Sensor 2 has a 6-mthick circular PVDF film with sensitive area of 20 cm². Each sensor will have four electronic thresholds, which determine four mass thresholds. A 16-bit counter for each mass threshold will record the number of impacts. The combined sensors will encompass the mass range from 10¹¹ to 10⁴ g.

The instrument is being procured from the Laboratory for Astrophysics and Space Research, Enrico Fermi Institute, University of Chicago. A similar instrument was carried on the two USSR Vega spacecraft which flew past Halley's comet in March 1986. The two instruments performed successfully throughout the missions and provided measurements of the dust fluxes and mass distribution in the coma of Comet Halley [8].

Extensive laboratory calibrations have been conducted to document the performance of the PVDF sensors [9, 10].

3.2 Anticipated Science Return

The wealth of data which will result from the STARDUST mission is due to its multi-faceted nature. It will collect interplanetary dust and, with mission implementation optimized, it will also collect extra-solar system grains, i.e. the interstellar dust. While cometary samples are of intrinsic interest for the entire comet science community, they hold considerable interest for exobiologists as well.

3.2.1 Cometary DustComets presumably formed in the outer solar nebula, where the temperature remained low enough that many intact interstellar grains (IGs) should have survived nebular processing [7]. At present it is not known what fraction of cometary dust is presolar, and what fraction was formed in the solar nebula and transported to the region of comet formation. It is also not known how the nebular accretion of IGs into larger aggregates may have changed their observable properties.

For comet samples that can be captured intact, it should be possible to determine the following:

(1) The mineralogical, elemental, and chemical composition of comets at the sub-µm scale.

(2) The extent that building materials of comets are found in interplanetary dust particles (IDPs) and meteorites.

(3) The state of water in comets-whether in ice or in hydrated minerals.

(4) Mixing of inner nebular materials (i.e. chondrule fragments) to the comet formation region.

(5) The presence of isotopic anomalies.

(6) The nature of the carbonaceous material and its relationship to silicates and other phases.

(7) Evidence for pre-accretional processing either in the interstellar medium or in the nebula (including cosmic ray tracks, sputtered rims, etc.).

3.2.2 Cometary Volatiles The CIDA carried on STARDUST will provide direct measurements of volatile species in the impacting dust samples and is expected to obtain much more information on complex molecules than for the Halley flybys because impacts with coma particles will be 100 times less energetic.

3.2.3 Interstellar DustAt present, astronomically derived information on IGs comes primarily from observations of extinction, scattering, polarization, and infrared emission. While such astronomical observations provide clues to the nature of IGs, they are not sufficiently definitive to confidently match the particles with theoretical models. Basic information, such as the abundance of SiC (from carbon stars), the abundance of graphite, grain morphology, silicate mineralogy, the role of radiation processing, grain ages, and the association of silicates and carbonaceous matter, is highly uncertain. Collection of even a few degraded particles would provide a unique and historic opportunity to directly examine solid matter that formed outside the solar system. This information would provide powerful constraints on grain models and provide insight in the relationship of presolar and meteoritic materials.

STARDUST will provide ground truth on interstellar grain models and perhaps reveal physical properties and effects of processes that were previously unforeseen. It will provide data on the degree of processing after initial formation in circumstellar regions, and information on the relative importance of oxygen-rich and carbon stars in producing interstellar dust. Isotopic ratios in the samples will yield information on nucleosynthetic processes in a variety of stars. In the case of hydrogen, isotopic fractionation will provide insight into the ion-molecule reactions that are a favored explanation for high deuterium/hydrogen ratios in some molecular clouds and trace components in meteorites and IDPs.

3.2.4 Exobiology ImplicationsComets are now known to contain large quantities of volatiles, including organic compounds and a rich variety of microparticles of various types (pure organic particles, silicates, sulfides, and mixed particles) with a graduation of sizes that extends to sub-µm diameters. With high surface areas, juxtaposed chemical constituents, and their easy transportability, these particulates may have been critically important for abiotic catalytic activity, macromolecular synthesis, and subsequent chemosynthetic pathways [5, 6]. These are the accepted prerequisite processes for the origin of life.

Comets, being rich in water and other volatiles, have been postulated to be transporters of volatile and biogenic elements to a young Earth. The study of cometary material is felt by many to be essential for understanding the formation of the solar system, and, most importantly to exobiology, the interstellar contribution of pristine, early-formed organic matter from several different environmental regions. How the biogenic elements entered the solar system, were transformed by processes operating therein, became distributed among planetary bodies, and what molecular and mineral forms they took during this history are questions of major importance for exobiology. Comparison of the compositions of the volatiles contained within cometary material with those found in carbonaceous meteorites and interplanetary dust will provide a basis for determining what commonalities in source regions can be attributed to the materials in these putatively related objects. The analysis of minerals like carbonates, clays, and sulfates in comet dust will also be significant for the history of interaction between water and minerals in the early solar system [4].

4.1 Trajectory

STARDUST's seven-year, three-loop, VEGA (Earth gravity assist) trajectory is designed (1) to fly by Wild 2 at a low velocity while it is active, (2) to maximize the time for favorable collection of interstellar dust, and (3) to minimize the C₃ (escape energy from Earth) and V requirements for the mission so that a small launch vehicle may be used. Figure 5 shows the spacecraft trajectory and the location of Earth, Wild 2, and the deep space maneuvers on orbit.

Figure 5. STARDUST Trajectory (E-E-Wild 2-E)

The STARDUST spacecraft will be launched in February 1999. The first orbital loop is a 2-year VEGA path with a 171 m/s ^{V trajectory correction maneuver (TCM) near aphelion. This V will set up the Earth swingby that will pump the orbit up to the 2.5-year loop, which the spacecraft will fly twice. At 160 days before encounter, a small V of 66 m/s will set up the Wild 2 flyby. This will occur on 1 Jan 2004, at 1.86 AU and 97.5 days past Wild 2 perihelion passage. The spacecraft will approach the comet at 6.2 km/s from sunside with a 70o} phase angle. Coma fly-through will be on the sun side at a planned miss distance of 100 km. Flyby is five years after launch, and Earth return, two years later.

Interstellar dust will be collected on two of the three post-aphelion legs of the orbits, where the spacecraft orientation makes it possible to collect IGs at low velocity. In these portions of the orbit, indicated by ISP 1 and 2, the vectors of the IGs and the spacecraft align favorably to yield lower relative velocities.

4.2 Flight System Performance

Table 1 shows propulsion parameters and Table 2, the launch capability for the baseline launch energy requirement of 26 km²/s².

V (m/s) 357

Isp (s) 220

Trapped Prop. (kg) 0.5

ACS Prop. (kg) 12

Pressurant (kg) 0.2

Despin Propellant (kg) 1

* Preliminary Design Review held Sept. '96.

4.3 Wild 2 Encounter Phase Design

4.3.1 Wild 2 Encounter GeometryThe spacecraft will encounter Wild 2 at 97.5 days past perihelion at 1.86 AU from the Sun when Wild 2 is far from its peak active period and relatively safe for a close flyby. The spacecraft will approach Wild 2 from above its orbital plane, then dip slightly below it. Figure 6 shows the geometry of the flyby, which will be at 100 km on the sun side.

Figure 6. Orbital Geometry at Closest Encounter, where RS is the orbital plane, R=radial direction, Sun to Wild 2, S=orthogonal to R along Wild 2 velocity direction.

Investigations of the navigation accuracy and the impacts on the entire encounter profile are based on this aim point which is regarded as worst-case targeting. Final selection of the aim point may be further out and the encounter date may be shifted depending on the results of the ground observations of Wild 2 in early 1997.

4.3.2 Navigation PlanSTARDUST will use both radio and optical navigation (OPNAV). Early knowledge of the orbital state of Wild 2 based on ground observations gives an estimated position uncertainty of about 1500 km (1 ). An improvement over this is expected at about E-50 days, after OPNAV has been in operation for some time. The adopted navigation plan can deliver with an accuracy of 8 km (1 , cross track ) and 11 seconds (1 , time of closest encounter). This plan is based on a ground-commanded trajectory correction maneuver (TCM). The last OPNAV image will be sent at E-12 hr and the last TCM executed at E-6 hr. The two-way light time will be 40 minutes, which will leave about 5 hours to prepare the last TCM command from the ground. To limit telemetry volume, on-board image data processing, "windowing," and 2:1 data compression are planned.

4.3.3 Encounter Phase Mission ScenariosThe encounter phase nominally starts at 150 d before and ends 150 d after comet encounter. Accurate delivery of spacecraft to the desired aim point is accomplished with OPNAV. On-board acquisition of comet images will begin at about E-150 d. Ground-based image processing and TCM commands are used until E-1 d. Thereafter, images will be processed on board in order to guide the mirror to contain comet images in the field of view (FOV) of the camera. A mirrored tracking device in the camera system will protect the optics during the coma fly-through and reduce image smear.

To attain low-cost operation, comet imaging goals, being secondary science, will not dictate the mission scenarios. Instead, imaging science will be acquired as the opportunity permits. The imaging science plan is for all data taken before E-4 min to be sent back as the OPNAV tracking schedule permits and all images taken from E-4 min to E+4 min to be recorded for delayed telemetry.

Only about 100 frames of the highest resolution images near closest encounter will be recorded, even though the data-link capability of the spacecraft (2 kilobits/s (kbps) on the 34-m diameter Deep Space Network (DSN) antenna and 8 kbps on the 70-m dish) makes it possible to acquire more than this. Spacecraft communications will be with the 34-m high-efficiency (HEF) stations during most of the encounter phase, except for a 30-hr period at closest approach. Continuous tracking by the 70-m stations is planned only during this critical period.

The Encounter Phase begins slowly and builds to an extremely fast pace centered around closest encounter. It is divided into four subphases: Far Encounter, Near Encounter, Close Encounter, and Post-encounter. Far Encounter involves acquisition of comet and coma science data. Near Encounter is the terminal guidance phase, and its science emphasizes high-resolution images of the coma and near-nucleus activities. Close Encounter is the core science period of STARDUST, focused on collecting samples and imaging the nucleus. Post-encounter is dedicated to assessing mission performance and downlinking comet images.

Figure 7 shows the timeline of key activities from E-100 d up to closest encounter. This figure also provides the resolution of images and sizes of the coma and the nucleus in the field of view (FOV) of the camera as a function of time.

4.3.4 Far Encounter Subphase (E-190 d to E-1 d)OPNAV will begin at about E-150 d when Wild 2 becomes detectable. The coma will be the focus of the imaging science during this period. Coma images acquired during this period will have resolutions of 32 to 6000 km per pixel. All eight filters will be used at each imaging episode and will be sent back at designated OPNAV telemetry time. Approximately thirty 4-hr passes of downlink time will be available during this period. At 1 kbps (50% link capability, 34-m dish), a data

volume amounting to 75 frames of 2:1 compressed images may be sent back. More can be accomplished by combining the onboard "windowing" process. This, in essence, offers an opportunity to obtain full color movies of the evolving coma. At E-1 d, the coma image begins to fill the FOV of the camera, and the focus of the imaging will be on the finer details.

4.3.5 Near Encounter Subphase (E-1 d to E-5 hr)STARDUST enters the terminal navigation phase with increased OPNAV activities. Continuous communication with Earth (70-m stations) will be established. At E-1 d the OPNAV picture rate will be increased to one per hour. All data acquired since the previous TCM (E-2 d) will be processed on the ground as each image is received for image location extraction, orbit determination, and the final TCM computation. We expect to obtain finer details of the coma when we image Wild 2 during this period. The Wild 2 nucleus will still be a pinpoint until the end of this phase when it begins to occupy about a pixel. Assuming a 50% link capability of the spacecraft, a real-time data volume transmission of 34 image frames with 2:1 compression is possible. Full-color images of Wild 2 with resolutions ranging from 5 to 32 km per pixel will be obtained during this period.

4.3.6 Close Encounter Subphase (E-5 hr to E+5 hr)This is the core science period of the mission. At E-5 hr the spacecraft will begin to enter the coma (100,000 km from Wild 2) and the nucleus will start to emerge as an extended body in the camera FOV. All comet science will be on. Continuous tracking of the spacecraft with the 70-m station is planned until the end of this mission subphase.

Dust collection will begin with the deployment of the dust collector after the last TCM at E-6 hr. The spacecraft dust shield and the collector array will orient perpendicular to the dust stream (spacecraft-comet relative velocity) to protect the spacecraft from the dust hazard while maximizing the collection area.

CIDA will provide information on comet particle composition during the fly-through. Data from up to 10,000 CIDA events will be compressed and stored on board. The data volume allocated is about 200 Mbits.

Continuous imaging and real-time transmission of data will be made from E-5 hr to E-4 min and again from E+4 min to E+5 hr. At E-4 min when the nucleus occupies 60 60 pixels, a final black and white picture surrounding the nucleus will be sent in real time. This will take no longer than 27 s. Any images taken after E-4 min will be stored on board. Figure 8 shows details of mission activities occurring from E-5 min to E+5 min. Due to the uncertainty in delivery, the image of the nucleus may spill out of the FOV of the camera beginning at about E-2 min. Although the scanning mirror can compensate for down-track and in-plane errors, only banking the spacecraft (by providing the second axis to the mirror) can correct out-of-plane errors. Because of this, temporary loss of high-gain lock to Earth during the ±3 min of the encounter is expected. The medium gain antenna will take over the critical communications function during this time.

4.3.7 Post-Encounter Subphase (E+5 hr to E+50 d)Post-encounter spacecraft health check, reconstruction of flyby conditions and downlink of recorded data will constitute the activities of this mission phase. DSN tracking similar to cruise-phase mode will resume.

4.4 Interstellar Dust Collection Phase Design

4.4.1 Interstellar Grain Impact ProfileBased on recent studies [3], IGs are assumed to enter the heliosphere with a velocity of 30 km/s from the upstream direction of 10_±10_, 280_±30_ ecliptic latitude and longitude. The flight paths of the IGs are modified by solar gravity, solar pressure, electromagnetic interaction with the interplanetary magnetic field, and various other complex processes not well or easily formulated. If one considers only the simple effects of solar gravity and solar pressure, the velocities of IGs of various sizes can be calculated as a function of , where is the ratio of solar pressure to solar gravity.

4.4.2 Interstellar Grain Collection StrategyThe strategy of IG collection is (1) to collect at the part of the spacecraft orbit where IG impact velocity is relatively low (<15 km/s), (2) to orient the collector in a specific direction so that the area for the desired IGs is maximized and the IG tracks indicating normal incidence may be tagged as the desired particle, and (3) to avoid pointing toward the sun in order not to intercept particles of interplanetary origin. Total duration of IG collection will be about two years.

Mission operation during the IG collection period is similar to cruise phase due the passive nature of the collector design. Although the IG collector will need to be steered in specific directions to maximize the area for intercepting desired IGs, tight attitude control is not required because the uncertainty in the IG radiant direction may be as large as 30_.

4.5 Sample Earth Return Phase Design

This phase of the STARDUST mission begins two weeks before Earth re-entry and ends when the SRC is transferred to its ground-handling team. The planned landing site is the Utah Test and Training Range (UTTR) as shown in Figure 9. There are two optional landing zones accessible by targeting the approach trajectory for either a posigrade position with respect to Earth or retrograde.

Figure 9. UTTR Footprints

Following touchdown, the SRC will be recovered by helicopter or ground vehicles and transported to a staging area at UTTR for retrieval of the sample canister. The canister will then be transported to the planetary materials curatorial facility at Johnson Space Center. The Earth Return is divided into four subphases: Earth Approach, Entry, Terminal Descent, and Recovery.

4.5.1 Earth Approach SubphaseEarth Approach begins with an increased tracking frequency of one 8-hour pass per day. During this period three TCMs are involved: at ER (Earth reentry) -13 d, ER-3 d and ER-3 hr. The SRC will be released soon after the last TCM and will enter the atmosphere at a nominal entry angle of -8_. The approach velocity to Earth will be approximately 6.4 km/s with a right ascension of 205.7_, a declination of 11.1_, and velocity at entry (assumed to be at an altitude of 125 km) of 12.8 km/s. The entry corridor control accuracy (3 ) attainable, based on the Navigation Plan, is 0.08_.

The spacecraft will perform a divert maneuver subsequent to the SRC release to avoid entering the atmosphere.

4.5.2 Entry SubphaseEntry begins when the spacecraft reorients for SRC release from the spacecraft bus and ends with parachute deployment. The SRC will be released from the spacecraft bus approximately 3 hours before entry. Significant activities during these 3 hours include slewing the spacecraft bus to the proper release attitude, settling and verifying spacecraft attitude, initiating the SRC on-board timer/sequencer, turning off spacecraft-bus-provided heater power to the SRC, and releasing the SRC.

The SRC will perform a direct entry at Earth. After entry the SRC will continue to free-fall until approximately 3 km, at which point the parachute deployment sequence will initiate. Elapsed time from entry to parachute deploy will be approximately 10 minutes.

4.5.3 Terminal Descent SubphaseDescent begins when the parachute deployment sequence initiates and continues until the SRC/parachute system has descended into the recovery zone, the UTTR.

The velocity of the SRC must be reduced from the initial entry velocity of 12.8 km/s to a level that permits soft landing.

The aeroshell removes over 99% of the initial kinetic energy of the vehicle to protect the sample canister against the resultant extreme aerodynamic heating. The heatshield is a 60_ half-angle blunt cone made of a graphite/epoxy composite covered with a thermal protection system. Ablative material on the backshell protects the lander from the effects of recirculation flow around the entry vehicle.

Taking into account SRC release and entry corridor uncertainties, vehicle aerodynamics uncertainties and atmospheric dispersions, the landing footprint ellipse for the SRC has been determined to be approximately 60 km by 6.5 km. The SRC will approach the UTTR on a heading of approximately 122_ on a north-west to south-east trajectory. Local time of landing will be approximately 3:00 am.

4.5.4 Recovery SubphaseRecovery begins a few hours before the SRC touches down. Retrieval is via ground transportation or helicopter.

Given the small size and mass of the SRC, it is not expected that its recovery and transportation will require extraordinary handling measures or hardware other than a specialized handling fixture to cradle the capsule during transport.

Transportation of the SRC to a staging area at the UTTR for extraction of the sample canister will follow. The sample canister then will be transported to its final destination, the planetary material curatorial facility at Johnson Space Center.

The STARDUST flight system is composed of the SRC and the spacecraft. Each employs elements of advanced technology in concert with flight-proven components to produce a cost-effective, lightweight spacecraft capable of operating reliably in deep space for long-duration missions. Designed from the ground up with cost and mass-efficiency in mind, the STARDUST flight system represents a new wave of small, lightweight spacecraft. Although STARDUST is representative of the new approach to faster, better, cheaper, it still embodies LMA's total commitment to mission success. STARDUST has been designed to eliminate all credible single-point failures from the system.

5.1 STARDUST Spacecraft

Propulsion on STARDUST depends on a single, simple, blowdown hydrazine system. The mission has been designed with minimal V requirements and very loose attitude-control requirements for the bulk of the mission. Therefore a single-tank monopropellant system is adequate to meet the propulsion requirements of STARDUST.

The telecommunications system on STARDUST consists of fully redundant X-band deep-space transponders, solid-state power amplifiers, and associated filters, couplers, switches, and waveguides. During comet encounter, the period of highest demand on telecommunications data rate, the geometry between comet, spacecraft, and Earth do not change significantly. Exploiting this fact, the STARDUST high-gain antenna is fixed-mounted without a gimbal mechanism, thereby saving cost, complexity, and mass. During the remainder of the mission, when the attitude of the spacecraft is not held within tight deadbands, communications are through a medium-gain antenna. Low-gain patch antennas are also integrated into the telecommunications system for use during the initialization and checkout phase of the mission and, if necessary, during safing modes.

Power for STARDUST (some 260 W at encounter) is provided by two solar arrays each covering 3 m². Each solar array is also fixed-mounted to the bus. Power conditioning, control, and distribution functions are provided by advanced avionics cards developed by LMA for deep space missions. Commonality of the electrical power system among several ongoing spacecraft programs at LMA provides a cost-effective strategy for implementing state-of-the-art avionics. The SMTS evolutionary test plan will concurrently integrate analysis, design, and testing to achieve high confidence in mission success.

Command and Data Handling (C&DH) for STARDUST is essentially inherited, but has been upgraded to include advanced circuitry now being developed for a number of ongoing programs at LMA. The central processor card is a RAD6000, an off-spring of the Mars Pathfinder Project, with 1 Gbit of on-card solid-state data-storage capacity. Interfaced to the processor card through a VME bus are the data I/O cards, payload interface cards, telecomm interface cards, and the A/B arbitration card, all with design heritage from previous programs such as Mars Pathfinder, Mars Surveyor '98, and Cassini. The C&DH system is block-redundant.

Attitude Determination and Control on STARDUST is accomplished with redundant ring-laser gyros, dual star cameras, dual analog sun sensors, and the OPNAV camera. As with the electrical power system and C&DH system, the attitude-control system design is in common with several ongoing programs at LMA, thereby enabling a large degree of synergy, shared risk mitigation, and cost efficiency.

5.2 Sample Return Capsule

Like the spacecraft structure, the SRC structure is also based on lightweight, advanced composites. STARDUST uses entry capsule design techniques developed by LMA for the Mars Pathfinder program to keep the structural mass of the entry capsule low. Within the composite entry probe is a sample canister which houses the aerogel grids and the mechanisms used to extend and retract them.

The project has successfully completed both Phase A and Phase B activities culminating in a confirmation to proceed into the full development Phases C/D. The confirmation was in October of 1996. Four elements were crucial in achieving confirmation. First, was the strong teambuilding that occurred through
use of the concurrent engineering architecture described earlier. Communications, data-sharing and response (decisions) to the many trades and options in the Phases A and B were rapid and team-generated. Second was a strong focus on implementing a performance measurement system (PMS) that has driven the many activities toward a proper partitioning for the "earned-value" metrics needed to understand progress, liens and cost controls. Third, it was possible to rephase, or "prefer," the required Phase C/D funding to achieve common-buys with the "sister" project Mars Surveyor '98 at LMA and to negotiate favorable subcontracts at fixed-prices early. Fourth, a strong inheritance of design elements in hardware and software has been achieved with Mars Pathfinder, Mars Global Surveyor, Mars Surveyor '98, and Cassini. It was also very important to have actual flight hardware in the case of the residual optics from the Voyager Camera, along with a flight-proven experiment provided by Germany, the CIDA.

While it remains, as usual, that mass and power margins are not as robust as desired, the preferral funding has provided a significant reduction in uncertainties, particularly in avionic components.

The STARDUST Project is committed to manage-to-budget by organizing through the TQM principle of concurrent engineering. The team esprit brings the spirit of "ownership of mission success" to every level in the project. The project engineering and integration team operates with a suite of high-tech communication and software tools to gain visibility and rapid-response involvement at the appropriate solution-point.

STARDUST's mission plan is set to return cometary and interstellar material to Earth where literally hundreds of experimenters can participate directly. This is a major paradigm-shift in deep-space exploration away from traditionally large expenditures to design a few experiments into severely constrained mass, power, and volume limits for delivery to the target environment. The focus on sample-return makes STARDUST a relative "bargain" at a price near 10% of the traditional approach. STARDUST will be faster, better, and cheaper by doing more with less. STARDUST's success will validate the Discovery Program goal of changing the way NASA does business by effecting a complete culture change for planetary exploration and space science.

This work was performed at the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.

[1] Y. A. Attia, M. S. Ahmed and M. Zhu, 4th Int. Aerogel Symp. (1994).

[2] M. Baguhl, D.P. Hamilton, E. Gr¸n, S.F. Dermott, H. Fechtig, M.S. Hanner, J. Kissel, B.-A. Lindblad, D. Linkert, G. Linkert, I. Mann, J.A.M. McDonnell, G.E. Morfill, C. Polanskey, R. Riemann, G. Schwehm, P. Staubach and H.A. Zook, Dust Measurements at High Ecliptic Latitudes, Science 268 1016-1019 (1995).

[3] D.E. Brownlee, P. Tsou, K.L. Atkins, C.-W. Yen, J.M. Vellinga, S. Price and B.C. Clark, STARDUST: Finessing Expensive Cometary Sample Return Missions, IAA International Conference on Low-Cost Planetary Missions, Laurel MD, April (1996).

[4] T. Bunch and S. Chang, Geo Chemica and Cosmic Chemica Acta, 44, 1543-1577 (1980).

[5] B. C. Clark, Primeval Procreative Comet Pond, Origins of Life 18, 209-238 (1988).

[6] B. C. Clark, Comets, Volcanism, the Salt-Rich Regolith, and Cycling of Volatiles on Mars, Icarus 71, 250-256 (1988).

[7] J. Mayo Greenberg and J. I. Hage, From Interstellar Dust to Comets: a Unification of Observational Constraints, Astrophys. J. 361, 260-274 (1990).

[8] J.A. Simpson et al., Astron. Astrophys.. 187, 742 (1987).

[9] J.A. Simpson et al., Nucl. Instr. & Methods. A279, 611 (1989).

[10] J.A. Simpson and A.J. Tuzzolino, Nucl. Instr. & Methods. A279, 625 (1989).

[11] P. Tsou, T. Bernatowicz, D. Burnett, A. Chutjian, P. Eberhardt, R. Mawhorter, M. Neugegbauer and A. L. Albee, 18th Lunar and Planetary Science Conference, (1987).

[12] P. Tsou, D. E. Brownlee and A. L. Albee, 24th Lunar and Planetary Science Conference, (1993).

[13] P. Tsou, D. J. Griffiths and A. L. Albee, LPI Report 93, (1993).

Ken Atkins joined STARDUST as Project Manager in June 1995 during its Phase A competition for selection as the Discovery 4 mission. He first worked at JPL as a member of the technical staff in propulsion and mission analysis focusing on small body missions, then managed U.S. options on missions to Halley's comet in the late '70s. He next managed the JPL Power Systems Section which, under his leadership, successfully delivered the Galileo power subsystem.

He later managed the Flight Command and Data Systems section focusing on deliveries of the Cassini command and data subsystem, development of flight software for the low-gain Galileo mission, and flight operations for Voyager, Galileo, and Mars Observer. He also managed the integrated avionics development for the Mars Pathfinder Discovery mission. He has a doctorate in aeronautical and astronautical engineering from the University of Illinois.