text
stringlengths
0
13.6k
The selected model philosophies are important in terms of both risk and cost and, as a maximum, a protoflight approach is acceptable for such programmes.
The advantages of small satellites are as follows:
(a) Orbital parameters optimized to individual instrument requirements;
(b) Augmentation of conventional satellite programmes, such as additional capability, redundancy for critical missions, or replacement of a failed instrument;
(c) Missions with limited lifetime and/or coverage requirements;
(d) Improved responsiveness to the end-user (more frequent launch opportunities and increased mission flexibility for individual instruments, plus schedule independence);
(e) Quick-reaction, launch-on-demand launches using low-cost dedicated vehicles (e.g. crisis monitoring, replacement after an in-orbit failure, or monitoring of unexpected environmental conditions);
(f) Relaxed reliability owing to the shorter lifetime or agreed, lower levels of product assurance or lower-quality parts as appropriate to lower development costs;
(g) Reduced satellite design complexity (e.g. simplified interfaces, optimized for instrument requirements), shorter development schedule and suitable test-bed for proving technique and/or technology.
There are three general classes of orbit that may be suitable for small satellites: the geostationary orbit (GSO), the highly elliptical orbit (HEO) and the low Earth orbit (LEO).
GSO is where the satellite appears fixed relative to the ground, thus allowing continuous visibility and simplifying the ground segment and operational requirements.
Because of the large space-to-ground distance involved, however, the data rates are small, or larger ground antennas and higher electrical power on board the spacecraft are required.
This orbit is usually reached from a standard geostationary transfer orbit (GTO) provided by a large launch vehicle.
The use of GTO is interesting as it could benefit from frequent piggyback launch opportunities while avoiding the complexity and extra costs associated with an apogee propulsion system.
LEO is generally preferred for small satellite missions.
Small launch vehicles may be used, offering flexibility in the selection of the orbit parameters; piggyback launches may also be used.
A low-energy transmitter on board is sufficient because of the short distance from the ground, but infrequent and short visibility periods are a drawback, leading to some ground-segment and operational complexity.
A distinction should be made between the near equatorial or low inclination orbits, for which the visibility zone will be limited to the topical zone, and the polar and quasi-polar (Sun-synchronous) orbits that allow accessibility to any point on Earth, either for communication (e.g. store and forward) or for remote sensing of Earth.
The current and future development of small satellites is closely linked to the appearance of new, low-cost launchers (Pegasus, Taurus etc.) and lower-cost launch opportunities on existing vehicles (for example, Ariane-4 or small canisters on the Space Shuttle).
The potential availability of cheap launchers has spurred much of the recent surge of interest in small satellites, which was initially driven largely by defence and global civil communication programmes of the United States of America.
Of the major low-cost launchers of European countries and the United States, only Pegasus and Taurus are flight-proven.
Conestoga is planned for flight in the near future, development of the Italian San Marco Scout has not yet started (although its forerunner, the United States Scout, has been operated for many years) and the Ariane-5 derivative programme should be completed in 1999.
In order to maximize their potential, small launcher developers must apply the same innovative, low-cost design approach used on small satellites.
Launch costs represent a large portion of total programme costs (generally over 25 per cent) and satellite mass and size must therefore be constrained to take full advantage of low-cost launch opportunities.
Options include:
(a) Small dedicated launchers;
(b) Multiple launch of several small satellites, nominally on Ariane-4 or Ariane-5 for European missions (e.g. Cluster scientific satellites of the European Space Agency (ESA));
(c) Flight opportunities on larger launchers (Arianespace has strongly promoted its vehicles for this purpose, with Ariane-4 offering: Ariane Structure for Auxiliary Payloads (ASAP) for piggyback microsatellites (up to six 50 kg satellites (300 kg total)); Ariane Radioamateur Satellite Enseignement Espace (ARSENE)-type configuration for satellites of up to 200 kg; and SPELDA Dedicated Satellite (SDS) for satellites in the range of 400-800 kg within the short Ariane Dual Launch External Bearing Structure (SPELDA) adapter.
Similar options are provided by other medium-sized or large launchers, for example, the United States Atlas Centaur and the Delta-2.
Requirements for the ground segment of a small satellite system vary enormously depending on the application area.
At one extreme, low data rate sensors with only local or regional coverage on missions with low tracking and command requirements impose relatively low demands on the ground segment, possibly comprising only 10 per cent or less of total programme costs.
More complex data retrieval and processing requirements could result in ground segment costs of up to 50 per cent of total programme costs.
Assuming that ground segment costs tend to average 25 per cent of total programme costs, it is clearly important to identify potential savings in the ground segment in concert with those of the space segment.
In attempting to reduce the ground segment costs, there are limits on simplification because it is still necessary to ensure the achievement of capabilities such as reliable operation, rapid response to critical commands and regular form of data on time and with low loss. The ground segment model, which forms the basis of any technical and cost assessments, must include not just the provision of the ground stations, but also ground communications infrastructure, mission control etc.
Very small, in some cases transportable, stations have been made commercially available by several suppliers in European countries and in the United States.
Commercial providers of remote sensing data are likely to press for the adoption of such approaches in attempts to reduce the cost of distributing and processing the data.
Space science activities are obviously valuable, and most space-faring nations have begun their involvement in this field with small scientific satellites.
A university environment is often an ideal environment for the development of space activities and, because such projects often require the creation of new laboratories, these facilities are a lasting, beneficial by-product of such projects.
Thus the usual spin-offs of a space programme, the acquisition of technology and the development of industrial organization and management methods, will begin to accumulate at the national level as students leave the university and enter local industry.
The first small scientific satellite of Argentina will be the Scientific Applications Satellite (SAC-B), which is being developed jointly by its national space agency, the National Commission for Space Activities (CONAE), and the National Aeronautics and Space Administration (NASA) of the United States.
The 190 kg satellite is to be launched in 1996 by a Pegasus rocket into a circular orbit of 550 km with an inclination of 37 degrees.
SAC-B will be inertially stabilized and permanently oriented to the Sun.
It will monitor energetic X-rays from solar flares and survey the sky with X-ray charge-coupled device (CCD) sensors along an axis perpendicular to the Sun line.
Between 1978 and 1991, scientific microsatellites weighing 15-50 kg were developed for the Magnetosphere-Ionosphere (MAGION) research programme in the former Czechoslovakia.
MAGION-1 was launched on 24 October 1978 as a subsatellite of the INTERCOSMOS-18 geophysical satellite.
Although it was designed for an operational lifetime of three weeks, MAGION-1 remained operational for three years.
MAGION-2 and MAGION-3 were launched into high-inclination low-eccentricity orbits (with an altitude of 500-3,200 km) as part of the ACTIVE and Ariane Passengers Experiment (APEX) mother-daughter active space missions launched on 28 September 1989 and on 18 December 1991, respectively.
The MAGION-4 subsatellite was successfully launched by a Molniya launcher from the cosmodrome at Plesetsk, Russian Federation, on 3 August 1995, as part of the INTERBALL-tail mission.
MAGION-5 is currently scheduled for a 1996 launch.
The Central European Satellite for Advanced Research (CESAR) is a spacecraft of about 300 kg that will fly in an orbit with a perigee of 400 km, an apogee of 1,000 km and an inclination of 70 degrees.
The scientific mission is related to the study of the magnetosphere, ionosphere and thermosphere of Earth.
Ten different experiments, provided by scientists from Austria, Czech Republic, Hungary, Poland and Slovakia, will be accommodated on the spacecraft, which is being funded by the Italian Space Agency (ASI) and designed by Alenia Spazio.
This mission is one of the objectives of the cooperation among the countries of the Central European Initiative.
Space industry and research institutions in Finland have acquired experience in satellite payloads and instrumentation through their associate membership in ESA and have long been active in remote sensing and other space-related disciplines.
To initiate a study of a Finnish small satellite (FS-1), Finnish institutes defined their interests by contacting selected institutes with an unofficial announcement of opportunity and by asking for proposals.
After the proposal phase, system design was performed for two alternatives: a scientific satellite and an Earth observation satellite.
Each satellite would contain a technology demonstration package in which new electronic components would be tested in the space environment.
The French National Centre for Space Studies (CNES) is currently considering the following small scientific satellites:
(a) Mission SAMBA: registration of the local fluctuations of 3 kelvin radiation from the big bang (similar to the Cosmic Background Explorer (COBE) satellite of the United States) and detailed measurement of possible anisotropies;
(b) Mission COROT: astroseismology, new data on the convection and internal rotation of stars by long-term measurement of stellar oscillations;
(c) Mission IBIZA: registration of the plasma accelerated in the geomagnetic auroral regions, interaction of ionized particles with the ionosphere and magnetosphere of Earth, creation of electromagnetic disturbances;
(d) QUICK-STEP: verification of the equivalence of the inertial and gravitational mass (theory of relativity) with a relative precision of 10-17.
On 3 February 1994, the small satellite of the University of Bremen, BREMSAT, was carried into orbit by the United States Space Shuttle Discovery.
The spacecraft, which weighed 63 kg, waited six days in its Get Away Special (GAS) container before it was deployed into its initial circular orbit of 350 km.
The satellite carried six experiments with different scientific objectives, including heat conductivity under microgravity, micrometeorite and dust particle distribution, atmospheric atomic oxygen mapping and re-entry pressures and temperatures.
The satellite functioned until its in-orbit decay on 12 February 1995.
In developing its indigenous launching capacity, India prepared a series of small technology development and scientific satellites called Rohini and the Stretched Rohini Scientific Satellite (SROSS) series.
The Rohini satellites were launched between 1980 and 1983 and carried a landmark sensor payload including a solid state camera.
More than 2,500 frames in both visible and infrared bands for identification of landmarks and altitude and orbit refinement were obtained. The orbital mass of the Rohini satellites was about 42 kg.
SROSS-C and SROSS-C2 were launched on 20 May 1992 and 4 May 1994, respectively.
They each carry two scientific payloads.
The first is the retarding potential analyser, consisting of two planar detectors to measure plasma parameters and to investigate the energetic structure of the equatorial ionosphere.
The second is the gamma-ray burst experiment, consisting of two scintillation detectors to study celestial gamma-ray bursts in the energy range of 20-3,000 kilo-electronvolts.
The Instituto Nacional de Tecnica Aeroespacial (INTA) of Spain, situated at Torrejón de Ardoz, has been entrusted by the Government with the direction of a research project for the development of a Spanish space system, called MINISAT.
The system will consist of a multi-purpose platform (service module), payload module and an associated ground segment.
Both the platform and the subsystems comprising it are modular in character.
The platform will be capable of receiving, integrating, operating and carrying on board a payload module by means of standard interfaces.
This will permit all the required adaptations for a particular mission to be easily conducted.
The platform will be able to carry payloads with masses that vary between 80 and 500 kg.
The first of these satellites will be a MINISAT mission carrying a payload module (PLM-1).
The first Swedish-built satellite was the 283 kg Viking, which was launched into low polar orbit in 1986 in a piggyback configuration with the French remote sensing satellite, satellite pour l ' observation de la Terre (SPOT).
The scientific objective of the Viking satellite was to study ionospheric and magnetospheric phenomena at high geomagnetic latitudes in the altitude region of up to about two Earth radii.
Simultaneous measurements were made of electric and magnetic fields, particle distributions, plasma composition and waves, as well as imaging in the ultraviolet of the aurora variations.
A more advanced small scientific satellite, called Freja, was launched on 6 October 1992 by a Chinese launcher.
This satellite of 214 kg is designed for auroral research and other related magnetospheric phenomena.
Because Swedish magnetospheric scientists are very interested in the possibilities of small satellites, a compact satellite platform has been developed that is 10 per cent of the mass of the Freja.
This new microsatellite, called Astrid, is shaped like a box with sides measuring approximately 50 x 50 cm and has a mass of 25 kg in its stowed configuration.
It is spin-stabilized with a Sun-pointing capability and deployable solar panels.
The first Astrid was launched on 24 January 1995 from the cosmodrome at Plesetsk on a Cosmos launcher.
The best example of the small satellite programmes in the United States is the NASA Small Explorer (SMEX) programme, which provides for frequent flight opportunities for highly focused and relatively inexpensive science missions.
Each SMEX spacecraft weighs approximately 250 kg and each mission is expected to cost approximately US$ 50 million for design, development and 30 days of in-orbit operations.
The first satellite of this series, the Solar, Anomalous and Magnetospheric Particle Explorer (SAMPEX), was launched on 3 July 1992.
It has been successfully investigating the composition of local interstellar matter and solar material and the transport of magnetospheric charged particles into the atmosphere of Earth.
The Submillimeter-Wave Astronomy Satellite (SWAS) is to be launched on a Pegasus rocket in 1995 or 1996.
Probably the most experienced research unit in the field of microsatellites is the Spacecraft Engineering Research Unit at the University of Surrey in the United Kingdom of Great Britain and Northern Ireland.
Since 1981, UOSAT and, more recently, the Surrey Satellite Technology Limited (SSTL) team have logged over 25 orbit-years of microsatellite operations.
A total of 10 University of Surrey Satellite Project (UOSAT) satellites were launched between 1981 and 1993.
The operational microsatellite S-80/T, which is based upon the UOSAT platform, was launched in August 1992 with a mission objective to explore communication possibilities of the very high frequency World Administrative Radio Conference (WARC-92) bands allocated to non-geostationary satellite systems.
The initial mission objective has been successfully achieved.
S-80/T completed its first operational year in October 1993 while continuing its flawless functioning.