Minimize SNAP-1

SNAP-1 (Surrey Nanosatellite Applications Program)

SNAP-1 is a project of SSTL (Surrey Satellite Technology Ltd.), a spin-off company of the University of Surrey, UK, and the Surrey Space Center, started in 1998. The objective is to develop a modular, multi-mission nanosatellite bus (mass range of 1-10 kg), to demonstrate the use of miniature electrical and mechanical COTS (Commercial-Off-The-Shelf) product technologies in space and their use as autonomous robots for observing orbiting space vehicles. Future applications of nanosatellites are seen in remote inspection of satellites and monitoring of deployment systems in orbit, and carrying small space science instruments requiring measurements with spatial diversity. 1) 2) 3) 4)

SNAP1_Auto12

Figure 1: Illustration of the SNAP-1 nanosatellite (image credit: SSTL)

Spacecraft:

SNAP-1 was designed and built as a low-cost research mission by a joint academic-commercial team at the Surrey Space Center and at SSTL, funded entirely by SSTL. The objective of SNAP-1 is to demonstrate the feasibility of a standardized modular nanosatellite bus.

• Provide a test-bed for novel microelectronic technologies - in particular a new GPS navigation system, APS camera technologies and RISC processors. The intent is to use SNAP-1 as a “remote inspector demonstrator.”

• Provide experimental data and imagery to the radio-amateur/amateur-scientific communities

• Provide a vehicle for the education and training of students in spacecraft engineering at undergraduate and post-graduate level

• The SNAP program is also intended to demonstrate the feasibility of using clusters of low-cost satellites that can fly in formation and conduct multipoint remote sensing.

The SNAP-1 nanosatellite is three-axis stabilized by a single Y-momentum wheel and magnetorquers as actuators (the magnetorquers are for nutation damping and wheel momentum management). The S/C attitude is sensed by a three-axis magnetometer and sun sensors. A GPS receiver (model: SGR-05 of SSTL) is used for autonomous orbit determination (onboard navigation parameters and timing). The best pointing accuracy achieved is < 1º. The SGR-05 performs differential orbit determinations in conjunction with Tsinghua-1.

The spacecraft bus features a modular design based on a tray structure (three 'nanotrays') formed around a triangular central bay. This central bay can host attitude actuators, propulsion units and payloads. The SNAP-1 primary structure consists mainly of the aluminum-alloy electronics module boxes. SNAP-1 uses three stacks of three modules arranged in a triangle. The end facets of the structure are closed with aluminum honeycomb panels, and four additional honeycomb panels are used to support the solar cells. The thermal control is passive, and is achieved via the appropriate thermo-optical tapes (first-surface and second-surface Kapton/ vacuum-deposited aluminum mirrors) being applied to the boxes and honeycomb panels.

A micro-propulsion system (liquified gas) is used for orbit maneuvers (rendezvous with other S/C).

EPS (Electrical Power Subsystem): Primary power to the satellite is supplied via 4 solar panels. The power from each of the four solar panels feed into a dedicated BCR (Battery Charge Regulator), i.e. one BCR per solar panel. The output of the BCRs is connected to a 6 cell, 1.4Ah NiCd battery, the PDM (Power Distribution Module) input and the PCM (Power Conditioning) input. 5)

The power system uses only a single string of systems, resulting in no redundancy. Although, a failure of a BCR can result in a graceful degradation of orbit average solar array power, a failure in the PCM, PDM or battery will result in the loss of the spacecraft. This same redundancy philosophy has been used for all of the SNAP subsystems.

The available power from the solar panels depends on the spacecraft orientation (spin, inertial, nadir or sun pointing). A spacecraft pointing constraint, based on the solar aspect angle, is observed in operations where the spacecraft -z-axis (the nadir face) is maintained at an angle greater than 30° from the sun-vector for any period longer than a small fraction of the orbit period. The solar arrays could be reconfigured on a mission-by-mission basis to tailor the power profile or enhance the power capability.

SNAP1_Auto11

Figure 2: Block diagram of the EPS (image credit: SSTL)

The body-mounted solar cells provide an average power of 6.5 W each. SNAP-1 has 4 W of average orbit power and 9.1 W of peak power. A six cell NiCd battery (Sanyo KR-1400AE) provides eclipse-phase energy of 1.4 Ah.

SNAP1_Auto10

Figure 3: Fully assembled SNAP-1 EPS (image credit: SSTL)

The S/C mass is 6.5 kg (the all up mass is 8.3 kg with about 1.8 kg for the attach fitting for mating onto the carrying vehicle). 6) 7) 8)

SNAP1_AutoF

Figure 4: Illustration of the internal tray structure of SNAP-1 (image credit: SSTL)

Background: In the design approach for SNAP-1, a simple standard electrical interface was prescribed for each module, consisting of a regulated 5 V and raw battery power connections, with a single bi-directional CAN (Controller Area Network) bus for data transfer. These connections are provided with a 9-way D-type connector, which is standard to all modules. An additional 44-way D-type connector was allowed for each module to provide for specific point-to-point connections. All modules used a standard CAN bus for onboard data handling. - A standard module box mechanical format was also defined at the beginning of the SNAP program. Thus, every module of SNAP-1 has the same external dimensions, sized to house a standard “Eurocard” printed circuit board (160 mm x 100 mm with about 13 mm of useable depth).

Parameter/name

Magnetometer

GPS

Magnetorquer rods

Momentum wheel

ACDS module

Manufacturer

Billingsley

SSTL

SSTL

SSTL

SSTL

Quantity

1

1

3

1

1

Type

Fluxgate

Mitel chipset

Nickel-alloy
core

Brushless DC
motor

C515 CAN
µcontroller

Range

±60 µTesla

12 channel
1 antenna

±0.122 Am2

0-5000 rpm
0-0.01 Nms

-

Resolution/
accuracy

±60 nT


< 15 m

10 ms min.
pulse

±5 rpm

-

Mass (gram)

117

43

36 each

80

360

Size (mm)

35 x 32 x 83

95 x 50 x 8

125 x 5 diameter

40 x 47 diameter

168 x 122 x 20

Power (mW)

150

1700

100

100-500

250

Table 1: Specification of the ADCS sensors and actuators

SNAP1_AutoE

Figure 5: Illustration of the ADCS hardware (image credit: SSTL)

S/C mass

6.5 kg (1.2 kg of payload mass)

S/C size

175 mm x 200 mm (diameter) hexagonal prism

Solar panels

Each panel 7 x 8 cells (40 mm x 40 mm GaAs)

S/C power

2.5 W at 5.0 V regulated, 6 cell NiCd (an advanced form of NiCd)

RF communications

VHF uplink, 750 mW, S-band downlink

Onboard data handling

Asynchronous uplink (9.6 kbit/s), downlink (76.8 or 38.4 kbit/s selectable)

Data rate

Uplink: 9600 bit/s, FSK modulated Downlink: 38.4 kbit/s nominal, 76.8 kbit/s max; BPSK & QPSK modulation, convolutional encoding on QPSK

On-Board Computer (OBC)

A 32-bit StrongARM OBC includes 16 MB FLASH and 4 MB of EDAC (Error Detection And Correction) protected code memory. CAN-bus onboard data handling network

ADCS (Attitude Determination and Control Subsystem)

Navigation, magnetometer, sun sensors, magnetorquers. GPS receiver accuracy <15 m. Attitude is estimated using a Kalman filter

Payloads

- SGR-05 based on the Orion GPS receiver of MITEL Semiconductors,
- APS camera (referred to as MVS)

Table 2: Specification of the SNAP-1 satellite

SNAP1_AutoD

Figure 6: The architecture of the EDAC subsystem used on the OBC and MVS (image credit: SSTL)

EDAC (Error Detection And Correction) technique: Up until the SNAP program, some SSTL computer systems had used the TMR (Triple Modular Redundancy) method to protect their volatile RAM from SEU corruption. Under this technique, when a processor writes data into the RAM, EDAC hardware between the processor and the RAM writes three identical copies of the data into three independent RAM banks. Then when the data is read back the EDAC hardware reads all three copies and performs majority voting between them. Therefore if the data from one of the banks is corrupted, the other two banks will out vote it and the correct data will be returned to the processor.

To minimize the mass on the nanosatellite, a new design approach was taken on the OBC and MVS (Machine Vision System) developed at SSTL. This technology is known as 16:8 EDAC. In this method, for every 8 bits that are written to RAM a further 8 bits of encoded parity data are also written into the RAM by the EDAC hardware. Using this technique it is possible to correct two errors in each 16 bit block of data and parity bits stored in the RAM. So its error correction ability is lower than TMR, but it has the advantage of having only a 100% overhead. The block diagram of the EDAC and RAM structure used on the SNAP OBC and MVS is shown in Figure 6. 9)

SNAP1_AutoC

Figure 7: Isometric drawing of the SNAP-1 spacecraft (image credit: SSTL)

 

Launch: The launch of SNAP-1, along with Tsinghua-1 of China (a microsatellite of 50 kg mass) as secondary payloads, took place on June 28, 2000 on a Russian Cosmos-3M launcher from the Plesetsk Cosmodrome, Russia. The primary payload on this launch was Nadezhda, a Russian COSPAS-S&RSAT (Search & Rescue Satellite) payload. 10)

RF communication: SNAP-1 carries a VHF receiver and an S-band downlink transmitter, as well as an intersatellite link, a UHF receiver, tuned to the downlink frequency of Tsinghua-1 (when within range, data can be transmitted from Tsinghua-1 to SNAP-1). The SNAP-1 S-band data rate is 38.4 kbit/s nominal or 76.8 kbit/s maximum. The S-band modulation is BPSK or QPSK (selectable), the convolutional encoding on QPSK is selectable. The uplink data rate is 9.6 kbit/s with FSK modulation. The mass of the S-band transmitter is 0.5 kg, the size is: 120 mm x 160 mm x 20 mm; DC power of 5.2 V @ 670 mA (330 mA idle). 11)

A VHF spread-spectrum communications payload is being tested for a commercial customer. The UHF inter-satellite link (ISL) is aimed at deriving relative position data via differential GPS to aid rendezvous maneuvers between SNAP-1 and Tsinghua-1.

Orbit: Sun-synchronous orbit, 700 km altitude, inclination=98º, period about 98.7 minutes.

SNAP1_AutoB

Figure 8: Artist's view of the SNAP-1 nanosatellite in orbit (image credit: SSTL)


 

Instrument complement: (MVS, MPS, SGR-05)

The module box for SNAP payloads was defined as being a box of dimensions 160 mm x 120 mm x 20mm.

 

MVS (Machine Vision System):

The objective of MVS is to enable SNAP-1 to act as a remote inspector. The intent is to obtain imagery of TV-quality of other S/C in the local vicinity of SNAP-1 and to relay this information to the ground. MVS, a CMOS video system with APS (Active Pixel Sensor) technology, consists of the following components: 12) 13) 14)

• Three wide-angle CMOS cameras (WAC, 3.6 mm), each with a 350 x 288 pixel detector, and each with a 90º FOV to cover an arc of 270º

• A single narrow-angle camera (NAC, 50 mm, detector of 350 x 288 pixels) co-aligned with the center wide-angle camera providing the capability of finer feature inspection. The NAC device also contained a near infra-red (NIR) pass filter to enhance the contrast between land and sea features on the Earth's surface

• Video digitizer circuitry to convert the composite camera video output into digital data

• A recording capability of 8 Mbit of 70 ns SRAM for code and image storage

• A processor of the type StrongARM 1100 (220 MHz) for video/image compression. The processor may also being used for optical navigation functions in support of target tracking or automated optical docking.

• A CAN (Control Area Network) bus interface for communication with the spacecraft.

MVS features a software digitizer design due to the lack of implementation time (six months from project start to launch). The video signals of the cameras are fed into an ADC (Analog Digital Converter) and recorded. The MVS processor scans the digital representation and extracts the video data. - MVS has performed to specifications. The images captured in flight show in detail the deployment sequence of both the SNAP-1 and the Tsinghua-1 satellites.

SNAP1_AutoA

Figure 9: Schematic layout of the cameras of the MVS modules on SNAP-1 (image credit: SSTL)

SNAP1_Auto9

Figure 10: Schematic layout of the MVS architecture (image credit: SSTL)

SNAP1_Auto8

Figure 11: Illustration of the MVS (image credit: SSTL)

 

MPS (Micro-Propulsion System):

MPS is an SSTL-funded technology development with the objective to demonstrate the capability of SNAP-1 as an inspection vehicle (rendezvous with Tsinghua-1). The MPS is a stand-alone instrument consisting of three boxes and interconnected in such a way as to form a triangular structure (140 mm side length). The thruster selected is a cold gas system from Polyflex Aerospace Ltd. with a maximum thrust of 100 mN at a chamber pressure of 4 bar (minimum impulse of < 1 mNs). It utilizes butane as propellant, operating in a cold gas mode. The propellant (32.6 grams) is stored in a formed titanium tube (coiled tube of 1.1 m in length with a volume of 65 cm3). The thruster valve is a solenoid-operated valve. The drive electronics are operated from a microcontroller, deriving its commands and feedback telemetry via a CAN (Control Area Network) interface. The microcontroller also controls the manifold heater, which ensures that the propellant is vaporized. MPS is rated as a 30 µN thruster with a delta-v capacity of 3 m/s. 15) 16) 17)

Initially, the propulsion system was intended to be used to maneuver SNAP-1 to rendezvous with Tsinghua-1 of the University of Beijing (sometime in Nov. 2000). However, the rendezvous did not occur in November. Still, MPS demonstrated the new nanosatellite's unique capability for controlled orbital maneuvers using its miniature propulsion system.

SNAP1_Auto7

Figure 12: MPS thruster and pipework (image credit: SSTL)

SNAP1_Auto6

Figure 13: Illustration of MPS device (image credit: SSTL)

 

SGR-05 (Space GPS Receiver-05):

SGR-05 is a technology demonstration device on SNAP-1. The instrument is essentially based on the Orion GPS GP2000 applications receiver of Mitel Semiconductors. SGR-05 comprises solely the core GPS engine (no features of the SGR-10/20 series). It uses a single antenna with 12 channels. The features of EDAC protection, CAN-bus support, and telemetry and command capability were disengaged in the trimmed-down version. The SGR-05 has a size of 100 mm x 60 mm x 15 mm and a mass of 50 g, the power consumption is 2 W. 18) 19) 20)

The SGR-05 was used to update the orbital elements. At the time is was the smallest spaceborne GPS receiver.

The SGR-05 is not a complete subsystem solution, but must be integrated into a host module. This enables, for example, the close integration of the GPS receiver with the ACS module and enables the absolute minimization of resources. The host is responsible for mechanical housing, EMC, interfaces, power supply, although SSTL is able to provide practical integration advice.

SNAP1_Auto5

Figure 14: Block diagram of SGR-05 with typical host application (image credit: SSTL)

SNAP1_Auto4

Figure 15: The GNC module (ACS+GPS) with the SGR-05 circuitry at the right half of the image (image credit: SSTL)


 

Discussion of maneuvers:

SNAP-1 and Tsinghua-1 were both released from the Cosmos launcher in slightly different directions specifically to avoid the possibility of accidental re-contact. SNAP-1 ended up in an orbit about 2 km below that of Tsinghua-1 and, being relatively light (6.5kg), suffered more from the effects of atmospheric drag than the much heavier (50 kg) Tsinghua-1 microsatellite. This meant that, relative to Tsinghua-1, SNAP-1 dropped in altitude more quickly. This was exacerbated by a very active sun (at solar maximum), causing the atmospheric density at 700 km to be higher than normal. SSTL measurements showed that on average, SNAP-1 was falling about 10 m per day with respect to Tsinghua-1. 21) 22)

Thus MPS was used firstly to demonstrate orbit control (the primary objective) by maintaining its altitude by overcoming the relative atmospheric drag effects, and then also to climb back up to an altitude about 1 km higher than that of Tsinghua-1. Most of the propellent was used for this climb in December 2000. In this period Tsinghua-1 separated from SNAP-1 by more than 10,000 km along their orbital tracks. - A long sequence of of thruster firings was initiated under the automatic control of the OBC, and the GPS navigation system was used to keep track of the orbital changes. Over a period of 30 days, the thruster was fired about four times per day, giving a change in velocity (delta v) of about 10 cm/s per day. Once SNAP-1 was higher than Tsinghua-1, the along-track gap began to close. In January 2001, SNAP-1 was about 300 m higher than Tsinghua-1 with the gap closing. In total, the propulsion system raised the altitude of SNAP-1 by about 4 km ( corresponding to a total delta v of 2.1 m/s), all done with just 32.6 grams of butane propellant.

At maximum separation, Tsinghua-1 and SNAP-1 were about 15,000 km apart. But by means of the propulsion maneuvers, SNAP-1 brought itself to within 2000 km of its target. Thus, while a true rendezvous was not achieved, the agility and maneuverability of SNAP-1 under automatic control was amply demonstrated, meeting its objectives of demonstrating that nanosatellites can be constructed rapidly to achieve sophisticated mission requirements. 23)

 

Summary of some SNAP-1 achievements:

Following launch into a 700 km sun-synchronous Earth orbit in June 2000 onboard a Cosmos rocket from Plesetsk, SNAP-1 had, by December 2000, achieved over 90% of its mission objectives - achieving a remarkable number of 'world firsts' in this class of nanosatellites. Regarding the MVS, there were of course the problems of light intensity and gain control, but overall the SNAP-1 MVS mission was a sounding success. 24) 25)

• The first fully 3-axis attitude stabilized nanosatellite

• The first nanosatellite with onboard propulsion demonstrating orbit control

• The first in-orbit images of another spacecraft from a nanosatellite (in June 2000, SNAP-1 imaged the Russian Nadezhda satellite as well as Tsinghua-1 in orbit shortly after deployment). - The MVS was used in the following months to acquire images and movie-sequences of the Earth under differing lighting conditions.

• The first successful use of a GPS receiver onboard a nanosatellite - used for orbit maneuvering.

SNAP1_Auto3

Figure 16: The Russian Nadezhda COSPAS-S&RSAT satellite imaged by SNAP-1 just 2 s after deployment (image credit: SSTL)

SNAP1_Auto2

Figure 17: The Tsinghua-1 microsatellite orbiting above the limb of the Earth imaged by SNAP-1 just seconds after its deployment (image credit: SSTL)

Test of the MPS propulsion system: On August 15, 2000, the propulsion system was used for the first time, and shortly afterwards orbital maneuvers were started to try to bring SNAP-1 and Tsinghua-1 back together.

This was made difficult by the differential effects of atmospheric drag, which meant that, unless the thruster is fired, SNAP-1 falls approximately 10 m per day with respect to Tsinghua-1. This, coupled with the initial orbital insertion conditions, meant that SNAP-1 was by now some ~2 km below Tsinghua-1, and some considerable distance ahead of it. Thus, the cold-gas thruster had to be used extensively to re-gain altitude, so as to slow SNAP-1 down with respect to Tsinghua-1, in order for them to be brought back together. A long sequence of firings was initiated under the automatic control of the OBC. The GPS navigation system was used to keep track of the orbital changes.

Over the following 30 days, the thruster was fired approximately 4 times per day, giving a change in velocity (ΔV) of ~10 cm/s per day, by which time SNAP-1 had climbed ~1 km above Tsinghua-1. In total, taking atmospheric drag effects into account, the propulsion system raised the altitude of SNAP-1 by the equivalent of ~4 km (with a corresponding total ΔV of 2.1 m/s) - all done using just 32.6 g of butane propellant.

At maximum separation, Tsinghua-1 and SNAP-1 were approximately 15,000 km apart. By means of these maneuvers, SNAP-1 passed Tsinghua-1's orbital altitude on March 18, 2001, at a minimum separation distance of approximately 2000 km. Thus, whilst true rendezvous was not achieved, the agility and maneuverability of SNAP-1 under automatic control was amply demonstrated.

Attitude control: After launcher separation SNAP-1 had a 5 rpm tumble rate. This initial high rate was damped within 1 day to zero values in the X and Z-body axes- after activation of the magnetorquer rate damping controller. The Y-rate was damped to about 2 rotations per orbit with a somewhat surprising result, i.e. the space pointing facet of the S/C (-Z axis) was almost perfectly tracking the local geomagnetic field vector (B-field). Although the magnetic controller was designed to put the satellite into a Y-Thomson spin, it was suppose to track a constant Y-spin rate reference of 10 rotations per orbit. The compass mode attitude response was then attributed to an internal unmodelled magnetic moment almost aligned with the spacecraft's Z-axis. 26)

The source of the disturbance was eventually traced to magnetic remanence in the dual solenoid valves of the propulsion thruster. The 2 solenoids were supposed to have been wired in opposite polarities, but somehow it never happened. So, instead of having residual magnetic moments that cancel, the 2 dipole moments were summed, resulting in a fairly large magnetic disturbance.

Figure 18 shows a typical B-field measurement onboard SNAP-1 during compass mode. It is clear from this figure that the negative Z-axis is roughly aligned to the B-field vector B. The direction of the disturbance magnetic moment will try to align itself to B in body coordinates, similar to a compass needle tracking the B-field lines. Therefore by determining the average direction over many orbits of the measured B vector in the body coordinates, the direction of the body fixed internal disturbance magnetic moment vector Md will be known.

SNAP1_Auto1

Figure 18: Compass mode measurements of the SNAP-1 attitude response (on Aug. 3, 2000), image credit: SSTL

Over the course of the following months the attitude control algorithms were refined and new software uploaded to the spacecraft, so that by late November, full nadir-pointing momentum-biased operation had been achieved using the pitch-axis momentum wheel.

It was realized that both the Y-wheel dynamics and the magnetic moment cross coupling influence, caused by the magnetorquer rods, complicated an accurate estimation of the internal disturbance. A new in-situ (with active magnetorquer compensation) magnetic moment estimator was developed and after a few iterations, SSTL managed to reduce the 3-axis attitude stability to acceptable levels. This resulted in a nadir pointing performance of ~ 3º (1σ) and a stability good enough for Earth imaging 24 hours per day.

SNAP1_Auto0

Figure 19: Final 3-axis performance with magnetic disturbance compensation (image credit: SSTL)


1) C. I. Underwood, G. Richardson, J. Savignol, “SNAP-1: A Low Cost Modular COTS-Based Nano-Satellite - Design, Construction, Launch and Early Operations Phase,” Proceedings of AIAA/USU Conference on Small Satellites, Logan, UT, Aug. 13-16, 2001, SSC01-VI-7

2) J. Singer, “US Eyes British Demonstration Satellite,” Space News, Oct. 23, 2000, pp. 3 and 19

3) A. Cropp, “The SNAP-1 NanoSat Project at Surrey - A New Generation of Satellites,” Proceedings of the 49th IAF Congress, Melbourne, Australia, Sept. 1998

4) “Surrey Nanosatellite Applications Platform,” SSTL, URL: http://microsat.sm.bmstu.ru/e-library/SSTL/SNAP.pdf

5) Craig S. Clark, Kevin W. Hall, “Power system design and performance on the world's most advanced in-orbit nanosatellite,” URL: http://www.clyde-space.com/documents/1503

6) H. Steyn, Y. Hashida, V. Lappas, “An Attitude Control System and Commissioning Results of the SNAP-1 Nanosatellite,” Proceedings of the 14th AIAA/USU Conference on Small Satellites, Logan, UT, Aug. 21-24, 2000, SSC00-VIII-8

7) Craig I. Underwood, Guy Richardson, Jerome Savignol, “In-orbit results from the SNAP-1 nanosatellite and its future potential,” Philosophical Transactions of The Royal Society, A 2003 361, 199-203, doi: 10.1098/rsta.2002.1123, URL: http://rsta.royalsocietypublishing.org/content/361/1802/199.full.pdf

8) Craig I. Underwood, Mike J. Crawford, Jeff W. Ward, “A Low-Cost Modular Nanosatellite Based On Commercial Technology,” Proceedings of the AIAA/USU Small Satellite Conference, Logan, UT, USA, Aug. 31 - Sept. 3, 1998, SSC98-V-4, URL: http://microsat.sm.bmstu.ru/e-library/Missions/lowcost.pdf

9) H. A. B. Tiggeler, M. S. Hodgart., “A (16,8) error correcting code (t=2) for critical memory applications,” Proceedings of DASIA (DAta Systems In Aerospace) 2000., Montreal, Canada, May 22-26, 2000

10) Cosmos-3M rocket launches with three satellites,” Spaceflight Now, June 28, 2000, URL: http://www.spaceflightnow.com/news/n0006/27cosmos3m/

11) Z. A. Wahl, K. L. Walker, J. Ward, “Modular and Reusable Miniature Subsystems for Small Satellites: An Example Describing Surrey's Nanosatellite S-Band Downlink,” Proceedings of the 14th Annual AIAA/USU Conference on Small Satellites, Logan, UT, Aug. 21-24, 2000, SSC00-IX-4

12) R. Lancaster, “An Optical Remote Inspection System for the Surrey Nanosatellite Applications Program,” Aug. 2001, URL: http://www.carrotworks.com/documents/20010822-1555-surrey-msc-thesis.pdf

13) R. Lancaster, C. I. Underwood, “The SNAP-1 Machine Vision System,” Proceedings of the 14th Annual AIAA/USU Conference on Small Satellites, Logan, UT, Aug. 21-24, 2000, SSC00-II-6

14) “SNAP Machine Vision System,” URL: http://microsat.sm.bmstu.ru/e-library/SSTL/SubSys_SNAP1_MVS.pdf

15) D. Gibbon, J. Ward, N. Kay, “The Design, Development and Testing of a Propulsion System for the SNAP-1 Nanosatellite,” Proceedings of the 14th Annual AIAA/USU Conference on Small Satellites, Logan, UT, Aug. 21-24, 2000

16) “SNAP-1:Propulsion System,” URL: http://centaur.sstl.co.uk/datasheets/SNAP-1%20Propulsion.pdf

17) D. Gibbon, L. Cowie, M. Paul, “COTS (Commercial Off The Shelf) Propulsion Equipment for Low Cost Small Spacecraft,” 38th AIAA/ASME/SAE/ASEE Joint Propulsion Conference and Exhibit, Indianapolis, IN, USA, July 7-10, 2002, AIAA-2002-3994

18) M. J. Unwin, P. L. Palmer, Y. Hashida, C. I. Underwood, “The SNAP-1 and Tsinghua-1 GPS Formation Flying Experiment,” ION GPS 2000, Sept. 19-22, 2000, Salt Lake City, UT, pp. 1608-1611

19) Takuji Ebinuma, Martin Unwin, Craig Underwood, Egemen Imre, “A Miniaturized GPS Receiver fpr Space Applications,” IFAC 2004,URL: http://tebinuma.tripod.com/publications/b10_ACA2004.pdf

20) “SSTL SGR-05 Space GPS Receiver,” SSTL, URL: http://microsat.sm.bmstu.ru/e-library/SSTL/Subsys_SGR05_HQ.pdf

21) Information provided by Craig Underwood and by Dave Gibbon of SSTL, Surrey, UK

22) “Surrey's SNAP-1 Nanosatellite Snaps Satellites in Orbit,” URL: http://www.surrey.ac.uk/news/releases/10-0100sstl.html

23) C. Underwood, V. Lappas, A. da Silva Curiel, M. Unwin, A. Baker, M. Sweeting, “Science Mission Scenarios using PalmSat Picosatellite Technology,” Proceedings of AIAA/USU Conference on Small Satellites, Logan, UT, Aug. 9-12, 2004, SSC04-VIII-3

24) SNAP-1: World's First 3-Axis Stabilized Nanosatellite,” Spacedaily, Jan. 16, 2001, URL: http://www.spacedaily.com/news/nanosat-01a.html

25) Alex da Silva Curiel, Phil Davies, Adam Baker, Craig Underwood, Tanya Vladimirova, “Towards Spacecraft-on-a-Chip,” URL: https://escies.org/download/webDocumentFile?id=1083

26) W. H. Steyn, Y. Hashida, “In-Orbit Attitude Performance of the 3-Axis Stabilized SNAP-1 Nanosatellite,” Proceedings of the AIAA/USU Conference on Small Satellites, Logan, UT, Aug. 13-16, 2001, SSC01-V-1, URL: http://staff.ee.sun.ac.za/whsteyn/Papers/USU2001_Snap2.pdf


The information compiled and edited in this article was provided by Herbert J. Kramer from his documentation of: ”Observation of the Earth and Its Environment: Survey of Missions and Sensors” (Springer Verlag) as well as many other sources after the publication of the 4th edition in 2002. - Comments and corrections to this article are always welcome for further updates.