Acta Astronautica 114 (2015) 65–78

Contents lists available at ScienceDirect

Acta Astronautica
journal homepage: www.elsevier.com/locate/actaastro

Launch and deployment of distributed small

satellite systems$
N.H. Crisp n, K. Smith, P. Hollingsworth
School of Mechanical, Aerospace and Civil Engineering, The University of Manchester, George Begg Building, Sackville St, Manchester M13
9PL, United Kingdom

a r t i c l e i n f o abstract

Article history: The rise in launch and use of small satellites in the past decade, a result of improved
Received 4 November 2014 functionality through technology miniaturisation and alternative design philosophies, has
Received in revised form spawned interest in the development of distributed systems or constellations of small
14 April 2015
satellites. However, whilst a variety of missions based on constellations of small satellites
Accepted 18 April 2015
have been proposed, issues relating to the launch and deployment of these distributed
Available online 29 April 2015
systems mean that few have actually been realised. A number of strategies have been
Keywords: proposed which enable multiple small satellites comprising a constellation to be launched
Small satellites together and efficiently separated on-orbit, thus reducing the total cost of launch. In this
Constellation deployment
paper, two such strategies which have the potential to significantly increase the viability
Nanosatellite
of small satellite constellations in Earth orbit are investigated. Deployment using natural
CubeSat
Nodal precession Earth perturbations to indirectly achieve plane separations is analysed using a developed
Earth–Moon Lagrange point L1 method and compared to deployment utilising the Earth–Moon Lagrange point L1 as a
staging area prior to return to LEO. The analysis of three example missions indicates that
these two strategies can facilitate the successful establishment of small satellite con-
stellations in Earth orbit whilst also reducing propulsive requirements, system complexity,
and/or cost. The study also found that the method of nodal precession is sensitive to the
effects of orbital decay due to drag and can result in long deployment times, and the use of
Lunar L1 is more suitable for constellation configurations where several satellites are
present in each orbital plane.
& 2015 IAA. Published by Elsevier Ltd. on behalf of IAA. This is an open access article under
the CC BY license (http://creativecommons.org/licenses/by/4.0/).

1. Introduction been primarily driven by the miniaturisation of electronics

and sensors [1] and the availability of commercial-off-the-
A growing interest in the use of distributed systems or shelf components with increasing capability, significantly
constellations of small satellites has been generated follow- reducing the cost of hardware development. The access-to-
ing the rise in popularity of small satellites, especially in the orbit and economy of these spacecraft is also improved
past decade. This growth in the use of small satellites has through availability of secondary payload launch opportu-
nities [2,3], especially for small satellites which conform to
standardised form factors such as CubeSat [4].
☆
This paper was presented during the 65th International Astronautical In recent years, the launch of successful small satellite
Congress in Toronto. missions, particularly nanosatellites, with valuable engi-
n
Corresponding author. neering/technology demonstration (e.g. CanX-6/NTS [5],
E-mail addresses: nicholas.crisp@manchester.ac.uk (N.H. Crisp),
kate.smith@manchester.ac.uk (K. Smith),
STRaND-1 [6]), scientific (e.g. O/OREOS [7], GeneSat-1 [8]),
peter.hollingsworth@manchester.ac.uk (P. Hollingsworth). military (e.g. SMDC-One [9], SENSE-1), and commercial

http://dx.doi.org/10.1016/j.actaastro.2015.04.015
0094-5765/& 2015 IAA. Published by Elsevier Ltd. on behalf of IAA. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).
66 N.H. Crisp et al. / Acta Astronautica 114 (2015) 65–78

(e.g. WNISAT-1 [10]) capabilities has now demonstrated ability of these methods to facilitate the establishment of
the utility of this class of spacecraft in independent these systems in low Earth orbit (LEO). Through the use of
operation. a developed methodology, described in detail in Section 4,
The use of small satellites in constellations has also been the relative effectiveness of deployment using natural
successfully demonstrated by a number of microsatellite- Earth perturbations and the Earth–Moon Lagrange point
class missions, including the Disaster Monitoring Constella- L1 are considered for different constellation missions.
tion (DMC) and RapidEye Earth observation missions and the
ORBCOMM [11] satellite communications system. 2. Launch of small satellites
The demonstration of small platform capability and
constellation operation has recently resulted in the genera- The absence of sufficiently small or inexpensive launch
tion of larger multi-plane constellations of smaller satellites. vehicles for the delivery of small satellites to orbit presents a
Two such examples of this new generation of small satellite significant barrier to the development of small satellite
constellation are the Planet Labs [12] (Flock-1a: 28 satellites, missions given their typically smaller budgets and develop-
Flock-1c: 11 satellites) and Skybox Imaging (24 satellites) ment time-scales. This issue of access-to-orbit is somewhat
Earth observation constellations which are currently in the addressed by secondary payload launch opportunities, where
process of being launched. satellite operators can either share launch vehicle capacity
A further value proposition of small satellite constellations, through clustering or rideshare agreements, or utilise excess
resulting from their lower cost of platform development, is the capacity on a commissioned launch of a larger satellite, a
ability to be launched in larger numbers and perform many practise termed piggybacking. Unless arranged through a
simultaneous and distributed measurements or observations. launch programme (e.g. NASA CubeSat Launch Initiative and
A key feature of multi-plane systems of these satellites is Educational Launch of Nanosatellites) with provided or
increased temporal resolution of collected data (i.e. shorter subsidised launch, the cost of secondary payload opportu-
revisit times) over single-plane or string-of-pearls configura- nities is generally greater than the specific cost ($/kg) of the
tions. Furthermore, the presence of multiple satellites in each launch vehicle itself [23]. However, these opportunities still
orbital plane can facilitate a more graceful degradation of allow small payloads to achieve access-to-orbit at a signifi-
system performance on the occasion of individual satellite cantly lower total expense than an independently commis-
failures [13]. sioned launch.
A variety of novel missions benefiting from these cap- The use of secondary payload opportunities is limited by
abilities have been proposed in the fields of meteorology the lack of control on the launch schedule and destination
[14]; climate-science [14,15]; disaster warning and detect- orbit of the vehicle, both controlled by the requirements of the
ion [16–18]; atmospheric, magnetospheric, and ionospheric primary payload or determined by a compromise between the
measurement/observation [14–17,19,20]; and gravity and payload operators in a rideshare launch. As a result, satellites
other Earth sciences [15]. Multi-satellite interplanetary launched as secondary payloads need to be flexible with
exploration missions and constellations in orbit about other regard to the orbit in which their mission can be performed.
central bodies utilising small satellites are also being con- For some missions, this flexibility may not be feasible or may
sidered [16,17,21]. be too costly to embed in the system design.
However, the current launch paradigm of secondary Further restrictions on the launch of small satellites
payload manifesting of small satellites limits the ability of utilising secondary payload opportunities can include the
these constellations to be successfully deployed into orbit. requirement to be compatible with a certain class of
In particular, the lack of control on launch schedule and deployment mechanism (e.g. P-POD, X-POD, ISIPOD), redu-
destination orbit prohibits the use of multiple secondary cing the level of certification required by the secondary
launch opportunities by constellations which require accu- payloads by isolating them from the launch vehicle and
rately coordinated orbits and multi-plane configurations. primary payload [4]. This can further constrain the mass
This issue is further compounded by technology, mass, and and volume of the satellite and any provision for deploy-
volume constraints on propulsion system capability to able surfaces such as solar arrays or wireless communica-
maintain low development and manufacturing costs and tion antennae. Constraints on volumes and pressures of
comply with launch vehicle regulations. These constraints stored propellant, nominally to protect the primary pay-
can be particularly restrictive for the smaller nanosatellite load, can also limit the capability of on-board propulsion
and picosatellite class platforms which are therefore systems, further restricting the ability of the secondary
typically limited in their ability to individually manoeuvre payloads to manoeuvre into more suitable or favourable
into their mission orbits [1,4,22]. mission orbits.
In order to enable the cost-effective realisation of small A number of new launch vehicles aiming to address the
satellite constellations a number of deployment strategies microsatellite and nanosatellite launch capability gap are
have been proposed which allow the launch of a complete currently in varying stages of development. The payload
multi-plane constellation on a single vehicle with satellite capability of these vehicles ranges from 12 to 300 kg with
distribution occurring on-orbit. Currently, the FORMOSAT-3/ specific launch costs in the range of current secondary payload
COSMIC mission is the only example of a multi-plane small opportunities. Notable examples include the Virgin Galactic
satellite constellation to be deployed from a single launch LauncherOne which will be air-launched from the White-
vehicle. KnightTwo carrier aircraft and will have a capacity on the
This paper investigates two deployment methods for order of 225 kg to LEO [24], a 10 kg payload launcher
constellations with multi-plane configurations and the deployed from the XCOR Aerospace Lynx Mk.III suborbital
 N.H. Crisp et al. / Acta Astronautica 114 (2015) 65–78 67

Fig. 1. Δn with drift time for a fixed plane separation of 601 at varying inclination.

vehicle [25], and the DARPA ALASA program involving several capability. Further technological, mass and volume con-
companies working towards the launch of a 45 kg payload to straints can compound this issue. The ability of small
orbit for less than $1M. satellites to simply manoeuvre from their insertion orbit into
These vehicles will support the dedicated launch of their required mission orbits is therefore restricted, espe-
microsatellites and nanosatellites, avoiding the potentially cially if expensive out-of-plane manoeuvres in Right Ascen-
mission critical issues related to secondary payload launch sion of Ascending Node (RAAN) or inclination are required.
opportunities. In order to achieve the payload distribution required by
the mission specification, various deployment strategies
2.1. Launch of small satellite constellations have been developed allowing multiple satellites to be
launched together and separated on-orbit. These include
Traditionally, constellations of satellites have been the use of nodal precession due to natural Earth perturba-
populated through many launches, one or more per orbital tions and the use of the Earth–Moon Lagrange point L1,
plane, or even one per satellite. However, due to the which will be examined in Sections 3.1 and 3.2.
prohibitive cost of launch in comparison to the develop-
ment cost of smaller satellites, launch in this manner is not
3.1. Nodal precession
economically viable for small satellite constellations.
For relatively high-budget missions of microsatellite
A method of constellation deployment using natural
constellations, the cluster launch method can currently
orbital perturbations to separate orbital planes in RAAN,
provide the opportunity to launch a constellation to orbit.
often termed indirect plane separation, was patented in
For example, the FORMOSAT-3/COSMIC mission of five
1993 by King and Beidleman [27]. The method utilises the

61 kg satellites was launched on an Orbital Sciences
differential rate of nodal precession due to the non-
Minotaur I vehicle [26]. However, due to the present lack
spherical geopotential of the Earth whereby orbits with
of sufficiently small launch vehicles, the launch of similar
different sizes, shapes, or orientation precesses at different
nanosatellite and picosatellite constellations, unless in
rates, allowing plane separations to be achieved without
extremely large numbers, becomes uneconomical. The
out-of-plane manoeuvring. Eq. (1) [13] shows the analy-
launch of a very small satellite constellation, the Planet _ , as a function
tical form for the rate of nodal precession Ω
Labs Flock constellation of 3U CubeSats, has thus far
of semi-major axis a, eccentricity e, and inclination i. In Eq.
achieved orbit by both manifestation on a re-supply
(1), only the secular effects due to the second degree zonal
launch to the ISS and subsequent deployment from the
harmonic J2 are shown for brevity:
Kibo module and deployment as secondary payloads from
a Dnepr launch, resulting in different orbital inclinations _ ¼  3 R2E
and altitudes [12]. Ω J2

2  J 2  n  cos i ð1Þ
2 a 1 e2
The emergence of new small launch vehicles, men-
tioned previously, would support the cluster launch of The process for a constellation deployment from a
very small satellite constellations. common insertion orbit initially requires an in-plane man-
oeuvre of a satellite into an orbit with a different rate of
3. Constellation deployment strategies nodal precession, nominally the mission orbit. A drift period
is then required in order to achieve the correct angular
For many missions a fixed constellation configuration of separation in RAAN, before a second satellite is manoeuvred
the payloads is required. If many small satellites comprising a into the mission orbit, fixing the developed plane separation
constellation are launched from a single launch vehicle, the between the first two payloads. This process is repeated for
payloads must be dispersed on-orbit into their respective all required planes in the constellation.
mission orbits. However, to reduce system complexity and The drifting time required for such a deployment is
development costs, if present, the propulsion systems on- dependent on the required separations and the differential
board these small platforms are typically limited in drift rate between the initial and modified orbits, and is
68 N.H. Crisp et al. / Acta Astronautica 114 (2015) 65–78

therefore affected by the propulsive capability of the satellites. expenditure for Earth orbit re-entry. An aerocapture man-
The relationship between time for a fixed plane separation oeuvre differs from aerobraking in that the total speed
and propulsive capability (Δn) for different orbital inclinations reduction is performed during only one pass through the
is down in Fig. 1. For modest Δn expenditures, drift periods atmosphere of the central body, utilising the high aero-
for the deployment of a complete constellation can be dynamic drag of an elliptical orbit with a low altitude
expected to be on the order of many months to years. These perigee. Aerobraking manoeuvres utilise a higher altitude
long drift-times may be undesirable from an operation or perigee with lower drag, therefore requiring more passes
mission impact perspective or may present issues regarding to bring the satellite into the correct orbit, but with lower
the lifetime of the satellite in orbit with respect to both orbital aerodynamic forces and heating effects.
decay and hardware reliability. Following direct launch to EML-1 or transfer from LEO, the
The deployment of multiple-satellites into each orbital propulsive requirements of the carrier vehicles for deploy-
plane can be facilitated by manifesting the payloads on ment using this method involve manoeuvres for Halo orbit
carrier vehicles, termed pallets by King and Beidleman injection and departure at EML-1, approximately 600 m s  1
[27]. These carrier vehicles, each equipped with a centra- each, and Earth orbit circularisation after the aerocapture
lised propulsion system, can perform the required man- manoeuvre (120 m s  1) [32,34]. Finally, phasing manoeuvres
oeuvre and drift strategy to enter the correct orbital plane must be performed by the individual satellites in order to
before releasing the individual satellites. achieve the in-plane separations.
The benefit of a propulsion system common to a group The primary benefit to this method of deployment is
of satellites is the alleviation of some mass and volume the significantly reduced time required in order to achieve
constraints. Some propulsion technologies which are not the mission configuration whilst still enabling the use of a
suitable for the individual satellite platforms due to size single launch. A period on the order of 20 days is required
may also become viable due to the increased scaling of the to return a compliment of payloads to Earth orbit with any
system. required spacing of the orbital planes in RAAN [33].
Finally, the satellites on each pallet can be distributed
about the orbit in each plane. Aside from individual phasing of 4. Analysis methods
the payloads, this can be achieved using differential spring
separation energies from the pallets, requiring a wait period of The feasibility of these deployment methods for sup-
up to 50 days and a small Δn contribution from each satellite porting missions of interest is primarily dependent on the
(up to 7.4 m s  1) in order to freeze the distribution [28]. time and energy required by the deployment strategy and
Differential drag methods can also be used to achieve the mission requirements of the constellation.
required in-plane separations, demonstrated by the Aero- Deployment of LEO constellations by nodal precession
Cube-4 mission [29] and Planet Labs Flock-1 constellation requires consideration of the effects of atmospheric drag
[12] and proposed for use on the NASA CYGNSS mission due on the satellites. Satellites in orbits of a different semi-
for launch in 2016 [30]. These systems use active attitude major axis and eccentricity can experience significantly
control methods and deployable surfaces to alter the pro- different levels of drag and will therefore decay at different
jected area of the satellite and therefore drag profile to achieve rates. For low altitude constellations, there may be a risk of
minor differences in semi-major axis, therefore enabling the satellite de-orbit before the deployment scheme has been
separation of the satellites within the plane. completed. Propagation of the satellites throughout the
This method of constellation deployment utilising dif- deployment procedure can be used to determine the effect
ferential nodal drift rates has thus far only been demon- of drag on the constellation.
strated by the FORMOSAT-3/COSMIC mission, launched in The physical characteristics of the satellites and optional
April 2006, in which 6 satellites were each deployed into a carrier vehicles can also influence the deployment process,
different plane [31]. A total Δn requirement of 147 m s  1 affecting the magnitude of drag experienced and propulsive
was estimated for each satellite, provided by multiple system requirements. Additional system masses, such as
thrust-burns of the Hydrazine monopropellant propulsion aerocapture devices and payload deployment systems, must
subsystem. A total period of approximately 20 months was also be considered.
required for the deployment of this constellation.
4.1. Satellite propagation
3.2. Lunar L1
A semi-analytical propagation method developed by
A method of LEO constellation deployment utilising the Liu and Alford [35], Semi-Analytical Liu Theory (SALT), was
first Earth–Moon Lagrange point (EML-1) was proposed by used in order to analyse the orbits of the satellites during
Chase et al. [32] and developed by Nadoushan and the deployment of the constellation using the differential
Novinzadeh [33]. In this method satellites for each nodal precession method. Accommodation of both effects
intended plane of the constellation are manifested on due to atmospheric drag and the non-spherical geopoten-
carrier vehicles which are all launched together on a single tial of the Earth are required in order to allow analysis in
vehicle to EML-1. The compliment of carrier vehicles is the LEO environment and determination of the rate of
inserted into a Halo orbit at EML-1 before being individu- nodal precession of the orbits [36].
ally returned to Earth orbit with the required inclination A semi-analytical method was chosen due to its speed
(up to 601) and ascending node. An aerocapture or aero- of execution in comparison to numerical propagation
braking manoeuvre can be used to reduce the Δn techniques, allowing for the analysis of a potentially large
 N.H. Crisp et al. / Acta Astronautica 114 (2015) 65–78 69

Fig. 2. Daily and 81-day centred average F10.7 solar radio flux and daily planetary amplitude, Ap.

number of satellites over long time periods. Furthermore, the requirement for forecast solar flux and geomagnetic
the increased precision of a fully numerically determined index data. Due to the highly dynamic nature of the Sun
solution would be dominated by the uncertainties asso- and near-Earth environment these parameters are difficult
ciated with the evaluation of future atmospheric density to predict, and are therefore associated with a high level of
using predicted solar flux and geomagnetic index data. In uncertainty.
contrast to purely analytical methods, a semi-analytical
propagation method can accommodate the use of a com- 4.2. Space weather indices
plex atmospheric density model rather than requiring a
simplified model or series-expansion. Solar flux incident on the upper atmosphere of the
SALT uses a combination of both general and special Earth, primarily Extreme Ultraviolet (EUV) radiation, has a
perturbation techniques to consider both effects of atmo- heating effect which affects the local atmospheric density.
spheric drag and perturbations due to non-spherical The variation of incident solar flux affects the level of
geopotential. Perturbations due to third body effects (lunar heating within the atmosphere, and therefore the drag
and solar) and solar radiation pressure are neglected due that a satellite in orbit will experience.
to their small magnitude in comparison to the primary Measurements of solar flux with a wavelength of
perturbations and domination by the uncertainty involved 10.7 cm (F10.7) are used as a surrogate for EUV radiation
in atmospheric density and ballistic coefficient determina- due to their similar level of production by the Sun and low
tion [35]. level of absorbance of F10.7 in the atmosphere. Ground-
The 2001 United States Naval Research Laboratory Mass based sensors can therefore be used to measure the level
Spectrometer and Incoherent Scatter Radar Exosphere of incident F10.7 on the Earth and approximate the effect of
(NRLMSISE-00) atmosphere model was used for density EUV on atmospheric density [36].
evaluation in the orbit propagator. The NRLMSISE-00 The level of F10.7 has both a number of periodic trends
model is a global and time-varying model of the Earth's and significant random variation, shown in Fig. 2. In the
atmosphere valid from the ground to 1000 km altitude long term, F10.7 varies with the 11-year solar cycle, leading
[37]. The required inputs to the model include longitude; to lengthy periods of high and low solar flux, solar maxima
geodetic altitude and latitude; date and time; Ap geomag- and minima. However, the magnitude and timing of the
netic index; and daily and 81-day centred average F10.7 solar cycle is difficult to predict, leading to significant
Solar radio flux. uncertainty in solar flux prediction. Furthermore, random
Whilst complex modern atmospheric density models variation in solar flux can be seen to increase during solar
can exhibit significant improvements over older techni- maxima in comparison to periods of solar minima. In the
ques (e.g. 1976 Standard Atmosphere Extended, Exponen- short-term, solar flux variability is primarily related to the
tial Model), a number of simplifications and assumptions 27-day rotation of the Sun, causing transient areas of high
are made in order to facilitate their development. As a and low activity to turn towards or away from the Earth
result, significant errors are present in the evaluation of [36,39].
atmospheric density, typically in the range 10–20% [38]. The prediction of geomagnetic index, a measurement
For analysis of proposed missions, the evaluation of of the interaction between energetic charged particles
atmospheric density and drag is further complicated by ejected from the Sun and atmospheric molecules, caus-
70 N.H. Crisp et al. / Acta Astronautica 114 (2015) 65–78

ing particle ionisation and a heating effect, is similarly properties of the spacecraft:
complex. Geomagnetic index is expressed as either plane-
CDA
tary index, Kp (on a scale of 0–9), or planetary amplitude ap BC ¼ ð2Þ
m
(0–400 nT), measured globally every 3 h. A daily planetary
amplitude, Ap, is generated by averaging the set of 3- Whilst the mass of a satellite is usually known or can be
hourly ap values. Due to the influence of the Sun, coupling calculated from propellant usage, the evaluation of drag
with the solar cycle is present in the magnitude of coefficient and cross-sectional area can be more complex.
geomagnetic index. The most pronounced effect is that The coefficient of drag for a spacecraft is both related to
variability increases somewhat during periods of solar the shape of the spacecraft and the molecular interaction
maxima, shown in Fig. 2. between atmospheric gas particles and the surfaces of the
For the deployment of satellite constellations in LEO, spacecraft. As a result, the coefficient of drag for the
the long-term forecasting of solar flux and geomagnetic spacecraft can vary significantly with altitude due to the
index is critical due to the length of analysis required. A varying levels of atmospheric constituents [43]. However,
number of approaches used for the prediction of these for long analysis periods this variation in drag coefficient
indices are examined by Vallado and Finkleman [40] and with altitude can be accounted for through the use of an
Vallado and Kelso [41,42]. In Fig. 3, historic NOAA/SWPC average value, the classic estimate of 2.2 is typically used
predictions are shown against measured values. A poly- [44].
nomial trend developed by Vallado [36], matched to the The cross-sectional area for ballistic coefficient deter-
magnitude and phasing of the past four solar cycles is also mination is taken normal to the velocity vector of the
shown. The performance of these predictions show a satellite and is therefore dependent on the attitude of the
capability to reasonably match the trend of solar flux spacecraft. If the attitude is known, either fixed or rotating,
through the remainder of a solar cycle. However, the the cross-sectional area with respect to the velocity vector
predictions are typically incapable of matching the trend can be calculated. However, for tumbling spacecraft with
spanning multiple solar cycles. This is demonstrated by the constantly varying attitude, an averaged cross-sectional
performance of prediction of the latest solar cycle which area can be calculated either by integrating the different
has a characteristically lower magnitude and later occur- cross-sectional areas over the range of possible attitudes
rence of solar maximum. or using a simplified flat-plate model [44].
The aggregated set of solar flux and geomagnetic index
data from NOAA/SWPC, available from the CelesTrak web-
4.4. Deployment analysis procedure
site, was chosen due to the completeness and availability
of up-to-date measured and forecast data.
Utilising the principles set out in Sections 4.1–4.3, the
deployment of a constellation of satellites using the
method of differential nodal precession from a single
launch insertion point can be analysed. The deployment
4.3. Ballistic coefficient analysis of a constellation with one satellite in each plane
can be represented by the basic procedure shown in Fig. 4.
The physical characteristics of orbiting satellites affect The semi-analytical propagation method enables the
the amount of drag which is experienced whilst in orbit. examination of satellite orbital decay through the use of
The coefficient of drag CD, spacecraft mass m, and cross- predicted Earth environment data and a complex atmo-
sectional area A can be mathematically combined to spheric density model. For constellations inserted or
generate the ballistic coefficient for the satellite, Eq. (2), operated in low altitude orbits, the decay of satellites prior
a measure of the effect of drag as a result of the physical to the complete deployment of the constellation would

Fig. 3. Comparison of measured and predicted F10.7 solar radio flux.

 N.H. Crisp et al. / Acta Astronautica 114 (2015) 65–78 71

Fig. 4. Procedure for deployment analysis of a constellation with one satellite per plane.

constitute an infeasible design. Furthermore, any change in Finally, the payloads or carrier vehicles must perform a
orbit shape and size alters the rate of nodal precession and re-circularisation manoeuvre for insertion into the mission
can therefore affect the rate of separation of planes in the orbit and any phasing manoeuvres required.
constellation.
For constellation configurations in which there is more 5. Constellation deployment analysis
than one satellite in each plane additional phasing man-
oeuvres are required to achieve the configuration of pay- The method of analysis for deployment using the
loads. In these circumstances, carrier vehicles can be used method of indirect plane changes can be validated against
in order to perform the more intensive in-plane man- the actual deployment of the FORMOSAT-3/COSMIC
oeuvres and then dispense the payloads into each plane. mission.
Analysis of other small satellite constellation missions
4.5. Analysis of Lunar L1 deployment with varying properties and configurations can be used to
investigate the implementation of the different deployment
Due to the complexity of trajectory design for missions methods and their relative capabilities and limitations.
involving EML-1, the corresponding analysis for deploy-
ment of constellations EML-1 for staging of return to LEO is 5.1. Deployment of FORMOSAT-3/COSMIC
treated in a more qualitative manner.
The first stage of an EML-1 deployment strategy The FORMOSAT-3/COSMIC constellation of six micro-
requires either a launch vehicle capable of direct launch satellites was launched in April 2006 on an Orbital
to EML-1 or transfer of payloads from LEO. Due to the Sciences Minotaur vehicle. To achieve the mission config-
relatively high Δn of transfer to EML-1 (
3.8 km s  1), uration, the six payloads were deployed into six orbital
system complexity can be reduced by manifesting the planes equispaced about 1801 in RAAN using the method
payloads or sub-carrier vehicles on a single transfer of natural nodal precession over a period of 20 months.
vehicle with a capable and centralised propulsion system. As this mission has been launched and deployed, the
On the return to Earth orbit, aerocapture or aerobraking actual deployment of the satellites, using Two-Line Ele-
manoeuvres are required to reduce the velocity of the ment (TLE) data can be compared to the deployment
payloads or carrier vehicles. The use of ballute devices simulated using the analysis technique previously detailed.
(inflated parachutes for high-speed aerodynamic braking) The satellite platform properties, insertion orbit, and
and drag-sails is typically proposed to perform these required mission orbits of the satellites are shown in
aerodynamic manoeuvres, significantly reducing the pro- Table 1. This information was used as the input variables
pulsive requirements of the system. However, additional for the deployment analysis. As only one satellite per plane
mass, system complexity, and cost may be associated with was required by this constellation, no benefit to the
the use of such devices and manoeuvres. Issues such as deployment process could be gained through the use of
thermal shielding and additional structural requirements carrier vehicles. Historical measured values for solar flux
due to aerodynamic forces and heating, increased attitude and geomagnetic activity were utilised in the propagation
control requirements, and radiation hardening of compo- method.
nents due to passes through the van Allen belts must be Comparison of the semi-major axis and RAAN profiles of
considered if aerocapture or aerobraking manoeuvres are the actual mission, shown in Fig. 5, and the analytically
to be used. determined deployment, shown in Fig. 6, indicates that the
72 N.H. Crisp et al. / Acta Astronautica 114 (2015) 65–78

developed analysis is capable of successfully characterising (b) The incomplete deployment of FORMOSAT-3D was due
the process of constellation deployment using nodal preces- to a propulsion system malfunction which prohibited
sion. However, several discrepancies exist between the two the satellite from achieving the correct mission orbit.
data-sets which are due to mission changes, anomalies, and (c) Numerous thrust-burn failures and an issue with
system failures in the actual deployment of the constellation: thrusting during sunlit periods are also contributing
factors to irregularities and discontinuities in the
(a) The discontinuous orbit-raising of FORMOSAT-3F was orbit-raising manoeuvres of the satellites during the
caused by an intentional change in the specified plane deployment of the constellation.
spacing from 241 to 301 after the manoeuvre had
begun. To enable the correct separation of all the The calculated and actual Δn requirement of each
orbital planes, an additional drift period was required satellite in the constellation is presented in Table 2 in
before the manoeuvre could be completed. order of satellite deployed. Given the same initial and final
orbit, the difference in analytically calculated Δn between
Table 1
the satellites is due to the amount of orbital decay
FORMOSAT-3/COSMIC mission specifications [31,26]. experienced. The satellites which stayed in the lower
altitude initial orbit for longer experienced a greater level
Property Value of drag for a longer period and therefore required a
marginally larger orbit-raising manoeuvre to achieve
No. of satellites 6
No. of planes 6 the final mission orbit. The actual Δn values were calcu-
Satellite dry mass 54 kg lated using the average increase in semi-major axis per
Satellite fuel mass 6.65 kg burn and number of thrust events recorded for each
Thrust, BoL to EoL 1.1 N to 0.2 N satellite by Fong et al. [31]. The actual Δn expenditure
Specific impulse 217 s to 194 s
Satellite area 0.5963 m2
calculated for FORMOSAT-3B using the published data
Satellite CD 2.2 (Assumed) appears to be incorrect, as this would not have been
sufficient to raise the satellite from the insertion orbit to
Insertion orbit
Semi-major axis, a 6893 km the mission orbit.
Eccentricity, e 0.00323 Aside from the propulsive failure for FORMOSAT-3D
Inclination, i 71.9921 and an anomalistic value for FORMOSAT-3B, the difference
RAAN, Ω 301.1581 between the predicted and actual Δn expenditure is less
Mission orbit than 1%.
Semi-major axis, ai 7178 km Deployment of the FORMOSAT-3/COSMIC mission using
Eccentricity, ei o 0.014
EML-1 can also be considered. However, due to the
Inclination, ii 71.9921
RAAN Ω1, Ω1  ð30; 60; 90; 120; 150Þ1 constellation configuration of only one satellite in each
orbital plane, each platform would require an individual

Fig. 5. Semi-major axis and RAAN of FORMOSAT-3/COSMIC mission from TLE data.
 N.H. Crisp et al. / Acta Astronautica 114 (2015) 65–78 73

Fig. 6. Semi-major axis and RAAN of FORMOSAT-3/COSMIC mission using developed analysis.

Table 2 5.2. Halo constellation deployment

Calculated Δn requirement for FORMOSAT-3/COSMIC deployment.
A number of novel and interesting configurations of
Satellite Analytical Δn (m s  1) Actual Δn (m s  1) [31]
small satellite constellations are proposed by Janson [45].
3C (FM5) 152.2 153.1 One of these constellations, termed a Halo constellation, is
3F (FM2) 152.6 154.0 a Walker-Delta configuration in which a large number of
3A (FM6) 153.0 153.0 payloads are equispaced about 3601 in RAAN with no
3E (FM4) 153.7 154.0
3D (FM3) 154.1 107.9a
relative-phasing. This results in a ‘bouncing’ motion of
3B (FM1) 154.5 129.8b the satellites, shown in Fig. 7 for a 601: 50/50/0 constella-
tion at an altitude of 1000 km. For a constellation of this
a
Propulsion system failure configuration the period of the bounce is 105.1 min [45].
b
Minimum Δn for transfer is 152.1 m s  1 Due to the fixed relationship between all the satellites in
the constellation, inter-satellite data-links can be reliably
created. The downlink of data from the complete constel-
ballute device (
50 kg [34]) to perform the aerocapture lation can therefore be performed twice during every
procedure and a significantly more capable propulsion bounce of the constellation.
system to perform the Earth orbit re-circularisation and The deployment of such a constellation using tradi-
Halo orbit ejection manoeuvres. tional methods would require many launches and would
The fuel mass to perform the required manoeuvres can therefore be prohibitively expensive. However, for nano-
be calculated using the rocket equation and an assumed satellites or picosatellites, the full complement of payloads
dry mass of the system. For a bipropellant propulsion could be launched by a single launch vehicle.
system with an Isp of 300 s, the mass of propellant required A sample deployment of a 3U CubeSat constellation
for the total Δn of approximately 1500 m s  1 is
75 kg. with this configuration has been investigated. The proper-
This results in a total satellite wet mass in excess of 190 kg, ties for this deployment are presented in Table 3. The
over three times greater than the original payloads. propulsion system for each satellite in this case has been
Additionally, the delivery of these heavier payloads to limited to a maximum Δn of 1 m s  1 in every 12 h period.
EML-1 requires a greater launch capability, increasing This thrusting performance is well within the capability of
costs further. microsatellite and nanosatellite propulsion technologies
However, deployment of the constellation using trans- such as miniaturised cold-gas, monopropellant, and bipro-
fer from EML-1 could be completed in a significantly pellant thrusters [22,46]. For a 3U CubeSat, a propulsion
reduced time compared to deployment performed using system such as this can be contained within a 1U section.
nodal precession which required a period of over The results of a sample deployment of a constellation of
20 months. Should the establishment of the constellation this configuration are shown in Fig. 8. Only the motion of
been more urgent, deployment utilising EML-1 may have every 10th satellite in the constellation is shown to
been the preferred method. maintain clarity. The complete time for deployment of all
74 N.H. Crisp et al. / Acta Astronautica 114 (2015) 65–78

Fig. 7. Halo constellation (601: 50/50/0).

50 satellites is 47 months, just under 4 years. This is a the satellites therefore does not vary significantly between
significant period of time for the deployment phase of a the first and last payloads deployed (87.0–87.2 m s  1).
constellation. However, given a greater difference between The deployment of a significant number of very small
the initial and mission orbits, the deployment can be satellites in a Halo constellation using EML-1 presents a
executed faster at the expense of additional propulsive number of significant issues which would prohibit the use
requirement. of this method for a constellation of this type. The primary
Due to the high altitude of the insertion orbit for this issue is due to the configuration of one payload per orbital
mission, the satellites do not decay significantly during the plane, similar to the deployment of the FORMOSAT-3/
drift phases of deployment. The range of Δn required by COSMIC mission discussed previously.
 N.H. Crisp et al. / Acta Astronautica 114 (2015) 65–78 75

To perform the Halo ejection and re-circularisation considered. In this constellation, multiple satellites are
manoeuvres, the propulsive capability of the satellites required in each orbital plane in order to achieve a certain
would have to be increased significantly. For example, level of global coverage or minimum revisit time.
the propellant mass required for a satellite of nominal The example constellation used has a configuration of
15 kg dry mass with a propulsion system Isp of 200 s would six equally spaced planes each containing five payloads at
be 17 kg. an inclination of 701. The size and mass of the SSTL
Very small satellites would also require significant Microsat-100 bus, used for the first DMC mission satellites,
system development in order to be capable of performing is used as a reference for the payloads in this study. The
the aerocapture manoeuvre, requiring highly accurate attitude constellation and satellite properties for this analysis are
control and significant heat-shielding due to their typically shown in Table 4.
low thermal inertia. Alternatively, an aerobraking manoeuvre The deployment profile of this mission is shown in Fig. 9.
utilising a sail device could be performed. In this case, due to Due to the lower initial orbit of the satellites, the decay due
the length of time spent in the Van Allen belts, radiation to atmospheric drag is considerable, indicated by the
hardening of components or a fault-tolerant approach to decrease in semi-major axis of the payloads in the initial
system design would be required, incurring additional devel- orbit. Over the period of deployment this orbital decay
opment costs or system complexity. results in the re-entry of the final group of payloads,
resulting in the establishment of an incomplete constellation.
5.3. Global microsatellite constellation To avoid this mission failure, the orbital height of the
initial orbit can be increased. Whilst Δn requirement for
The deployment of a constellation of microsatellites for the in-plane manoeuvre would be reduced, this would
communications or Earth observation missions can also be result in longer drift periods to achieve the correct plane
separations and therefore an extended total time for the
Table 3 deployment of the constellation.
Mission properties for Halo constellation.
Alternatively, the satellites could be inserted into an
Property Value orbit with a larger semi-major axis and/or eccentricity
than the final mission orbit thus avoiding the issue of
No. of satellites 50 decay and increasing the rate of plane separation. How-
No. of planes 50 ever, this may result in greater launch vehicle costs.
Satellite dry mass 5 kg
Satellite area 0.035 m2
Due to the configuration of multiple payloads in each
Satellite CD 2.2 orbital plane, the deployment of this constellation using
EML-1 supports the use of carrier vehicles in order to
Insertion orbit
Semi-major axis, a 7200 km reduce the propulsive requirements for each individual
Eccentricity, e 0.001 payload. For this deployment, each carrier vehicle can
Inclination, i 601 manifest the five payloads specified for each plane.
RAAN, Ω 01 The carrier vehicle can perform the required Halo orbit
Mission orbit injection and ejection manoeuvres at EML-1 and re-
Semi-major axis, ai 7371 km circularisation in Earth orbit using a centralised propulsion
Eccentricity, ei 0.001
module. A higher thrust, and greater specific impulse
Inclination, ii 601
RAAN Equispaced about 3601 system can be developed due to the lower mass and
volume constraints of a system common to the set of

Fig. 8. Semi-major axis of selected satellites through analysed deployment of a sample Halo constellation.
76 N.H. Crisp et al. / Acta Astronautica 114 (2015) 65–78

payloads. Once in the required orbital plane, the payloads determined for the FORMOSAT-3/COSMIC mission in
can be dispensed from the carrier vehicle with varying which only a single satellite was required in each plane.
relative velocities using differential spring energies. The
required phasing of the satellites about the orbital plane
can thus be achieved with minimal propulsion from each 6. Summary
individual payload [28].
Further mass and costs efficiencies can be achieved The deployment of constellations using the indirect
through the use of carrier vehicles as the high-accuracy method of plane separation by differential nodal precession
ADCS and aerocapture device can be common to the set of was shown to enable the deployment of a novel type of very
satellites whilst also less restricted by the mass and small satellite constellation whilst maintaining a low level of
volume of the individual payloads. propulsive requirement of each payload. However, the draw-
For a set of five satellites with a nominal mass of 75 kg, back of this method was found to be the significant time
the dry mass of a suitable carrier vehicle can be approxi- required to execute the deployment of the payloads over
mated to 190 kg (50% of payload mass [13]) plus an large plane separations. For constellations in low altitude
aerocapture device (ballute) of 100 kg [34]. The additional orbits, due to the extended drift periods, the decay of
propellant mass for such a system is calculated to be satellites before the deployment has been completed must

405 kg assuming an Isp of 320 s due to the increase in also be carefully considered. Furthermore, the lifetime of the
size and capability of the system. A total mass of
1070 kg satellites in orbit following deployment should be investi-
is subsequently obtained for the mass of each manifested gated to ensure that decay and potential de-orbit does not
carrier vehicle, resulting in an increase in mass of each compromise the mission.
original satellite (100 kg) by a factor of 2.1. This is con- Conversely, the investigation of deployment using EML-1
siderably less than the increase in mass previously to dispense satellites into different orbital planes found that
this method was most suitable for constellations which
Table 4 require responsive set-up. Due to the significant magnitude
Mission properties for microsatellite constellation. of the manoeuvres involved, the propulsive requirements to
Property Value
enable this deployment strategy are much greater than the
indirect method. Furthermore, due to the use of aerocapture
No. of satellites 30 or aerobraking manoeuvres and multi-payload carrier vehi-
No. of planes 6 cles to reduce the total and individual payload propulsive
Satellite dry mass 100 kg
requirements, additional system development is needed to
Satellite area 0.975 m2
Satellite CD 2.2 enable deployment by this method. For constellations where
only a single satellite is required in each orbital plane, this
Insertion orbit
Semi-major axis, a 6825 km
method was found to be inefficient. However, in the analysis
Eccentricity, e 0.01 of constellations with multiple satellites in each plane, where
Inclination, i 701 carrier vehicles could be used effectively to reduce the
RAAN, Ω 01 individual propulsive requirements of the satellites, this
Mission orbit method was shown to be more suitable.
Semi-major axis, ai 7078 km Nevertheless, the performed analysis of these different
Eccentricity, ei 0.001 small satellite constellation deployment methods has
Inclination, ii 701
RAAN Equispaced about 3601
indicated that the establishment of such systems on a
variety of scales in Earth orbit is within the current

Fig. 9. Semi-major axis and orbit decay profile of microsatellite constellation deployment.
 N.H. Crisp et al. / Acta Astronautica 114 (2015) 65–78 77

capability of such systems and can be performed using [5] F.M. Pranajaya, R.E. Zee, Nanosatellite tracking ships: from concept
existing payload launch opportunities. to launch in 7 months, in: 23rd Annual AIAA/USU Conference on
Small Satellites, American Institute of Aeronautics and Astronautics
(AIAA), Logan, UT, 2009, pp. 1–6.
6.1. Future work [6] S. Kenyon, C.P. Bridges, STRaND-1: use of a $500 smartphone as the
central avionics of a nanosatellite, in: The 62nd International
Astronautical Congress, International Astronautical Federation
Given the relationships between constellation configura- (IAF), Cape Town, SA, 2011.
tion, payload design, and feasibility of deployment shown by [7] W.L. Nicholson, A.J. Ricco, E. Agasid, C. Beasley, M. Diaz-Aguado,
the analyses performed, the study of constellation deployment P. Ehrenfreund, C. Friedericks, S. Ghassemieh, M. Henschke,
J.W. Hines, C. Kitts, E. Luzzi, D. Ly, N. Mai, R. Mancinelli,
should be performed concurrently throughout the design
M. McIntyre, G. Minelli, M. Neumann, M. Parra, M. Piccini, R.
process to ensure the development of feasible system and M. Rasay, R. Ricks, O. Santos, A. Schooley, D. Squires, L. Timucin,
mission designs. Currently, these methods of deployment are B. Yost, A. Young, The O/OREOS mission: first science data from the
not considered by existing design processes. The investigation space environment survivability of living organisms (SESLO) pay-
load, Astrobiology 11 (10) (2011) 951–958, http://dx.doi.org/10.1089/
of deployment of small satellite constellations is therefore ast.2011.0714. ISSN 1557-8070.
currently performed on an ad hoc basis without complete [8] C. Kitts, J.W. Hines, E. Agasid, A.J. Ricco, B. Yost, K. Ronzano, J. Puig-
analysis. Awareness and knowledge of the solutions to and Suari, The GeneSat-1 microsatellite mission: a challenge in small
satellite design, in: The 20th Annual AIAA/USU Conference on Small
effects of different constellation deployment strategies may
Satellites, American Institute of Aeronautics and Astronautics
also influence other aspects of the constellation, payload, or (AIAA), Logan, UT, 2006, pp. 1–6.
platform design resulting in the synthesis of improved overall [9] J.R. London III, M.E. Ray, D.J. Weeks, A.B. Marley, The first US army
system or mission designs. The use of design space explora- satellite in fifty years: SMDC-ONE first flight results, in: The 25th
Annual AIAA/USU Conference on Small Satellites, American Institute
tion or optimisation techniques may present a route by which of Aeronautics and Astronautics (AIAA), Logan, UT, 2011.
constellation deployment can be investigated and integrated [10] S. Kim, T. Eishima, N. Miyashita, Y. Nojiri, Y. Nakamura, WNISAT—
into the system design process. nanosatellite for north arctic routes and atmosphere monitoring, in:
To more thoroughly investigate the feasibility and The 24th Annual AIAA/USU Conference on Small Satellites, American
Institute of Aeronautics and Astronautics (AIAA), Logan, UT, 2010.
potential benefits of constellation deployment using [11] B.T. Patel, S. Schroll, A. Lewin, On-orbit performance of the ORB-
EML-1, an increased understanding of the trajectory design COMM spacecraft constellation, in: The 13th Annual AIAA/USU
and subsystem requirements and design is required. The Conference on Small Satellites, American Institute of Aeronautics
and Astronautics (AIAA), Logan, UT, 1999.
development of aerocapture or aerobraking devices for [12] C.R. Boshuizen, J. Mason, P. Klupar, S. Spanhake, Results from the
very small satellites and investigation of the aerodynamic Planet Labs Flock constellation in: The 28th Annual AIAA/USU
effects of these manoeuvres on these satellites may also Conference on Small Satellites, American Institute of Aeronautics
enable the use of this method for a greater range of and Astronautics (AIAA), Logan, UT, 2014.
[13] J.R. Wertz, W.J. Larson, Space Mission Analysis and Design, 3rd ed.
mission types. Microcosm Press, Kluwer Academic Publishers, El Segundo, CA, 1999
ISBN 978-1-881883-10-4.
[14] J. Esper, P.V. Panetta, M. Ryschkewitsch, W. Wiscombe, S. Neeck,
Acknowledgements NASA-GSFC nano-satellite technology for earth science missions,
Acta Astronaut. 46 (2–6) (2000) 287–296, http://dx.doi.org/10.1016/
S0094-5765(99)00214-3. ISSN 00945765.
This work was supported by the Doctoral Training [15] L. Dyrud, S. Slagowski, J. Fentzke, W. Wiscombe, B. Gunter, K. Cahoy,
Partnership (DTP) between the University of Manchester G. Bust, A. Rogers, B. Erlandson, L. Paxton, S. Arnold, Small-sat
science constellations: why and how, in: The 27th Annual AIAA/USU
and the UK Engineering and Physical Sciences Research Conference on Small Satellites, American Institute of Aeronautics
Council (EPSRC) under grant EP/J50032X/1. and Astronautics (AIAA), Logan, UT, 2013.
[16] D.J. Barnhart, T. Vladimirova, M.N. Sweeting, Very-small-satellite design
for distributed space missions, J. Spacecr. Rockets 44 (6) (2007)
Appendix A. Supplementary data 1294–1306, http://dx.doi.org/10.2514/1.28678. ISSN 0022-4650.
[17] D.J. Barnhart, T. Vladimirova, A.M. Baker, M.N. Sweeting, A low-cost
Supplementary data associated with this article can be femtosatellite to enable distributed space missions, Acta Astronaut.
64 (11–12) (2009) 1123–1143, http://dx.doi.org/10.1016/j.actaas-
found in the online version at http://dx.doi.org/10.1016/j.
tro.2009.01.025. ISSN 00945765.
actaastro.2015.04.015. [18] R. Sandau, K. Brieß, M. DErrico, Small satellites for global coverage:
potential and limits, ISPRS J. Photogramm. Remote Sens. 65 (6)
(2010) 492–504, http://dx.doi.org/10.1016/j.isprsjprs.2010.09.003.
ISSN 09242716.
References [19] W.W. Saylor, K. Smaagard, N. Nordby, D.J. Barnhart, New scientific
capabilities enabled by autonomous constellations of smallsats, in:
[1] H.J. Kramer, A.P. Cracknell, An overview of small satellites in remote The 21st Annual AIAA/USU Conference on Small Satellites, 719,
sensing, Int. J. Remote Sens. 29 (15) (2008) 4285–4337, http://dx.doi. American Institute of Aeronautics and Astronautics (AIAA), Logan,
org/10.1080/01431160801914952. ISSN 0143-1161. UT, 2007.
[2] D. DePasquale, A. Charania, H. Kanayama, S. Matsuda, Analysis of the [20] A. da Silva Curiel, M. Lambert, D. Liddle, M.N. Sweeting, C.-H. Chu, C.-J.
earth-to-orbit launch market for nano and microsatellites, in: AIAA Fong, G.-S. Chang, Introduction to FORMOSAT-7/COSMIC-2 mission, in:
SPACE 2010 Conference and Exposition, American Institute of Aero- The 27th Annual AIAA/USU Conference on Small Satellites, American
nautics and Astronautics (AIAA), Anaheim, CA, 2010. Institute of Aeronautics and Astronautics (AIAA), Logan, UT, 2013.
[3] G. Webb, A. da Silva Curiel, The changing launcher solutions of the [21] M. Bille, P. Kolodziejski, T. Hunsaker, Distant horizons: smallsat
small satellite sector, in: The 62nd International Astronautical evolution in the mid-to-far term, in: The 25th Annual AIAA/USU
Congress, International Astronautical Federation (IAF), Cape Town, Conference on Small Satellites, American Institute of Aeronautics
SA, 2011. and Astronautics (AIAA), Logan, UT, 2011.
[4] K. Woellert, P. Ehrenfreund, A.J. Ricco, H.R. Hertzfeld, Cubesats: [22] L. Lemmerman, C. Raymond, R. Shotwell, J. Chase, K. Bhasin,
cost-effective science and technology platforms for emerging and R. Connerton, Advanced platform technologies for Earth science,
developing nations, Adv. Space Res. 47 (4) (2011) 663–684, http://dx. Acta Astronaut. 56 (1–2) (2005) 199–208, http://dx.doi.org/10.1016/
doi.org/10.1016/j.asr.2010.10.009. ISSN 02731177. j.actaastro.2004.09.031. ISSN 00945765.
78 N.H. Crisp et al. / Acta Astronautica 114 (2015) 65–78

[23] N.H. Crisp, K.L. Smith, P.M. Hollingsworth, Small satellite launch to (1) (2014) 155–160, http://dx.doi.org/10.1177/0954410013476615, ISSN
LEO: a review of current and future launch systems, Trans. Jpn. Soc. 0954-4100.
Aeronaut. Space Sci. Aerosp Technol. Jpn. 12(ists29), ISSN 1884- [34] N. Chow, E. Gralla, N.J. Kasdin, J. Chase, Low Earth Orbit Constella-
0485, http://dx.doi.org/10.2322/tastj.12.Tf_39 . tion Design Using the Earth–Moon L1 Point, Technical Report,
[24] A. Charania, S. Isakowitz, W. Pomerantz, B. Morse, K. Sagis, Launch- Princeton University, Princeton, NJ, 2004.
erOne: revolutionary orbital transport for small satellites, in: The [35] J. Liu, R. Alford, Semianalytic theory for a close-earth artificial
27th Annual AIAA/USU Conference on Small Satellites, vol. 626, satellite, J. Guid. Control Dyn. 3 (4) (1980) 304–311.
American Institute of Aeronautics and Astronautics (AIAA), Logan, [36] D.A. Vallado, Fundamentals of Astrodynamics and Applications, 4th
UT, 2013. ed. Microcosm Press, Springer, Hawthorne, CA, 2013.
[25] K. Rodway, K. Papadopoulos, J. Greason, Utility and application of [37] J. Picone, A. Hedin, D. Drob, A. Aikin, NRLMSISE-00 empirical model
XCOR's commercial reusable suborbital vehicle Lynx for small of the atmosphere: statistical comparisons and scientific issues,
satellite launch, in: The 65th International Astronautical Congress, J. Geophys. Res. 107(A12), http://dx.doi.org/10.1029/2002JA009430,
International Astronautical Federation (IAF), Toronto, Canada, 2014. ISSN 0148-0227.
[26] C.-J. Fong, C.-Y. Huang, V. Chu, N. Yen, Y.-H. Kuo, Y.-A. Liou, S. Chi, [38] C. Pardini, L. Anselmo, Comparison and accuracy assessment of
Mission results from FORMOSAT-3/COSMIC constellation system, semi-empirical atmopshere models through the orbital decay of
J. Spacecr. Rock. 45 (6) (2008) 1293–1302, http://dx.doi.org/10.2514/ spherical satellites, J. Astronaut. Sci. 49 (2) (2001).
[39] D. King-Hele, Satellite Orbits in an Atmosphere: Theory and Appli-
1.34427. ISSN 0022-4650.
cations, Blackie and Son Ltd, Glasgow, UK, 1987.
[27] J. King, N. Beidleman, Method and Apparatus for Deploying a
[40] D.A. Vallado, D. Finkleman, A critical assessment of satellite drag and
Satellite Network, 1993.
atmospheric density modeling, in: AIAA/AAS Astrodynamics Spe-
[28] J. Puig-Suari, G. Zohar, K. Leveque, Deployment of CubeSat constella-
cialist Conference and Exhibit, The American Institute of Aeronau-
tions utilizing current launch opportunities, in: The 27th Annual AIAA/
tics and Astronautics (AIAA), Honolulu, HI, 2008.
USU Conference on Small Satellites, vol. 805, American Institute of
[41] D.A. Vallado, T. Kelso, Using EOP and space weather data for satellite
Aeronautics and Astronautics (AIAA), Logan, UT, 2013.
operations, in: The 15th AAS/AIAA Astrodynamics Specialist Con-
[29] J.W. Gangestad, B.S. Hardy, D.A. Hinkley, Operations, orbit determi-
ference, American Institute of Aeronautics and Astronautics (AIAA),
nation, and formation control of the AeroCube-4 CubeSats, in: The
Lake Tahoe, CA, 2005.
27th Annual AIAA/USU Conference on Small Satellites, American
[42] D.A. Vallado, T. Kelso, Earth orientation parameter and space
Institute of Aeronautics and Astronautics (AIAA), Logan, UT, 2013. weather data for flight operations, in: The 23rd AAS/AIAA Space
[30] R. Rose, W. Wells, J. Redfern, D. Rose, J. Dickinson, C. Ruf, A. Ridley, K. Flight Mechanics Meeting, American Institute of Aeronautics and
Nave, NASA's Cyclone Global Navigation Satellite System (CYGNSS) Astronautics (AIAA), Kauai, HI, 2013.
mission—temporal resolution of a constellation enabled by micro- [43] K. Moe, M.M. Moe, Gas-surface interactions and satellite drag
satellite technology, in: The 27th Annual AIAA/USU Conference on coefficients, Planet. Space Sci. 53 (8) (2005) 793–801, http://dx.
Small Satellites, American Institute of Aeronautics and Astronautics doi.org/10.1016/j.pss.2005.03.005. ISSN 00320633.
(AIAA), Logan, UT, 2013, ISBN 7347646561. [44] D.L. Oltrogge, K. Leveque, An evaluation of CubeSat orbital decay, in:
[31] C.-J. Fong, W.-T. Shiau, C.-T. Lin, T.-C. Kuo, C.-H. Chu, S.-K. Yang, The 25th Annual AIAA/USU Conference on Small Satellites, Amer-
N.L. Yen, S.-S. Chen, Y.-H. Kuo, Y.-A. Liou, S. Chi, Constellation ican Institute of Aeronautics and Astronautics (AIAA), Logan, UT,
deployment for the FORMOSAT-3/COSMIC mission, IEEE Trans. 2011, ISBN 6508594621.
Geosci. Remote Sens. 46 (11) (2008) 3367–3379. [45] S.W. Janson, The future of small satellites, in: H. Helvajian, S.W.
[32] J. Chase, N. Chow, E. Gralla, N.J. Kasdin, LEO constellation design Janson (Eds.), Small Satellites: Past, Present, and Future, The Aero-
using the lunar L1 point, in: The 14th AAS/AIAA Space Flight space Press, El Segundo, CA, 2008 (Chapter 23).
Mechanics Meeting, vol. 609, American Astronautical Society, Maui, [46] D.B. Scharfe, A.D. Ketsdever, A review of high thrust, high delta-V
HI, 2004. options for microsatellite missions, in: The 45th AIAA Joint Propul-
[33] M.J. Nadoushan, A.B. Novinzadeh, Satellite constellation build-up via sion Conference and Exhibit, vol. 15, American Institute of Aero-
three-body dynamics, Proc. Inst. Mech. Eng. Part G: J. Aerosp. Eng. 228 nautics and Astronautics (AIAA), Denver, CO, 2009.