Swarm satellites

Swarm satellites
Design, methods and applications
Radio astronomers, interested in low frequency signals are
troubled by Earth’s ionosphere, as it distorts and even
blocks signals below 20 MHz. This realisation lead to
studies to attempt to construct radio telescopes in space.
Aerospace Engineering
Two precursor missions were also launched (RAE1 and
RAE-2) in the early 70’s. They discovered Earth’s
ionosphere is also a very strong emittor or Radio
Frequency Interference. Also man-made signals were
detectable from space, causing RAE-2 to be launched into
lunar orbit. It detected that behind the Moon (relative to
Earth), the RFI was significantly reduced, to below the
detection threshold of the satellite. Launching a large
satellite into lunar orbit or beyond was deemed un-feasible
back then.
Recently, with the advent of distributed Earth-based radio
telescopes such as the WSRT, SKA and LOFAR, we have
seen a revival in studies of low frequency space-based
radio telescopes. The latest in this series is a proposal
called OLFAR (Orbiting Low Frequency Antennas for Radio
Astronomy), which proposes a swarm of nano-satellites to
form a virtual, distributed radio telescope in Lunar orbit or
beyond. The main difference in the OLFAR studies, and its
competitors is that it attempts to use (beyond) state-ofthe-art technologies, rather than proven space-qualified
technology, which would drive up the cost of the mission.
This in turn renders access to modern, high performance
processing and fast inter-satellite links, which allows for an
increase in the number of nodes in the array. The reduced
form factor is therefore less of a nuissance than it would
initially seem.
This causes a great deal of confusion as to what a satellite
swarm is, and one of my research goals is therefore to
formally distinguish satellite swarms from other forms of
distributed architectures, such as constellations and
formation flights.
The current consensus as to what a satellite swarm is, is
defined as:
“A satellite swarm is ideally defined as a
distributed space system consisting of many
identical, low-cost spacecraft, autonomously
cooperating to achieve a common global goal”.
What defines a swarm satellite?
Using the above definition, distinguishing features of a
swarm satellite can be identified. This allows defining the
properties of the individual satellites through a novel
“hybrid” systems engineering appoach, which uses a topdown requirements definition process, yet matches it with
a bottom-up element design process. This was required
due to an effect called “emergent behaviour”, which is
(unexpected) behaviour which only emerges due to
interactions between elements, and is therefore very
difficult to design for. A bottom-up approach can allow for
simulations to identifiy the interactions which cause
emergent behaviour, which can then be tuned to suit the
Swarm satellites
Satellite swarms are a novelty. My PhD research focusses
on design methods for satellite swarm elements (i.e. the
individual spacecraft which, when combined, form the
swarm). In order to structure this, several research
questions are defined. They can be categorised as follows:
What is a satellite swarm?
Unlike the mass-based definitions of monolithic satellites,
which have grown over the years to become an accepted
standard, distributed space systems are still in their
infancy, and terms such as swarms, clusters and
formations are used rather arbitrarily.
Preliminary element design of an OLFAR element
In general, a swarm satellite should be as simple as
possible. An overview of the preliminary element design for
an OLFAR spacecraft can be seen on the right. Fivemostly
independent functionality groups can be identified:
Artist impression of OLFAR in Lunar orbit
PhD Candidate: Steven Engelen
Department: SpE
Section: Space Systems Engineering
Supervisor: C.J.M. Verhoeven
Promoter: E.K.A. Gill
Start date: 1-4-2010
Funding: STW
Cooperations: UTwente, ASTRON
The power system
Attitude control and locomotion
Communication, intelligence and control
The payload
An end-of-life device
These can function almost independently, and should be
present in one form or other in any swarm satellite. The
end-of-life device is a special addition, to comply with
space-junk regulations, as satellite swarms will contain
many potentially unreliable elements, which would need
How to design the most basic swarm satellite (for the
OLFAR swarm)
This is the crucial question. A novel systems engineering
approach has been proposed, and it is being applied to the
OLFAR element requirements as defined by the OLFAR
team. Monte-carlo simulations of the lifetime estimates
have been performed for OLFAR-like swarms and swarm
satellites, rendering a tool to define the system- and
element (useful) lifetime, taking reduced operational states
into account, as well as the element reliability.
One such result is shown on the right. The satellites in this
siumation had an expected internal component lifetime of
around 3 years, which results in a useful lifetime of less
than 0,4 years, and an extended lifetime up to 0,45 year.
This lead to the conclusion the swarm satellite design used
in this model is fundamentally flawed, and an improved
version is currently being investigated.
Swarm lifetime simulation result for a swarm
of 100 satellites
Progress and Objectives
• Currently, a workable definition of a satellite swarm is
defined, and the impications are being investigated. This
would, in time, allow for an unambiguous distinction
between the different types of distributed satellite
• A novel systems engineering approach, tailored to
satellite swarms, has been proposed. It will still have to
prove its merits , yet initial results are promising.
• A throughough invesitgation into the reliability aspects
of using low-cost (nano-) satellites in swarms has been
performed, and is submitted to a Journal for peerreview. The tools developed will allow tailoring the
element design to the lifetime and relibaility
requirements of the overall swarm.
• The OLFAR satellites are currently undergoing their
preliminary design phase. Crucial elements, such as the
payload, are being developed. Those will be tested in
pathfinder missions, which will also determine the RFI
environment in the various candidate orbits.
• Using a satellite swarm for OLFAR is extremely
promising, but a lot of additional research and design
has to be performed before the mission can be
launched. Precursor missions are therefore already
being outlined.
Select publications
- S. Engelen, E. Gill, C. Verhoeven, “On the reliability of spacecraft swarms”, Proceedings of the 4S Conference, Portoroz, Slovenia, 2012
- S. Engelen, E.Gill, C. Verhoeven, “On the Reliability, Availability and Throughput of Satellite Swarms”, IEEE Transactions on Aerospace and Electronic Systems (March 2013, Paper
- C.J.M. Verhoeven, M.J. Bentum, J. Rotteveel, B. Monna and J. Guo, “On the Origin of satellite Swarms,” Acta Astronautica, 68 (7-8). pp 1392-1395, 2011. ISSN 0094-5765
- E. Dekens, S. Engelen, R. Noomen, “A Satellite Swarm for Radio Astronomy”, IWSCFF 7th International Workshop on Satellite Constellations and Formation Flying, March 13-15, Lisbon,
- S. Engelen, C.J.M. Verhoeven and E.K.A. Gill, “Systems Engineering Challenges for Satellite Swarms”, IEEE Aerospace Conference, March 5-12, 2011, Big Sky, Montana, US
- R.T. Rajan, S. Engelen, M.J.Bentum and C.J.M. Verhoeven, “Orbiting Low Frequency Array for radio astronomy”, IEEE Aerospace Conference, March 5-12, 2011, Big Sky, Montana, US

similar documents