Chapter 5

Miniaturization and Micro/ Nanotechnology in

Space Robotics
Evangelos Papadopoulos
1,2
, Iosif Paraskevas
1,3
, and Thaleia Flessa
1,4
1
Department of Mechanical Engineering, National Technical University of Athens
15780 Athens, Greece
email:
1
egpapado@central.ntua.gr,
3
isparas@mail.ntua.gr,
4
tflessa@mail.ntua.gr
Abstract
Space is an exciting but fundamentally unfriendly environment for humans. Space
robotic systems (robots in orbit, planetary rovers or even satellites) are of great
importance to space exploration and perform tasks hazardous or impossible for
humans. Using micro and nano technologies in space robotic systems results either
in miniaturized systems in terms of volume and mass, while retaining or increas-
ing their capabilities, or in space robots with increased capabilities while retaining
their size due to the nature of their tasks. Examples of miniaturization possibilities
for space robots and satellites are given, focusing on the challenges and the ena-
bling technologies. The miniaturization process and the use of advanced nano and
microtechnologies in space will have a large beneficial impact in the years to
come.
Keywords
Miniaturization, space robotics, satellites, microtechnology, nanotechnology,
MEMS, micromechanisms, nanorobots, microrobots, scaling, microsensors, mi-
croinstruments, sensor islands, nanosatellites, micropropulsion.
5.1 Introduction
Space is an exciting area of activities for mankind. These activities allow us an-
swering fundamental questions about the origins of the universe, assist our life on
earth (e.g. meteorological and GPS satellites) and improve our scientific and tech-
nological capabilities resulting in a wide range of inventions and new processes.
However, space is a highly unfriendly environment for humans. Harmful radia-
tions, extreme temperatures, lack of suitable or lack of any atmosphere, huge dis-
tances to be covered, and long communication lags and interruptions, are only a
few of the factors that render space a hostile environment for humans. It is well
known that the need for astronaut extravehicular activities (EVA) increases the
cost of a mission dramatically, due to the life support systems and precautions that
must be taken to ensure astronaut safety during EVA.
A solution to these problems is the use of robotic devices, capable of operating
for long times with minimum supervision or even autonomously. Such devices
have been developed during the last fifty years and include robotic spacecraft, on-
orbit robotic arms, rovers for planetary exploration, robots for space structures,
and satellites. Additional plans for robotic space hardware of all types are in the
development phase. Clearly, the future of space exploration and commercializa-
tion will include robots as a vital enabling technology.
Space robotic devices tend to be very big in size. For example, in orbital sys-
tems, the Canadarm of the Space Shuttle is 15 m long, weighs 431 kg, and handles
payloads up to 14 tons [1]. The reason is that it must be able to handle satellites
and other equipment, which are of large dimensions due to the instruments they
carry, or due to the need for being compatible with the human scale. For explora-
tion systems, the small Sojourner Mars rover was also at the human scale, weigh-
ing about 10.6 kg with its dimensions being 0.65 m high, 0.30 m wide and 0.48 m
long [2]. Subsequent rovers were bigger and heavier by at least one order of mag-
nitude due to the need for carrying instruments and for travelling larger distances.
On the other hand, on earth an important current trend is miniaturization of de-
vices and processes and the capability of acting at very small scales, including and
in some cases less than the nano scale. Such miniaturization offers significant
gains in volume, mass, and power for devices, leads to materials with unparalleled
properties, and capabilities of intervening at the cell level. Miniature mobile ro-
bots at the scale of a few cm and with resolution of a few microns have been de-
veloped to provide mobile microinstrumentation, to assist in cell manipulation in
biological experiments, and in micromechanism or microelectromechanical sys-
tem (MEMS) assembly. Besides mobility, these robots include capabilities such as
object manipulation with force feedback, two way communications, on board con-
trollers, visual feedback, and significant power autonomy. In addition to the above
developments, smaller robots operating at the nanometer scale are being envis-
aged. The size of such nanorobots may be in the range of 0.1-10 micrometers and
could be implemented through the use of controlled biological microorganisms [3]

or the use of magnetic nanocapsules steered by improved gradient coils provided
by Magnetic Resonance Imaging (MRI) systems [4].
A natural question appears then regarding the degree to which space robotic ac-
tivities are affected by the miniaturization trend observed on terrestrial applica-
tions in general, and more specifically in robotics. One can identify similarities
and differences. For example, in robotics the terms micro or nano refer either to
the size of the robot itself in terms of micrometers or nanometers, or to the resolu-
tion in micrometers or nanometers that this robot has in dealing with its environ-
ment. This is because human scale robots that operate in human scale tasks are
taken as the basis for comparison. However, in space the scale of tasks is very dif-
ferent and to be more exact, much bigger. Both space exploration and orbital sys-
tems scales in terms of orbits, distances, power requirements, etc. are very large
compared to the terrestrial human scale. Therefore, despite the evolving miniaturi-
zation process, the same terms refer in most cases to different scales of magnitude.
This can be seen clearly in the case of nano or even picosatellites, both of which
will not qualify as such according to terrestrial robotic terminology.
Miniaturization affects space robotic systems in two fundamental ways (a) by
reducing the size and weight of components, thus allowing a space robotic system
to increase its overall capabilities by maintaining more or less its size, and (b) by
reducing its overall size, maintaining the same capabilities. In the first case, the
robotic system cannot be reduced in size, because the objects it operates on remain
of large size, for example the case of the International Space Station (ISS) robotic
manipulators. In the second case, the size can be reduced by maintaining more or
less, the initial capabilities; this is the case of nanosatellites. As mentioned earlier,
the base scale is large enough so that any size reductions at the device or compo-
nent scale will not qualify as nanodevices on earth.
In the subsequent sections, we examine the existing space robotics applications
and classify the robots that are being used. The ways in which these robots can be
miniaturized, using our terrestrial experience are addressed. The interesting sub-
ject of satellite miniaturization and subsequent proliferation is presented next. Fi-
nally, we outline the future trends for miniaturization of space robotics systems
and components and address briefly the challenges ahead.
5.2 Space Robotics Applications
Usually the term space robotics is used to characterize semi or fully autonomous,
teleoperated exploration and servicing systems, and a few specialized experi-
mental systems. However on a broader sense, almost everything that has been sent
into space, integrates a form of automation in some degree, and therefore can be
characterized as automated or robotic. The emphasis here is given to systems
that are flight proven, they still operate, or are in the final construction phases and
the state-of-the-art in their class.
5.2.1 Exploration
From the very beginning of the space exploration age, rovers were considered as
the main exploration platform and appeared early in the space programs of both
the USA (with the Surveyor Lunar Rover Vehicle, 1963) and the USSR (with
Lunokhod, 1970). The former program was cancelled and substituted by the Lunar
Roving Vehicles of the Apollo program, while the latter (Lunokhod 1 in 1970)
was the first remote controlled rover to land on the moon [5]. NASA has also sent
to the Moon stationary explorers (Rangers and Surveyors [6]).
Since the late 70s, the exploration focus turned to Mars and other distant bod-
ies. At first, it was necessary to achieve the goal of reaching the target, e.g. the
Viking Landers in 1975 [7]; however it appeared soon that the scientific objectives
should not be limited to the landing area. Time delays due to the distance to Mars
rendered teleoperation impossible; this paved the way for semi and fully autono-
mous systems. Due to technological limitations, early designs such as Marsokhod
and Robby were massive and frequently impossible to fit in the available launch-
ers [8]. Smaller systems could not be autonomous and/ or integrate adequate sci-
entific instrumentation.
The radical changes in microtechnology enabled prototype miniaturization,
which lead in 1996 to the first flight proven planetary explorer with advanced ca-
pabilities, the Sojourner Rover [2]. After some unsuccessful efforts, the Mars Ex-
ploration Rovers Spirit and Opportunity were sent to Mars in 2003 [9]. The two
identical rovers were larger than the Sojourner; however they included more sci-
entific instrumentation and were capable to travel for kilometers. Today, rovers
are a main programmatic goal of all space agencies. Primary missions include the
Mars Science Laboratory (MSL, 2011) [10], and the NASA and ESA Exomars
and Mars Sample Return cooperative missions, still in the design phase [11]. Sta-
tionary explorers, such as ESAs Beagle 2 [12] and the successful Phoenix Mars
Lander [13] were and still are in agency programmatics.
A number of alternative concepts are tested for planetary exploration, which
include surface explorers based on legged locomotion [14], [15], the exploitation
of swarm capabilities [16], [17], or flying systems [18]. However, no such work-
ing prototype that has operated in space exists currently. Planetary orbital explor-
ers, such as the Mars Odyssey, are examples of relatively small autonomous sys-
tems [19]. Other interesting concepts in this category include those that release a
landing probe (to a planet, asteroid or comet) like Cassini-Huygens [20], Deep
ImpactEPOXI [21] and Rosetta [22], or that land themselves, such as NEAR
Shoemaker [23] and Hayabusa [24].
From the above examples, a trend is observed, from the large and massive sys-
tems to the small and lightweight ones and again towards larger exploration rov-
ers, the difference in mass is obvious; Table 5.1 and Fig. 5.1 show clearly this
trend as exhibited by NASAs rovers. The first rovers were large because of the
then available technology. The development of microtechnology enabled minia-

turization and development of designs that could be accommodated in a launcher.
At the same time, the requirements of the scientific community were the main de-
sign consideration for the development of larger systems. The size of the MSL is
comparable to that of the Rocky I; however, earlier designs had almost no space
for scientific payloads.
Table 5.1 - Characteristics of different rover generations from JPL.
Rover
Mass
(kg)
Instrument
Mass (kg)
Average
Speed (m/h)
Distance
Travelled (km)
Largest Object
Over (m)
Approx.
Volume (m
3
)
Sojourner (1996) 10.6 < 4.5 3.5 0.1 N/A 0.7x0.5x0.3
MER (2003) 176.5 6.8 34.0 Spirit 8, Op-
portunity >28
0.26 1.5x1.6x2.3
MSL (2011) 900.0 80.0 30.0 N/A 0.75 2.7x2.9x2.2


Fig. 5.1 - Full-scale models of three generations of exploration rovers. Sojourner (centre),
MER (left) and MSL (right). (Courtesy of NASA/ JPL-Caltech).
5.2.2 Servicing
To sustain a constant presence in space and to enable longer exploration missions,
it is necessary to develop servicing systems. Such systems are used in servicing or
removal of malfunctioning satellites, removal of space debris, inspection and con-
struction of space structures, or, in the future, in astronaut assistance at planetary
outposts. These robotic systems must be able to manipulate objects with various
characteristics, making them more complex.
Currently, the most successful space robots, such as the Shuttle Remote Ma-
nipulator System (SMRM) or Canadarm, the Space Station Remote Manipulator
System (SSMRM) or Canadarm 2 and the Special Purpose Dexterous Manipula-
tor (SPDM) or Dextre on ISS [1], Fig. 5.2a, are mainly teleoperated via dedicat-
ed remote interfaces. The European Robotic Arm (ERA) [25] (est. launch in 2012)
and the Japanese Experiment Module Remote Manipulator System (JEMRMS)
[26], currently at the ISS, have a greater level of autonomy but still require a hu-
man operator.
To overcome the increasingly important problem of space debris and malfunc-
tioning satellites, a greater level of autonomy is required. The first mission that
demonstrated autonomous in space servicing was JAXAs ETS-VII; Fig. 5.2b
[27]. NASA and DARPA extended the concept, initially with the unsuccessful
DART mission and later with the successful Orbital Express mission [28]. ESA
and others pursue similar designs for on-orbit servicing, like the TECSAS and the
DEOS [29]. Additionally, ESA and JAXA have developed fully autonomous hu-
man-class systems for logistic purposes for the ISS. Both the Automated Transfer
Vehicle (ATV) [30] and the H-II Transfer vehicle operated smoothly [31].


(a) (b)
Fig. 5.2 - (a) Dextre on ISS (Courtesy of NASA) and (b) ETS-VII prior to launch (Courtesy
of JAXA).
5.2.3 Experimental Systems
Space agencies, R&D institutes, universities and companies have developed a
number of designs to test future space robotic systems and enabling technologies.
The systems that have been tested in space are presented here.

During the STS 55 mission (1993), various control modes in space were test-
ed using DLRs ROTEX [32], a small, six-axis robot equipped with a gripper and
mounted inside a space-lab rack. On behalf of DLR, the Robotic Components Ver-
ification on the ISS (ROKVISS) was installed in 2005, and removed in 2010. Its
main purpose was to update and validate space robot dynamic models, to verify
DLRs proprietary modular joints and to verify telepresence methods [33]. The
most recent experimental system in space is the Robonaut 2 by NASA and GM
[34], which will test the performance of an anthropomorphic robot, resembling fu-
ture robots that can replace astronauts, while retaining at the same time the dexter-
ity of the human body.
The need for miniature robotic systems to assist the astronauts and capable of
performing a number of tasks, lead to the development of the AERCam Sprint
[35] and the MIT SPHERES [36], Fig. 5.3. Their performance proved that small
autonomous robots could indeed assist humans. Various small robots have been
proposed since then, but few designs have been tested. An interesting example is
Robyspace [37], which was developed to examine the capabilities of small robotic
systems operating in specialized nets. The integration of a large number of elec-
tronics for sensing and actuation in a small volume enabled the development of all
these experimental systems. However, a greater degree of integration is necessary
in order to increase redundancy and functionality.


(a) (b)
Fig. 5.3 - (a) AERCam Sprint, (b) SPHERES in on-orbit experiments. (Courtesy of NASA).
5.3 Robotics Miniaturization in Space
5.3.1 Motivation
The wealth of available micro and nanotechnologies (MNT), such as MEMS and
microelectronics, the exciting and innovative prospects they present combined
with the importance of space robotics for space exploration, leads us to the con-
clusion that the introduction of micro and nanotechnologies in space robotics will
significantly augment the capabilities of space robotics and will pave the way for
further space robotic exploration missions. Space robotics is an area of particular
importance to future missions; the introduction of MNT, either sourced from ter-
restrial applications (spinning-in) or specifically designed for space robotics, re-
sults in their overall miniaturization. However, as noted earlier, the term miniatur-
ization when applied to space robotics has a somewhat different meaning than the
traditional one. Therefore, for such systems the miniaturization process refers to
(a) reduction of component size and mass, with increased functionality and un-
changed total size and (b) reduction in overall size, while maintaining capabilities.
The terms MNT and microsystems are used in the present chapter with a
broader meaning; they encompass MEMS technologies, nanotechnologies and
miniature or microrobotic technologies, collectively referred as MNT or microsys-
tems. Electronic devices (e.g. processors) are not generally recognized as exam-
ples of microsystems, however, in order to benefit from the reductions achieved
with microsystems and from the integration of devices and electronics, they are
considered here as microsystems. Additionally, any enabling technologies that al-
low for the miniaturization of space robotics are also of interest. Further clarifica-
tion on the terms MEMS and microsystems can be found at [38].
There are several reasons why the miniaturization of space robotics is desira-
ble. An important motivation for the systematic miniaturization of robotic systems
is that launch vehicles have tight constraints with respect to the payloads mass
and volume characteristics and therefore successful miniaturization directly results
in an improved, more compact and less expensive system. The cost to place one
kg in Low Earth Orbit is approximately $10K and can rise up to $20K; this cost is
significantly increased for long space exploration missions (e.g. Moon, Mars, as-
teroids, etc.), therefore any weight and volume reduction would significantly de-
crease the launch cost. Another benefit of the miniaturization process is that the
total resources required for space systems (e.g. mass, volume, power) are substan-
tially reduced. Furthermore, systems such as proprioceptive and exteroceptive
sensors, wireless communications, control units, power generation and transmis-
sion units can be integrated into small packages at a system level, thus allowing
for a substantial increase in payload, reduction in power losses and more efficient

thermal management. Integrating several microsystems into a silicon wafer and in-
troducing redundancy by design, results in increased reliability and flexibility,
lower risk and greater functionality compared to conventional robotic space sys-
tems. Silicon wafer microsystems are subject to economies of scale and therefore
reduce the overall cost. Additional benefits of terrestrial microtechnologies in par-
ticular often include better performance compared to those in space, significantly
smaller development costs since investments on non space technology exceed by
orders of magnitude those of space, and sustainability of capabilities, reliability
and strict quality procedures due to the presence of strong markets.
A successful miniaturization (i) results in a compact system, (ii) reduces the re-
quired power budget, (iii) reduces the development cost, (iv) requires fewer re-
sources and smaller testing facilities, (iv) lessens complexity and improves overall
performance. These benefits are particularly important for space applications,
therefore the adaptation and implementation of micro technologies in space robot-
ics is of great interest and expected to yield substantial benefits.
The field of MEMS is the most important in terms of commercialisation, range
of applications and technological maturity. The worldwide market for MEMS,
boosted by automotive applications and by mobile handsets, gaming controllers,
digital cameras and other consumer electronics devices, exhibits nowadays an in-
creasing growth and has expanded to cover nearly all critical technological do-
mains. The MEMS market reached $6.9B in 2009, approximately $8B in 2010,
and is expected to be $9B in 2011, with an expected Compound Annual Growth
Rate (CAGR) for the next five years equal to 13% [39]. The MEMS accelerome-
ter, gyroscope and inertial measurement unit (IMU) market and in general the mo-
tion sensing industry is especially robust. For 2011, the MEMS gyroscope market
is estimated at $1B and the MEMS accelerometer market at $1.3B [40]. This
growth will be mostly driven by the deployment of more motion control user in-
terfaces on consumer electronics and drop detection and protection features in
portable systems. Table 5.2 shows the MEMS market volume in millions of units.
Table 5.2 - MEMS Market Forecast 2011 2015 in Millions of Units [39].
MEMS Device 2011 2012 2013 2014 2015
Accelerometers 1362 1630 1838 2002 2194
RF MEMS 805 987 1190 1551 2023
Inkjet Heads 673 683 691 706 719
Microphones 579 816 995 1186 1381
Pressure Sensors 389 488 571 633 689
Digital Compass 340 441 542 652 770
Gyroscopes 325 402 510 587 722
Microfluidics for IVD (In Vitro Diagnostics) 312 394 493 621 785
Oscillators 72 138 248 421 674
Microdispensers 55 68 118 141 168
Projection Systems 9 21 44 79 136
Optical MEMS 2.2 3.7 6.6 12.8 18.3
Micro displays 2 5 10 15 20
Microbolometers 0.3 0.4 0.4 0.6 0.7
Other (Microstructures, microtips, flow meters,
micro speakers, microfluidics for research)
6 15 43 112 209

The major terrestrial technological areas that employ MEMS are automotive,
aerospace, defence, industrial processes, consumer products, biotechnology and
telecommunications. Solutions provided by MEMS are finding their way into an
increasing number of automation and robotics applications, such as motion sens-
ing, impact detection, and rollover prevention. Commercial microsystem indus-
tries provide reliability procedures and yield management systems for micro-
systems and microsensor systems that offer excellent data return performance, re-
dundancy and reduced power budget integrated in a very small package. A good
example are inertial navigation systems, where inertial measurement sensors from
the automotive industry and complete inertial measurement units from oil drilling,
offer increased performance, and reduced power and space requirements. MEMS
and microsystems have a large and diverse market that enables cost reduction and
sustainability of available resources. It is also worth noting that the fraction of the
MEMS market solely dedicated to space applications is very small, often the
MEMS that have been used in space have been developed exclusively for the tar-
geted application; the situation is directly opposite to the case of solar cells, where
solar technology was first used in space and then on terrestrial applications.
5.3.2 State-of-the-Art
Microtechnologies for space systems in general and robotics in particular, have at-
tracted the interest of the scientific community during the past decade. The Jet
Propulsion Laboratory (JPL) discussed the role of MEMS in the development of
smaller robotic systems in 1999 [41] and the MEMS technology developments at
JPL, such as LIGA based devices, micro-propulsion, microvalves, optics, microac-
tuators, system on a chip, microinstruments, biomedical devices and packaging
were analyzed in [42]. A 1998 work also discusses the use of MEMS and micro-
technologies in propulsion, inertial navigation and wireless sensors in space sys-
tems [43]. More recently, the use of microrobotics in space, and the use of MEMS,
NEMS (Nano Electro Mechanical Systems) and microtechnologies towards a min-
iaturized robot in terms of mass and volume has attracted a lot of interest [44].
While not strictly related to the use of microtechnologies [45] discusses the use of
length scaling in space dynamics; a method for simulating the orbit and attitude of
small objects and therefore provide insights for the dynamics of very small space-

crafts ESA has organized several roundtables on MNT for Space applications [46].
Finally, the results of a recent ESA initiative for the introduction of terrestrial
MNT to space robotics can be found at [47]. In addition to the small exploration
systems, several studies exist on the design of miniaturized space systems, such
as: the mobile micro-robot Alice [48] developed at cole Polytechnique Fdrale
de Lausanne (EPFL), ESAs Nanokhod exploration rover [49] and the spider in-
spired climbing robot (Abigaille-I) for space at [50].
Despite these reports and studies and the benefits highlighted in the above sec-
tion, microtechnology is still sporadically used in space robotics. An overview of
the more developed areas of microtechnology in space is at [51] and [52] provides
a thorough review of the current and future use of MEMS and microsystems in
space systems. The most developed areas of use in space are inertial navigation,
where accelerometers and gyroscopes (e.g in the current Mars rovers) are sourced
from commercial and military applications, atomic force microscopes and propul-
sion. MEMS based propulsion that produces small thrusts in the order of to 1N
(micropropulsion), especially cold gas thrusters and ion thrusters (colloid and
FEEP thrusters), is particularly suitable for fine control, positioning formation fly-
ing applications and for primary acceleration of small spacecrafts [53-58]. NASA
has used miniature science instruments for the Phoenix Mars Mission (2008) and
for the Mars Science Laboratory (MSL) mission, which will land and operate the
Curiosity rover and was launched in November 2011 [59]. Finally, the James
Webb Space Telescope (est. launch 2014) will use a MEMS based microshutter
array for the Near InfraRed Spectrometer (NIRSpec) of the telescope [60, 61]. Ta-
ble 5.3 provides a summary of the MEMS technologies flown in space and their
technology readiness levels (TRL) [51]. Devices that were used in satellites (Cu-
beSats, ST5, Delfi C3) are discussed in Section 4.
Table 5.3 - MEMS in space applications and estimated TRL [51].
MEMS Device Flown in space? Estimated TRL
Inertial Navigation Yes High
Pressure Sensors Yes (Launch vehicles, propulsion) High
Magnetometer Yes (CubeSats) High
Atomic Force Microscope Yes (Phoenix mission) Medium High
Sun sensor Yes (Delfi C3) Medium High
Micro-fluidics Yes (Space shuttle, satellites) Medium
Bolometer Yes (Planck 2009) Medium High
Optical Switching No Medium High
Propulsion: ion, cold gas, colloid, solid Yes (ST5, small satellites) Medium
Thermal Control Yes Medium
RF switch and variable capacitor Yes (2000, OPAL picosatellites) Low Medium
Adaptive Optics & MOEMS instruments James Webb Telescope (est. 2014) Low Medium
MEMS Oscillator No Low Medium
5.3.3 Miniaturization Challenges
In order to assess the areas where the introduction of miniaturization in space ro-
bots is essential, it is necessary to identify the main challenges during design and
operation. The candidate areas for miniaturization in space robotics should be
searched within their subsystems components. The difference between the target
applications for microrobotics on Earth and space robotics must be stressed again;
a robot cannot be miniaturized if this negatively affects its objectives, e.g. a rover
should be able to travel for kilometers and gather samples, this is not possible with
a microrover. The miniaturization challenges interest greatly the space agencies,
such as the work presented in [47].
Space systems in general and space robotics in particular have the following
sub-systems: Power, Propulsion, Structure, Attitude and Orbital Control (AOCS),
On Board Data Handling (OBDH), Locomotion, Guidance, Navigation and Con-
trol (GNC), Communication, Thermal, Manipulators and End - Effectors [62].
Depending on the application of each space robot, a subsystem might not be appli-
cable, for example a rover requires no propulsion; however this categorization is
standard for all space systems. The subsystems that have the greatest demands (i)
in terms of mass and volume are Power, Propulsion and Structure, (ii) in terms of
required computational power are OBDH, AOCS, GNC and the motion of manip-
ulators, and (iii) in terms of power demands from the Power subsystem are Ther-
mal, Locomotion and Propulsion. Therefore, the focus of R&D in Micro & Nano-
technology (MNT) should be given in these areas to efficiently address the current
challenges. Table 5.4 presents the main challenges per space robot class and where
the introduction of MNT is expected to have the highest impact.
Table 5.4 - Challenges for MNT R&D per space robot class.
Class Subsystem Challenges for Micro- & Nano- Technology
1. Rovers/ Other
Means of
Locomotion
Power
Solution for solar cell efficiency decrease by dust
Large and heavy batteries: higher density required
OBDH
Low computational power: More efficient electronics, decen-
tralized architectures
Thermal Improved materials, spot cooling/ heating
Actuation
Low eff. DC motors: Better materials required
Low integration of electronics
Navigation Slow in rough terrains due to computational restrictions
2. Stationary
Planetary Explorers
Power See Class 1 above
Propulsion
Landing with retros has mass and volume penalty: Better
properties for fuels
Mechanisms See Class 6 below
Thermal See Class 1 above
3.Orbital Planetary
Explorers
Power and
OBDH
See Class 1 above
AOCS
High electrical power consumption: better electronics re-
quired

More efficient and smaller sensors required
Comms High power consumption: better electronics req.
Propulsion
50% of mass & volume of the system: better fuel properties
Electrical propulsion requires high power reqs.
Structure
20% of total mass: lightweight materials
Enhanced properties required: active materials
4.Aerobots/
Balloons
Power
No efficient flexible solar cells
Large Batteries: See Class 1
Structure
Restrictive space for all the necessary subsystems
Corrosive environments: protection required
AOCS,
OBDH,
Comms,
Thermal
High integration needed: better electronics required
High efficiency required: better sensors required
Propulsion Use mainly of propellers: See Actuation of Class 1
Power Highly efficient flexible cells required
5.Orbital Servicers
AOCS
See Class 3 above
Reqs. during rendez-vous, docking and manipulation
Comms See Class 3 above
OBDH
See Class 1
Reqs. during rendez-vous, docking and manipulation
Propulsion See Class 3 above
Structure See Class 3 above
GNC
Massive sensors
High power consumption
Manipulators
See Class 6 below
Computational intensive: complex dynamics
6. Manipulators
Sensors Integration of more sensors for high autonomy
Structure
Large mass
Massive and confusing cabling
Increased stiffness of cabling affects movement
Actuation See Class 1 above
End Effector
End effectors designed for specific tasks
Higher dexterity and sensory information required
5.3.4 General Selection Criteria
Not all MNT components can be used in miniaturizing space robotic devices, as
their reliability in space condition varies. This is especially pertinent to those sys-
tems that are sourced from terrestrial applications. A set of criteria for the selec-
tion of microsystems (MEMS, micro and nanotechnologies, etc.) is presented here.
To access the compatibility of the selected MNT components with the space
environment and functionality requirements of space robotics the following crite-
ria are proposed: (i) applicability to space robotics, (ii) launch conditions, (iii) ex-
ternal space environment requirements (LEO, GEO, Mars, Moon, Near Earth As-
teroid), (iv) required technical lifetime. Additionally, microsystems that are part of
the scientific payload must also comply with the scientific objectives of the exper-
iment. Each component should be able to withstand mechanical shocks of 6000 to
10000g and be able to at least operate in a temperature range of -50C to +80C. A
study on the reliability of MEMS under vibration and shock can be found at [63].
Vacuum conditions are detrimental to MEMS performance and out-gassing in a
vacuum environment has also adverse effects on a devices performance; howev-
er, it has been observed that a nitrogen atmosphere inside the MEMS packaging
has a positive effect on the devices performance and reduces drift. An additional
criterion is the maximum operating voltage, which is limited at approximately 2
kV for space applications due to electrical insulation specifications. Space radia-
tion is an important problem that is experienced by all structures operating in
space. The high energy particles present in space radiation can trigger single-event
effects (SEE) in all digital electronics. MEMS based on capacitive sensors (accel-
erometers, gyroscopes, proximity sensors) exhibit certain problems when exposed
to radiation, due to their operating principle; radiation effects result in creating
output drifts and generate noise; packaging is not always a sufficient solution, es-
pecially when volume and mass limitations are imposed. The effect of radiation on
devices in space is a subject of great interest [52, 64]. Reference [64] provides also
a list of radiation tested MEMS and microsystem devices.
For each microsystem, it is important to consider the development risk, time,
and cost required to reach the maturity necessary for use in space. Currently, there
is no general qualification process for space MNT, it is done on a case by case ba-
sis and usually there is no volume production of those devices that are space com-
pliant. The general standards for European space activities can be found at [65],
while a study on the reliability of MEMS in space can be found at [66]. Problems
are usually addressed by correct design while the applicability of packaging tech-
niques of MEMS devices for space is limited; for example the packaging of heter-
ogeneous MEMS is problematic because the metallic parts cannot withstand the
high temperatures of the packaging process. The reliability of MEMS is discussed
in [67]. As a final guideline, a terrestrial component that has been tested and veri-
fied by being used in the industrial or commercial sector would require an addi-
tional 10% cost to be made space compliant. If the terrestrial component or tech-
nology has not been extensively tested, the cost of technology transfer to space in-
creases above this 10% figure, and in proportion to the number of tests required.
The proposed set of criteria will aid in the selection of MNT components for
space robotics, however ideally a streamlined selection process would enable the
miniaturization of space robotics to a significantly greater degree.
5.3.5 Enabling Technologies
MNT can greatly benefit space robotics; the miniaturization of critical components
of the subsystems provide to the designers more solutions and flexibility during
development. The efficient, systematic introduction of MNT to space robotics re-

quires the fusion of the challenges and requirements of the future robotic systems.
The most important technologies and how they will affect space robots and their
subsystems are outlined here.
Sensor Islands: Sensor Islands are known as Power and Computational Auton-
omous Remote Sensors, a research area highly pursued and MNT dependent. This
concept is very important because it can increase the autonomy and flexibility of
space systems. A sensor island should be able to: (a) receive or harvest power with
minimal cabling, (b) have high electronics integration, (c) be computational au-
tonomous and perform data fusion and signal processing, without need to send or
receive any data, except for the final data packages, (d) wirelessly communicate
with a central computer for the overall control and (e) integrate sensors of differ-
ent functionalities. In this way, the computational architecture becomes complete-
ly decentralized and the overall system more compact and robust.
Power: The upgrade of current power subsystems is strongly related to the ad-
vances in power density for batteries (Ah/kg) and higher efficiency for solar cells.
Nanomaterials and microelectronics can harvest and exploit more energy than cur-
rent technological solutions, reducing the mass and weight of current power pro-
duction and storage systems. Efficient electronics reduce the power requirements
and have positive impact in power management.
Structure: The target technology is the development of structural elements with
advanced capabilities. Robust but lightweight materials like carbon fiber rein-
forced plastics (CFRP) are already available and in production; however their ca-
pabilities should be augmented by ejecting specialised nanoparticles and general
use of nanotechnology and nanomaterials. These new structural elements can be
combined with techniques for embedding sensors, cables and piping inside the
structure, thus lowering mass penalties and thermal losses and increasing flexibil-
ity and environmental protection. The technology of electroactive polymers (EAP)
and piezoelectric elements can also be used as sensors and/ or actuators (e.g. as
vibration suppressors).
OBDH & GNC: Developments here should be aimed towards a decentralized
architecture; therefore the enabling technologies are based on those for Sensor Is-
lands. Dedicated image processors would lower the computational burden, and
miniaturized cameras and optics would render the navigation capabilities more ef-
ficient. There are MNT systems with small footprints and low consumption offer-
ing superior functionalities for commonly used sensors, such as GPS, gyroscopes
and IMUs. Additionally, for electronics and microcontrollers 64 bit solutions
would increase the overall computational power.
Actuators: Reduction of motor volume and mass require advanced materials
enhanced with nanoparticles. They new materials could lower the power losses
and increase the magnetic flux and therefore the produced torque and/ or speed.
Additionally, higher integration would minimize the essential electronics volume.
All subsystems could benefit by the introduction of MNT and terrestrial MNTs
exhibit a significant potential for use in space systems. In general MNTs can lead
to smaller, lighter, less power consuming and with higher functionality parts,
which in turn means: (a) lighter and more compact systems (without affecting ca-
pabilities), (b) more space for payload (e.g. scientific instrumentation, cargo, etc.),
(c) higher autonomy capabilities, (d) higher security, operational flexibility, great-
er redundancy, (e) lower development costs and time and (f) lower launch costs.
5.4 Micro/ Nano Satellites
5.4.1 State of the Art
MNTs and miniaturization in terms of weight and volume also have been intro-
duced in satellites. Although satellites are not strictly considered as robotics sys-
tems, they are automated space systems, and therefore pertinent to this chapter.
The satellite market is the most mature and well-known segment of space systems.
Since the launch of Sputnik 1 in 1957, more than 4900 launches have placed ap-
proximately 6000 satellites into orbit, of which, as of April 2011, about 957 are
operational per the latest available satellite database available at [68]. Of those op-
erational, 463 (49%) are in Low Earth Orbit, (LEO, 160 2000 km), 397 (41%)
are in Geostationary Orbit, (GEO, 36000 km), 63 (6%) are in Medium Earth Orbit,
(MEO, 2000 36000 km) and 34 (4%) are in Elliptical Orbits. These satellites are
of government, military, commercial or civil nature. Their uses in orbit are shown
in Table 5.5, using data from [68]. The majority (93%) of commercial satellites
are communication satellites, 4% are Earth Observation/Remote Sensing satellites,
2% are Technology Development, and 1% are for Navigation Demonstration (pro-
totype satellites for the Galileo system).
Table 5.5 - Function of satellites on orbit.
Function Percentage
Communication 59%
Earth Observation/ Remote Sensing 9%
Navigation 8%
Military Surveillance 7%
Astrophysics / Space Science 5%
Earth Science/ Meteorology 4%
Other 7%

For the next decade, there will be an average of 122 satellites launches per
year, a 60% increase compared to the average annual rate of 77 per year in the
2000s, with a total of 1220 satellites build in the decade 2010 2020. The total

revenue from the manufacturing and launch of these 1220 satellites will reach
$194 billion worldwide for the decade 2010-2020, while currently 60% of the total
5B annual revenue of the European space industry comes from the manufacture
and launch of communications satellites. The average satellite mass is estimated to
be 1890 kg in the coming decade [69, 70]. For the satellites currently in orbit, the
average wet mass at launch (mass including fuel at launch) is 2139 kg and the av-
erage dry mass (mass in orbit) is 1190 kg. Table 5.6 shows the mass (wet mass)
distribution of the satellites in orbit using data from [68].
Table 5.6 - Mass distribution of satellites on orbit.
Weight (kg) (wet mass) Percentage
<500 26.40%
500 - 1000 13.69%
1000 - 1500 8.98%
1500 - 2000 5.37%
2000 - 2500 9.75%
2500 - 4000 18.07%
4000 -5000 11.39%
>5000 6.35%
5.4.2 Miniaturization efforts in satellites
It is clear from the above statistics that the satellite market is growing fast and the
average mass is quite high. Similarly to space robotic systems, the cost/ launch is
proportional to the wet mass of each satellite and higher mass means higher sys-
tems and development complexity. It is clear that any mass and volume reduction
would significantly decrease the launch cost. Within this context, the term minia-
turization of satellites in terms of mass and volume and has a different meaning
compared to the traditional one, used in terrestrial applications.
The miniaturization in satellites is achieved in three ways: (a) by scaling down
the satellites mass and volume while retaining functionality (small satellites), (b)
by implementing micro/ nano technologies for the subsystems, mostly in the form
of microelectronics and MEMS and (c) by combining (a) and (b). A small satellite
is defined as a satellite of wet mass of less than 500 kg. Within this range, a mi-
crosatellite has a wet mass between 10 and 100 kg, a nanosatellite between 1 and
10 kg and a picosatellite between 0.1 and 1 kg. Compared to the average wet mass
of 2139 kg or with an average telecommunication satellite with a mass of 1000-
5000 kg, a 100 kg satellite is at 10 to 50 times smaller, a significant weight de-
crease. The development of nano, micro and picosatellites requires decreased in-
frastructures and cost, making them ideal candidates for academic institutions and
research centres that have a limited budget for space activities, and for novel tech-
nology demonstrations which would be otherwise difficult and costly to put in or-
bit. The importance of small satellites, especially in the weight range of 1 to 30 kg,
is also recognised by the United Nations, which led to the establishment in 2009
of the Basic Space Technology Initiative, a new area of activity of the United Na-
tions Programme on Space Applications [71].
The majority of the miniaturization efforts have been concentrated in case (a),
where MEMS and microelectronics are mostly used. In general the implementa-
tion of MNT is still sporadic and mostly occurs at small satellites. A review of
MEMS used in pico to microsatellites can be found at [51]. A successful example
of the miniaturization of satellites is the CubeSat standard [72, 73]. A CubeSat is a
miniaturized satellite that weighs no more than 1.33 kg, therefore it is classified as
a nanosatellite; its dimensions are 10x10x10 cm and usually uses commercial off-
the-shelf (COTS) electronics components. CubeSats are scalable in 1U incre-
ments; a 2U CubeSat is 201010 cm and a 3U CubeSat is 301010 cm. The
CubeSat standard was developed in 1999 by the California Polytechnic State Uni-
versity (Cal Poly) and Stanford University, with the aim of providing a standard
design for picosatellites, while reducing cost and development time, increasing ac-
cessibility to space and sustaining frequent launches. CubeSats are used for educa-
tional purposes and technology demonstrations, such as testing microtechnologies
in space, earth remote sensing, tethers and biological experiments.
CubeSats are launched and deployed using a common deployment system, the
Poly-PicoSatellite Orbital Deployer (P-POD). The P-PODs are mounted on the
launch vehicle and carry a maximum of three CubeSats into orbit. Additionally,
CubeSats are usually piggy-back launches; the launch vehicle is used for anoth-
er purpose (e.g. for a commercial, full-size satellite) and the CubeSats are put into
orbit once the main spacecraft has been deployed. The minimized mass and vol-
ume, in conjunction to the use of COTS components and the piggy-back launch,
results in a significantly less expensive system that is developed much faster com-
pared to bigger satellites. As a guideline, a CubeSat costs approximately $40K, in-
cluding launch costs and has an average development time of 1-2 years, whereas a
500 kg satellite requires 3 years and bigger ones require 5 and more years. How-
ever, CubeSats have a smaller lifetime compared to that of bigger satellites
(weight > 500 kg), since they use COTS that have not been fully tested and quali-
fied for the harsh space environment (vacuum, radiations, extreme thermal condi-
tions); for example the typical lifetime of a telecommunications satellite is 15
years, whereas CubeSats have a lifetime of months to 3 years. It should be noted,
that the benefits of CubeSats are also applicable to micro and nanosatellites.
From 2003 to 2009, more than 45 CubeSat missions have been successfully
launched, such as ESAS SSETI Express [74] and the SwissCube. The EPFL de-
veloped and successfully launched in 2009 the SwissCube, with the aim of taking
pictures of the atmospheric airglow using a small low cost earth sensor [75]. The
earth sensor weighted less than 50 gr, the optics volume was 30x30x65 mm
3
, and
the payload board was 80x35x15 mm
3
. To acquire quality airglow images, the re-
quired attitude determination accuracy was better than 1.

There are several small satellite missions that employ microtechnologies.
MEMS are mostly used in the CubeSat attitude determination subsystem, such as
inertial sensors (gyroscopes), magnetometers, and optical sensors (sun sensors,
star trackers). Magnetometers and gyroscopes are typically COTS devices, while
sun sensors and star trackers are space specific. The SwissCube-1 mission used
the MEMS gyroscope ADXRS614 manufactured by Analog Device and the 3 axis
Honeywell magnetometer HMC1053. Another example is the AAUSAT-II Cu-
besat developed at Aalborg University, Denmark (2008), which used 6 Analog
Devices single chip yaw rate gyroscopes (model ADXRS401), and one 3 axis
magnetometer (model HMC1053) from Honeywell as part of the attitude determi-
nation system [51]. In both cases, the sensors operated satisfactorily in orbit.
MEMS based propulsion that produces small thrusts in the order of to 1N (mi-
cropropulsion) has also been used in pico to nanosatellites.
The PRISMA mission (2010) used a MEMS micropropulsion system. The mis-
sion consisted of two satellites, Mango (140 kg) and Tango (40 kg), with the aim
of demonstrating autonomous satellite formation flying. Mango was equipped
with a hydrazine propulsion system, a high performance green propellant (HPGP)
system and a MEMS cold gas micropropulsion system manufactured by Nano-
space. The system consisted of a four thruster array, orthogonally distributed in
the equator plane of the golf ball sized thruster module, with a thrust range of 10
N to 1 mN and used nitrogen as propellant. The micropropulsion system was
used successfully during the mission and is candidate for future missions where
extremely low and accurate thrust is required [76]. The Delft University of Tech-
nology in the Netherlands launched Delfi-C3, a 3 unit CubeSat in 2008 and its fol-
low up will be Delfi-n3Xt, which test several innovative technologies, including a
micropropulsion system and micro sun sensors [77]. The University of Torontos
Institute of Aerospace Studies Space Flight Laboratory has successfully launched
a number of nanosatellites, such as MOST (2003) [78]. MOST incorporated a
small optical telescope (15 cm aperture) equipped with a CCD photometer de-
signed to return high photometric precision and frequency on stars other than the
Sun and successfully demonstrated the capabilities of a significantly smaller tele-
scope. NASAs Space Technology 5 mission (2006) demonstrated the operation of
three 25kg, fully functional spacecrafts that functioned as a single constellation
and implemented multiple new technologies and miniaturized components [79].
The miniaturized technologies that were successfully validated include the follow-
ing: cold gas microthrusters (CGMT), designed by Marotta Scientific Controls,
Inc., variable emittance coatings for thermal control, which consisted of an electri-
cally tuneable coating that could change properties (absorbing heat when cool to
reflecting or emitting heat when in the Sun) [80] and CULPRiT, a microelectronic
device that allows circuits to operate at 0.5 Volts, a technology that is expected to
reduce power consumption while achieving a high radiation and latch-up immuni-
ty. Recent examples of successful nanosatellite missions are NASAs NanoSail-D
(2010), Fig. 5.4a, and the O/OREOS (2010), Fig. 5.4b, [81, 82]. NanoSail-Ds ob-
jective was the experimental validation of solar sail capabilities, with the sail
packed in-side the 9.9x9.9x37.9 cm
3
satellite. NanoSail-D successfully deployed
the 100 square feet polymer sail in January 2011. O/OREOS is the first nanosatel-
lite to operate in the exosphere, conducting autonomous biological and chemical
measurements, weighs 5.5 kg and will use a propellant-less mechanism for de-
orbiting.


(a) (b)
Fig.5.4 (a) Nanosail D stowed and ready for deployment test and (b) test of an early pro-
totype of O/OREOS bus (Courtesy of NASA).
5.5 Future Trends
As discussed above, the design of robotic devices for space applications is in gen-
eral affected by the ongoing miniaturization efforts in two ways. Those systems
that must be of certain size become more capable, while the rest shrink in size.
However, the size of all space systems remains much larger from that of the ter-
restrial systems, due to the different scale of the actual space tasks, the conserva-
tism of the space industry [76] (it may lag ten years with respect to the same tech-
nologies on earth), the need for extreme reliability and the inability for in-situ
repairs, and the requirement for survivability in extreme space conditions (radia-
tion, atomic oxygen, extreme atmospheres).
It is expected though that a number of drastically smaller devices will be con-
sidered and employed in space applications. For these to be adopted, the operating
scenarios will have to exploit the capabilities of micro and nanotechnologies in in-
novative ways. For example, an alternative to some functions provided by lander-
deployed rovers can be the deployment of a large number of microrobotic planet
monitoring modules with low flying balloons. These can cover great areas, estab-
lish a redundant communications network, transmit temperature, seismic or other
data, and even change their position periodically using spring loaded mechanisms,
with all their functions powered by harvesting solar energy. A similar scenario in-
cludes a parachute-deployed network of interconnected channels, covering vast
planetary areas and containing sensing bionano robots [83]. Micro or nanorobots

can also be envisioned to act as rover or mother station linked disposable scouts,
exploring hazardous zones in one-shot missions [44]. Another possibility is us-
ing swarm nanorobots for inspection and repair of space structures on orbit or on
planets, or for checking the status of spacecraft thermal shielding before reentry.
Researchers also envision applications such as spacesuit repairs by suit embedded
nanorobots [84].
The driving forces that will further strengthen the trend towards miniaturization
of robotic devices for space applications are many. The demand for small volume
and mass of space systems will continue to be important, due to launch volume
and mass technology constraints. As mentioned earlier, the cost of launching a
kilogram exceeds $10K, reaching even twice as more. Clearly, adding mass in-
creases the cost. Miniaturization will continue occurring also due to component
size reduction and to higher level of integration between platforms providing mo-
bility (such as spacecrafts or rovers) and instruments or sensors [85]. Downscaling
of systems and components has some interesting properties. For example, the iner-
tia forces on a component are proportional to the cube of its characteristic length
(size), while its stiffness is inversely proportional to it. Therefore, a decrease in
size reduces the inertial forces and increases its rigidity, with an obvious benefit to
its overall robustness to shocks [44]. Also, since the ratio of area to volume is in-
versely proportional to length, smaller systems can have higher power densities
and can dissipate power more effectively [86]. It is also important to note that a
large number of inexpensive robotic devices are obviously more effective against
failures versus a single large and expensive one, as the failure of a few of the min-
iature devices will not jeopardize the entire mission.
Miniaturization will have to overcome formidable technological obstacles. For
example, it is very difficult to have high voltages required in electron spectrome-
try in a very confined space [85], to produce radiation hardened chips with the
same capabilities as those for terrestrial applications, or to cover great distances at
reasonable times with millimeter size rovers. Despite factors that hinder the prolif-
eration of miniaturized robotic devices in space, the trend is clear and will contin-
ue for many years to come.
Acknowledgments
Part of the work presented in this chapter was conducted under the ESA Project
Identification and Assessment of Existing Terrestrial Micro-systems and Micro-
technologies for Space Robotics, 22110/08/NL/RA. Mr. Paraskevas has been fi-
nanced by the European Union (European Social Fund ESF) and Greek national
funds through the Operational Program Education and Lifelong Learning of the
National Strategic Reference Framework (NSRF) - Research Funding Program:
Heracleitus II. Investing in knowledge society through the European Social Fund.
References
[1] http://sm.mdacorporation.com/what_we_do/robotics.html. Accessed May 2011.
[2] http://www.jpl.nasa.gov/missions/missiondetails.cfm?mission=Pathfinder. Accessed May
2011.
[3] Martel S, Felfoul O, Mathieu J-B, Chanu A, Tamaz S, Mohammadi M, Mankiewicz M, and
Tabatabaei N (2009) MRI-based nanorobotic platform for the control of magnetic nanopar-
ticles and flagellated bacteria for target interventions in human capillaries. International
Journal of Robotics Research (IJRR), Special Issue on Medical Robotics, vol 28, no 9, pp
1169-1182.
[4] Vartholomeos P, Fruchard M, Ferreira A, Mavroidis C (2011) MRI-guided nanorobotic
systems for therapeutic and diagnostic applications. Annual Review of Biomedical Engi-
neering, vol 13, pp 157184.
[5] Seeni A, Schfer B, Hirzinger G (2010) Robot Mobility Systems for Planetary Surface Ex-
ploration State-of-the-Art and Future Outlook: A Literature Survey. In Thawar TA (ed)
Aerospace Technologies Advancements. InTech.
[6] http://www.jpl.nasa.gov/jplhistory/mission/apollo-t.php. Accessed May 2011.
[7] http://nssdc.gsfc.nasa.gov/planetary/viking.html. Accessed May 2011.
[8] Muirhead BK (2004) Mars rovers, past and future. Proc IEEE Aerospace Conf, Big Sky,
Montana, USA. doi: 10.1109/AERO.2004.1367598.
[9] http://marsrovers.jpl.nasa.gov/home/index.html. Accessed May 2011.
[10] http://marsprogram.jpl.nasa.gov/msl/. Accessed May 2011.
[11] Baglioni P, Fisackerly R, Gardini B, Gianfiglio G, Pradier AL, Santovincenzo A, Vago JL,
van Winnendael M (2006) The Mars exploration plans of ESA. IEEE Robotics & Automa-
tion Magazine, vol 13, pp 83-89.
[12] http://www.beagle2.com/index.htm. Accessed May 2011.
[13] http://www.jpl.nasa.gov/news/phoenix/main.php. Accessed May 2011.
[14] Ji LY, Bartsch S, Kirchner F (2011) Distributed Compliance Controller for Legged-Robot
with Geared Brushless DC Joints. 11th Symposium on Advanced Space Technologies in
Robotics and Automation (ASTRA), ESTEC, Noordwijk, The Netherlands.
[15] Silva MF, Machado JAT (2010) A Survey of Technologies and Applications for Climbing
Robots Locomotion and Adhesion. In: Miripour D (ed) Climbing and Walking Robots.
InTech.
[16] Curtis SA, Brandt M, Bowers G et al (2006) Mobile Science Platforms for Impassable Ter-
rain. Proceedings IEEE Aerospace Conference, Big Sky, Montana, USA. doi: 10.1109/
AERO.2006.1655722.
[17] Truszkowski W, Hallock H, Rouff C, Karlin J, Rash J, Hinchey M, Sterritt R (2009)
Swarms in Space Missions. Autonomous and Autonomic Systems: With Applications to
NASA Intelligent Spacecraft Operations and Exploration Systems, 1st Edition. Springer,
London.
[18] Kerzhanovich VV, Hall JL, Faibrother D, Said M (2007) Progress in Planetary Aerobot
Technologies. Proc IEEE Aerospace Conf, Big Sky, Montana, USA. doi:
10.1109/AERO.2007.352742.
[19] http://mars.jpl.nasa.gov/odyssey/. Accessed May 2011.
[20] http://saturn.jpl.nasa.gov/. Accessed May 2011.
[21] http://www.nasa.gov/mission_pages/epoxi/index.html. Accessed May 2011.
[22] Ulamec S, Biele J (2010) From the Rosetta Lander Philae to an Asteroid Hopper: Lander
Concepts for Small Bodies Missions. 7th Int Planetary Probe Workshop (IPPW), Barcelo-
na, Spain.
[23] http://near.jhuapl.edu/. Accessed May 2011.
[24] Kawaguchi J, Uesugi K, Fujiwara A (2008) Hayabusa - Its technology and science accom-
plishment summary and Hayabusa-2. Acta Astronautica, vol 62, issue 10-11, pp 639-647.

[25] http://www.esa.int/esaHS/ESAQEI0VMOC_iss_0.html. Accessed May 2011.
[26] Sato N, Wakabayashi Y (2001) JEMRMS Design Features and Topics from Testing. 6th Int
Symposium on Artificial Intelligence, Robotics and Automation in Space (iSAIRAS), Que-
bec, Canada.
[27] Oda M (2001) ETS-VII: Achievements, Troubles and Future. 6th Int Symposium on Artifi-
cial Intelligence, Robotics and Automation in Space (iSAIRAS), Quebec, Canada.
[28] Ogilvie A, Allport J, Hannah M, Lymer J (2008) Autonomous Satellite Servicing Using the
Orbital Express Demonstration Manipulator System. 9th Int Symposium on Artificial Intel-
ligence, Robotics and Automation in Space (i-SAIRAS), Los Angeles, USA
[29] http://www.dlr.de/rm/en/desktopdefault.aspx/tabid-3825/5963_read-8759/. Accessed May
2011.
[30] http://www.esa.int/SPECIALS/ATV/index.html. Accessed May 2011.
[31] http://www.jaxa.jp/projects/rockets/htv/index_e.html. Accessed May 2011.
[32] Hirzinger G, Brunner B, Dietrich J, Heindl J (1993) Sensor-based space robotics-ROTEX
and its telerobotic features. IEEE Transactions on Robotics and Automation, vol 9, no 5, pp
649-663. doi: 10.1109/70.258056.
[33] Rebele B, Schaefer B, Albu-Schaeffer A, Bertleff W, Landzettel K (2006) Robotic Joints
and Contact Dynamics Experiments: Lessons Learned from ROKVISS. 9th ESA Workshop
on Advanced Space Technologies for Robotics and Automation (ASTRA), ESTEC,
Noordwijk, The Netherlands.
[34] http://robonaut.jsc.nasa.gov/default.asp. Accessed May 2011.
[35] Choset H, Knepper R, Flasher J, Walker S, Alford A, Jackson D, Kortenkamp D, Burridge
R, Fernandez J (1999) Path planning and control for AERCam, a free-flying inspection ro-
bot in space. Proc IEEE Robotics and Automation Int Conf, Detroit, MI, USA. doi:
10.1109/ROBOT.1999.772556.
[36] Nolet S, Saenz-Otero A, Miller DW, Fejzic A (2007) SPHERES Operations aboard the ISS:
Maturation of GN&C Algorithms in Microgravity. AAS Guidance, Navigation & Control,
Breckeridge, CO, USA.
[37] Summerer L, Putz B, Kopacek P, Kaya N (2006) Robots Moving on a Loose Net in Micro-
gravity - Results from the Japanese Furoshiki Sounding Rocket Experiment. 9th ESA
Workshop on Advanced Space Technologies for Robotics and Automation (ASTRA),
ESTEC, Noordwijk, The Netherlands.
[38] http://www.wtec.org/loyola/mcc/mems_eu/Pages/Chapter-3.html. Accessed May 2011.
[39] Yole Dveloppement (2010) Status of the MEMS Industry. http://www.yole.fr. Accessed
May 2011.
[40] Yole Dveloppement (2009) MEMS Accelerometer, Gyroscope and IMY Market, 2008
2013. http://www.yole.fr. Accessed May 2011.
[41] Miller LM (1999) MEMS for space applications. Proc SPIE 3680, 2.
doi:10.1117/12.341193.
[42] George T, Bae Y, Chakraborty I, Cherry H et al (2000) MEMS technology at NASA's Jet
Propulsion Laboratory. Proc SPIE 4134, 16. doi:10.1117/12.405340.
[43] Janson S, Helvajian H, Amimoto S, Smit G, Mayer D, Feuerstein S (1998) Micro-
technology for space systems. IEEE Proc, vol 1, pp 409-418. doi:
10.1109/AERO.1998.686938.
[44] Corradi P, Menciassi A, Dario P (2005) Space Applications of Micro-Robotics: A Prelimi-
nary Investigation of Technological Challenges and Scenarios. Proc 5th Round Table on
Micro/ Nano Technologies for Space, Noordwijk, The Netherlands.
[45] Atchison J (2011) Length Scaling in Spacecraft Dynamics. J of Guidance, Control, and Dy-
namics, vo.34, no 1, pp 231-246. doi: 10.2514/1.49383.
[46] http://www.congrex.nl/10C03/. Accessed May 2011.
[47] Paraskevas I, Flessa T, Papadopoulos E (2011) Spinning-In of Terrestrial Micro Systems
and Technologies to Space Robotics: Results and Technology Roadmaps. 11th Symposium
on Advanced Space Technologies in Robotics and Automation, ESTEC, Noordwijk, The
Netherlands.
[48] Caprari G, Siegwart R (2005) Mobile micro-robots ready to use: Alice. Proc of IEEE/RSJ
International Conference on Intelligent Robots and Systems, pp 3295- 3300. doi:
10.1109/IROS.2005.1545568.
[49] Schiele A, Romstedt J, Lee C, Henkel H, Klinkner S et al (2008) NanoKhod Exploration
Rover - A Rugged Rover Suited for Small, Low-Cost, Planetary Lander Mission. Robotics
& Automation Magazine, IEEE, vol 15, no 2, pp 96-107. doi: 10.1109/MRA.2008.917888.
[50] Menon C, Li Y, Sameoto D, Martens C (2008) Abigaille-I: Towards the development of a
spider-inspired climbing robot for space use. 2nd IEEE RAS & EMBS Int. Conf. on Bio-
medical Robotics & Biomechatronics, pp. 384-389. doi: 10.1109/BIOROB.2008.4762903.
[51] Shea HR (2009) MEMS for pico- to micro-satellites. Proc of SPIE 7208.
doi:10.1117/12.810997.
[52] Osiander R, Darrin M.A.G, Champion J.L. (2006) MEMS and Microstructures in Aero-
space Applications. CRC Press.
[53] Khler J, Bejhed J, Kratz H et al (2002) A hybrid cold gas microthruster system for space-
craft. Sensors and Actuators A: Physical, vol 97-98, pp. 587-598. doi: 10.1016/S0924-
4247(01)00805-6
[54] Mitterauer J (2002) Micropropulsion for small spacecraft: a new challenge for field effect
electric propulsion and microstructured liquid metal ion sources. Surface and Interface
Analysis, 36, pp 380386. doi: 10.1002/sia.1693.
[55] van der List MCAM, van Put PAG, Yce V, Kuiper J (2006) Next Generation Electrical
Propulsion Feed Systems and Spin-Off Micro-Propulsion Components. ISU 2006 Confer-
ence, Strasbourg, France.
[56] Rossi C, Briandb D., Dumonteuila M, Campsa T, Phamb PQ, de Rooijb NF (2006) Matrix
of 1010 addressed solid propellant microthrusters: Review of the technologies. Sensors &
Actuators A: Physical, vol 126, no 1, pp 241-252. doi:10.1016/j.sna.2005.08.024.
[57] Krpoun R, Rdber M, Shea HR (2008) Microfabrication and test of an integrated colloid
thruster. Proc 21st IEEE Int Conf on Micro Electro Mechanical Systems, pp 964967, Tuc-
son, AZ. doi: 10.1109/MEMSYS.2008.4443818.
[58] Snyder JS, Randolph TM, Hofer RR, Goebel DM (2009) Simplified Ion Thruster Xenon
Feed System for NASA Science Missions. 31st Int Electric Propulsion Conf, Ann Arbor,
Michigan, USA.
[59] Kim SS, Hayati S, Lavery D, McBride KS (2006) Mars miniature science instruments. Proc
IEEE Aerospace Conf, Big Sky, Montana, USA. doi: 10.1109/AERO.2006.1655786.
[60] Kutyrev AS, Arendt R, Moseley SH, Boucarut RA et al (2004) Programmable microshutter
arrays for the JWST NIRSpec: Optical Performance. IEEE J of Selected Topics in Quantum
Electronics, vol 10, no 3, pp 652- 661. doi: 10.1109/JSTQE.2004.830614.
[61] Buchner S, Rapchun DA, Moseley H, Meyer SE et al (2007) Response of a MEMS Mi-
croshutter operating at 60 K to ionizing radiation. IEEE Transactions on Nuclear Science,
vol 54, no 6, pp 2463 - 2467. doi: 10.1109/TNS.2007.910040.
[62] Sellers JJ (2005) Understanding Space An Introduction to Astronautics, 3rd Edition.
McGraw-Hill.
[63] Sundaram S, Tormen M, Timotijevic B, Lockhart R et al (2011) Vibration and shock relia-
bility of MEMS: modeling and experimental validation. J of Micromechanics and Microen-
gineering, vol 21, no 4. doi: 10.1088/0960-1317/21/4/045022.
[64] Shea H (2011) Effects of Radiation on MEMS. Proc of SPIE 7928. doi: 10.1117/12.876968
[65] www.ecss.nl. Accessed May 2011.
[66] Shea H (2006) Reliability of MEMS for space applications. Proc SPIE 6111.
doi:10.1117/12.651008.
[67] Hartzell AL, da Silva MG, Shea HR (2011) MEMS Reliability, 1st edition. Springer

[68] UCS Satellite Database (1 February 2011).
http://www.ucsusa.org/nuclear_weapons_and_global_security/space_weapons/technical_iss
ues/ucs-satellite-database.html.Accessed May 2011.
[69] Euroconsult (2010) Satellites to be Built & Launched by 2019, World Market Survey, 13th
edition. http://www.euroconsult-ec.com/research-reports/space-industry-reports/satellites-
to-be-built-launched-by-2019-38-29.html. Accessed May 2011.
[70] ADS Eurospace (2010) European Space Industry activity - Facts and Figures, 14 Edition.
http://www.eurospace.org/. Accessed May 2011.
[71] UN General Assembly Committee on the Peaceful Uses of Outer Space (2010) Report on
the second United Nations/Austria/European Space Agency Symposium on Small Satellite
Programmes for Sustainable Development: Payloads for Small Satellite Programmes.
[72] http://cubesat.org/. Accessed May 2011.
[73] Toorian A, Diaz K, Lee S (2008) The CubeSat Approach to Space Access. IEEE Aerospace
Conf, Big Sky, Montana, USA. doi: 10.1109/AERO.2008.4526293.
[74] http://www.esa.int/SPECIALS/sseti_express/index.html. Accessed May 2011.
[75] http://swisscube.epfl.ch/. Accessed May 2011.
[76] Gronland T, Rangsten P, Nese M, Lang M (2007) Miniaturization of components and sys-
tems for space using MEMS-technology. Acta Astronautica, vol 61, issues1-6, pp 228-233.
doi:10.1016/j.actaastro.2007.01.029.
[77] Bouwmeester J, Brouwer GF, Gill EKA, Monna GLE, Rotteveel J (2010) Design Status of
the Delfi-Next Nanosatellite Project. Proceedings of the 61th Int. Astronautical Congress,
Prague, Czech Republic.
[78] http://www.utias-sfl.net/microsatellites/MOST/MOSTIndex.html. Accessed May 2011.
[79] http://nmp.nasa.gov/st5/ABOUT/about-index.html. Accessed May 2011.
[80] Beasley MA, Firebaugh SL, Edwards RL, Keeney AC, Osiander R (2004) MEMS thermal
switch for spacecraft thermal control. Proc SPIE 5344, 98. doi:10.1117/12.530906.
[81] http://www.nasa.gov/mission_pages/smallsats/nanosaild.html. Accessed May 2011.
[82] http://www.nasa.gov/mission_pages/smallsats/ooreos/main/. Accessed May 2011.
[83] Mavroidis C, Ummat A (2005) Space Bionanorobotic Systems: Design and Applications.
Proc of the 7th NASA/DoD Conf on Evolvable Hardware (EH-2005), Washington DC.
[84] Chui B, Kissner L (2000) Nanorobots for Mars EVA Repair. Int Conf On Environmental
Systems, Toulouse, France. doi: 10.4271/2000-01-2478.
[85] Dorsky LI (2001) Trends in instrument systems for deep space exploration. IEEE Aero-
space and Electronic Systems Magazine, vol 16, no 12, pp 3-12. doi: 10.1109/62.974833.
[86] Menon C, Ayre M, Ellery A (2006) Biomimetics - A New Approach for Space System De-
sign. ESA Bulletin 125, vol 5, pp 20-26.