PROJECT DAEDALUS, ROTOR CONTROLLED DESCENT AND LANDING ON REXUS23

Full text

PROJECT DAEDALUS, ROTOR CONTROLLED DESCENT AND LANDING ON REXUS23
Clemens Riegler1, Ivaylo Angelov1, Florian Kohmann1, Tobias Neumann1, Abdurrahman Bilican1, Kai
Hofmann1, Jessica Gutierrez Pielucha1, Alexander B¨
ohm1, Barbara Fischbach1, Tim Appelt1, Lisa Willand1,
Oliver Wizemann1, Sarah Menninger1, Jan von Pichowski1, Jonas Staus2, Erik Hemmelmann2, Sebastian Seisl3,
Christoph Fr¨
ohlich3, Christian Plausonig3, Alexander Hartl3, Patrick Kappl3, and Reinhard Rath3
1JMU W¨
urzburg - Space Technology, W¨
urzburg, Germany
2University of Applied Sciences W¨
urzburg-Schweinfurt, W¨
urzburg, Germany
3TU Wien SpaceTeam, Vienna, Austria
ABSTRACT
Daedalus goal is it to build and test an alternative form
of descent mechanism for drops from very high altitudes.
Since parachutes are not always the best option and some-
times hard to handle in space applications, there is a need
for new and innovative ways to descend through an at-
mosphere without using any propellant at all. Inspired by
the bionic design of the maple seed which gently falls to
the ground using its ingenious natural design, Daedalus is
building a SpaceSeed applying the same idea of a gentle
descent for Re-entry missions. Such a mechanism can be
more robust towards supersonic re-entry than a parachute
and can also be less affected by the space environment or
heat compared to traditional options. This can be espe-
cially useful when looking towards the challenges a fu-
ture Mars mission would bring to the table, but applica-
tions of this are numerous, for example, could designs
like this also have terrestrial usages like weather probes
which would be able to withstand harsher winds than a
parachute could. This is directly usable for a turbulent at-
mosphere like on Venus or even Jupiter. Besides, payload
returns from LEO could be made with similar designs
thereby removing some of the challenges one would face
with a parachute or an actively propelled landing system.
Daedalus wants to prove that this alternative is usable for
real-life applications on earth and other planets.
Key words: Re-entry, Landing, Rotor, Rotochute.
1. INTRODUCTION
The REXUS Project Daedalus developed a mechanism
that can achieve soft and target landing without utiliz-
ing any active propellant or engines of any kind. Hence
this can be a superior option to propulsive landing or
landing utilizing parachutes. SpaceSeed is the name of
the vehicle that served as a technology demonstrator of
the first iteration of this landing technology. To develop
the technology fully, more than the 2-year cycle of the
REXUS Program is needed, hence the team only devel-
oped a demonstrator to validate the very basic principle of
this technology and to gather data to optimize future ver-
sions of the technology. The technologic principle upon
this lander is based is auto-rotation. This is a very well
known flight characteristic of helicopters, used by these
vehicles when the engine fails and a safe landing must
be performed. It is clear that such a technology can be
used for many re-entry scenarios. One example would
be a sample return from an asteroid to land payload on a
ship where it can immediately be accessed by scientists as
these samples are desired to be immediately stored to pre-
serve them ideally. Another option would be to land dif-
ferent stages of rockets with rotors to offer a downrange
landing technology which is not reliant to propellant but
only to electric power for rotor actuators. For Daedalus,
many components had to be created and these shall be
addressed in the following paper, as well as the most im-
portant data along with an evaluation of said data. The
software development and communication strategies of
the SpaceSeeds are described in section 2. The electron-
ics and the infrastructure needed for the software are de-
scribed in section 3. The mechanical setup, including the
deployment mechanism mounted on the REXUS rocket,
is shown in 4. The flight data and the scientific return
of the recorded data points are discussed in 5. The final
conclusions of the project and the future goals are sum-
marized in 6.
2. SOFTWARE AND TELEMETRY
The telemetry, depicted in figure 1, has two tasks in the
Daedalus Project. The main task is to recover the Space-
Seeds. To achieve this, it is necessary to send GNSS
data to the search team, which is looking for the landed
SpaceSeeds. To achieve this, the Iridium satellite net-
work was used to send GPS data to the ground station
which was specifically developed for this mission. One
part is a server accepting the data from the Iridium gate-
way, which collects the data received from the satellites.
On the other hand, the ground station also collects the

received data from the board computer mounted on the
rocket core to monitor it. As a fallback, if no Space-
Seed has been found, at least some data would have been
sent via X-Bee for further analysis. The data set collected
this way is not a complete redundancy, since the X-Bee
connection only ranges up to 4km, whereas the Space-
Seeds and the rocket core will have a greater distance at
some point. Additionally to this, we could also receive
GPS Position via X-Bee using a search device. However,
this case would be only possible if the search team flies
near the space seed during the recovery of the main rocket
core. Luckily, valid position data for all SpaceSeeds has
been received via Iridium, which was the primary solu-
tion and in the end, were physically recovered.
2.1. Data Management
Data management is mandatory for the project. In gen-
eral, all data is saved locally on the FMS internal flash in
high frequency, in lower frequency the FMS sends its data
to the ORBC where this data is redundantly stored on the
SD storage card and parts are forwarded the ground sta-
tion for a live overview. The Data flow can be separated
into 3 major phases:
Pre-Ejection , during countdown and until the ejection
the Seeds have a wired bidirectional UART connection to
the ORBC with a baud rate of 115200 Bd. Until liftoff the
ORBC also has a wired bidirectional connection via the
RXSM to the ground station, after liftoff this connection
is unidirectional from the ORBC via the RXSMs Radio
to the ground station, so no commanding will be possible
after liftoff. In this phase all data is saved locally, and
parts are forwarded to the ground station
Post-Ejection , after the ejection the connection be-
tween the ORBC and the Seeds FMS is wireless by us-
ing the XBees, this connection is held until there are
out of range, while the range is maximized by using the
DigiMesh Network technology and a short delay between
the ejection of the seeds.
Landing/Ground , after landing, the SpaceSeed will
send his GPS location using the Iridium Satellite Net-
work to the server as shown in figure 1. As a backup in
case of connection problems, a heartbeat is implemented
using the XBees to send the location to a search device
equipped with the same XBee.
3. ELECTRONICS AND INFRASTRUCTURE
The whole experiment consists of several independent
subsystems, so a proper power and data control are
mandatory. Therefore, a lot of work has been done to
achieve a working experiment. The major systems are the
three free falling units (FFU); they must be completely
independent including with their own power and data
management systems. As well as the On Rocket Board
Computer (ORBC) underneath the ejection mechanism,
controlling the GoPro and managing telecommand and
telemetry data of the SpaceSeeds both using wired and
wireless connections.
3.1. SpaceSeed
The electronics of the SpaceSeeds have several purposes.
The Trajectory of the whole flight and other data needs to
be monitored to validate the underlying re-entry and land-
ing concept. This is done by an IMU accompanied by a
barometer, a temperature sensor and a GPS device. The
processor and the sensors are located on a single PCB
called Flight Management System FMS. This is a com-
plete flight controller designed for small sounding rock-
ets developed by the TU Wien SpaceTeam. To adapt the
FMS from a sounding rocket to our SpaceSeeds, some
changes need to be done and two additional boards had
to be designed. First, a Power-Supply-Board that houses
the charging circuit for the NiMh Batteries, as LiIon were
not suitable for our purpose, together with a heating sys-
tem to preheat the batteries before the start had to be de-
signed. Secondly, the communication has to be adapted
in form of another LowRange Communication module, in
this case, the XBee 868LP stacked onto the FMS, as well
as a second board, the Iridium motherboard, holding the
transmission module together with capacitors, due to the
intense power usage during transmission to the Iridium
network. The main task of the SpaceSeed software is the
high-frequency storage of the sensor data. Other tasks are
the management of communication to the ORBC as well
as the Iridium network. The connection to the ORBC is
for redundancy reasons as the data had to be forwarded
to the ORBC-Storage using the LowRange XBee mod-
ules while they are in range. This is done to have a kind
of backup of the data in case of losing a SpaceSeed. To
increase the chance of finding a FFU, the primary solu-
tion is the Iridium module, which is sending the GPS po-
sition via this satellite connection, the secondary solution
is the Heartbeat mode. It is a backup strategy based on
the XBee modules in conjunction with a search device.
3.2. On Rocket Board Comupter (ORBC)
The On Rocket Board Computer is the interface be-
tween the main REXUS Service Module (RXSM) and
the SpaceSeeds. Furthermore, it controls and supplies the
GoPro camera. The GoPro has no interface available for
control, therefore the on/off button is simply bypassed by
a transistor. The transmitted video signal confirms the
proper function before the start. The main task of the
ORBC is to establish communication between the ground
station and the SpaceSeeds and to log all the data. Until
lift-off, the connection from the ORBC and the RXSM is

Figure 1. Telecommunications Strategy in order to maximize recoverability of the SpaceSeeds
Figure 2. Block Diagram for the general electrical setup of the whole experiment
bidirectional to be able to command the ORBC as well
as the SpaceSeeds. After liftoff this connection becomes
unidirectional from the ORBC to the RXSM, so no more
commanding can be done, only telemetry data can be sent
or forwarded to the ground station. The other connection
to be managed is the communication to the FMS inside
the Seeds. This is done by a wired connection until the
ejection and afterwards, wireless using the XBee mod-
ules. The wired connection is a six-core cable holding
power for charging, a UART connection for communica-
tion, as well as a signal line to wake up the FMS. Under
nominal conditions, the wake-up signal is triggered by a
telemetry command of the ORBC, but as a backup, if the
ORBC has any problems this line is also connected to the
lift-off signal of the RXSM.
The ORBC is programmed to be a gateway between the
SpaceSeeds and the ground station, using the available
connection. This means switching from a wired connec-
tion to the wireless connection after the ejection to re-
ceive sensor data. Then storing this data locally as well
as forwarding it to the ground station. While all received
data is stored locally, a smaller amount is forwarded to
the ground station in adaptation to the downlink speed. A
complete overview of all subsystems can be seen in figure
2.
4. MECHANICAL SETUP
There are two main mechanical parts for the complete
experiment setup. The first one is the deployment mech-
anism. It stores and releases the SpaceSeeds at apogee.
The second, but most important part of the mechanical

design was the SpaceSeed itself, especially with its fold-
away wings and easy-to-access electronics, yet robust de-
sign to withstand the flight and landing loads.
4.1. The Deployment Mechanism
The most challenging aspect of the deployer was the
space within the 1:4 14” ogive nose cone. With the tubes
that hold the SpaceSeeds in place having an inner diam-
eter of 130mm, the design constraints were stressed to
their maximum in order to fit the glass fibre tubes into
the nosecone. The SpaceSeeds were held down by a steel
cable which was cut at apogee. A pre-tensioned spring
then pushed the experiments out of their storage posi-
tions and away from the rocket to start the free flight. Not
only the SpaceSeeds needed to be accommodated but also
the ORBC and the GoPro. As the volumetric constraints
were already stressed these components moved below the
ground plate, along with the pyro cutters and steal cable
retainers. The REXUS-Rocket offers a roughly 90mm
high comportment below the adapter plate which was
more than sufficient to fit all necessary elements there. In
figure 3, 4 and 5, the deployer can be seen from different
angles including the SpaceSeeds in their storage position.
Figure 3. Top view of the Deployer, including the separa-
tion mechanism and tip of the nosecone
4.2. The SpaceSeeds
The heart of the experiment were the SpaceSeeds, as seen
in figure 6, named after the idea of maple seeds also using
auto rotation on their descent. The mechanical challenges
of these rotor-descent vehicles were numerous. The fol-
lowing elements had to be taken into account:
•Flight loads, especially in the transonic regime
•Minimum weight
•Volumetric constraints
•Natural aerodynamic stability
•Antenna positioning
•Radio translucent materials at corresponding areas
•Landing shock loads
Figure 4. Bottom view of the Deployer, showing the
ORBC-Box (top right), the GoPro-Box (Bottom), the
through holes for all necessary cables and the 6 pyro-
mounts which for the release mechanism
Figure 5. A side view with the exposed SpaceSeeds in
storage position showing the pre-tensioned springs. The
speration mechanism of the nosecone is also visible
•Accommodation off on-board electronics
The points that proved to be most critical shall be ad-
dressed in the following paragraphs, including figures
of the most important aspects of the SpaceSeed. For
a deeper understanding of the mechanics of the Space-
Seed please refer to the 2nd SSEA Symposium, Paper
ID: SSEA-2018-95, ”The DAEDALUS Project” [Re18].
This paper is focusing more on flight data evaluation
which can be found in section 5.
Volumetric constraints and the foldaway Wings The
rocket only offers limited space and therefore, the biggest
implication on the SpaceSeed were the foldaway wings.
This was achieved with the help of the so-called ”inter-
face ring” as seen in figure 7 and 8. It is a part that
could only be produced by Metal 3D Printing which also
brought weight optimization with it. The idea was to un-
fold the wings and lock them as soon as the SpaceSeeds

Figure 6. This figure depicts a fully constructed Space-
Seed with deployed wings, as it would be seen in flight
conditions
leave their storage tubes in the deployer, hence no special
hold down and release mechanism was needed, simplify-
ing the design.
Flight loads Especially challenging was the fact that
the wings were deployed from the beginning of the flight,
and had to withstand massive forces during re-entry. Vari-
ous simulations lead to the result that Carbon Fiber wings
where the only way of guaranteeing sufficient strength
for the wings during re-entry [Hem18]. Unfortunately,
this could not be tested during flight, as the SpaceSeeds
went into a flat spin and the load profile for the wings
was lower than expected when oriented correctly during
the phase of maximum deceleration.
Natural aerodynamic stability This was only partially
achieved. By lowering the centre of mass as low as pos-
sible toward the nose natural stability should have been
the result. As well as the wings excreting control author-
ity on the whole device to stabilize it further. This was
not observed in the upper atmosphere only in the lower
atmosphere, which is further explained in section 5.
5. FLIGHT DATA EVALUATION
During the mission the general robustness and approach
of the design was tested under realistic conditions. There
were 5 main goals as the result of our mission statement.
1. Construct and build a rotor based descent mecha-
nism
Figure 7. This is the prototype of the interface ring
printed using a thermoplast but already with the mounted
folding mechanism for the wings
2. Construct and build a structure that withstands the
impact
3. Construct and build a deployment mechanism
4. Film the deployment
5. Evaluate the flight data and compare it with simula-
tions
These mission statements can be evaluated as follows:
Construct and build a rotor based descent mechanism
As a result of the experiment, it was shown that the rotor-
based descent mechanism is a feasible alternative to the
typical re-entry options. It was able to reduce the Space-
Seeds velocity to 25-30 m/s. The SpaceSeeds experi-
enced flat-spin in the upper atmosphere since the centre
of gravity was not optimally positioned.
Construct and build a structure that withstands the
impact All three Space Seeds survived the descent
from nearly 80km. The fuselage and electronics were
fully intact, but every SpaceSeed lost some wings. The
data shows that the wings were probably clipped on im-
pact. It is expected to be caused by the collision with the

Figure 8. Here a cut view is shown of the interface ring
mounted in the SpaceSeed, the way the wings move when
deployed can be well observed in this figure
trees on the ground. A stable flight of all SpaceSeeds was
observed towards the end of the flight, suggesting that no
wings where missing, hence, the wings were lost upon
landing.
Construct and build a deployment mechanism All
three SpaceSeeds got deployed from the deployment
mechanism successfully.
Film the deployment Footage of the deployment of the
SpaceSeeds was successfully recorded and sent down. It
was mediately clear that the wings had locked in their
designated position. Unfortunately, due to the hard land-
ing of RX23, the SD-Card was destroyed, and no HD
footage of the deployment is available.
Evaluate the flight data and compare it with simula-
tions During its flights, every SpaceSeed captured data
from the various instruments installed. This will be ex-
plained in further detail in the following subsection.
5.1. Flight Data vs. Simulation
Different simulations of various flight parameters have
been made. These simulations can now be compared
with the recorded sensor values in order to validate and
refine the model in the future. It shall be stated that the
data shown will only be from SpaceSeed number 3. The
recorded data points look very similar for all 3 vehicles
but they key difference is the acquisition of the GPS Sig-
nal, which was mainly used for recovery but also nice to
have in the critical flight phase. Only Vehicle Number 3
got a GPS lock. This is probably due to the high speeds
and the breach of the ITAR-Limits during the flight. Once
on the ground, all SpaceSeeds started sending GPS Data,
thereby the GPS modules fulfilled their primary task.
Figure 9. Angular velocities as predicted by the devel-
oped model
Figure 10. Angular momentum from the actual flight,
deep pass filtered to make the data more readable

As shown in figure 9, which represents a precalculated
simulation, a smooth transition of angular momentum
was expected. The recorded data, represented by figure
10, shows that the flight was less stable than anticipated.
This was since the SpaceSeeds experienced flat-spin in
the beginning and later the wings gained control author-
ity, stabilizing the flight as it was intended. The main con-
clusion is that our model overestimated the control au-
thority and the centre of mass was not low enough given
the lower than expected wing influence in the upper at-
mosphere.
Figure 11. Altitude over time, as predicted by the simula-
tion
Figure 12. Altitude over time, combined with the plot
of predicted data and then the SpaceSeed data, as ITAR
regulations did not allow for GPS Data above a certain
height
As shown in the simulated and actual height in figure 11
and 12 respectively, the time of flight and descent rate
were modelled very accurately. Due to ITAR regulations,
the recording only started at a height of roughly 35km and
the SpaceSeeds couldn’t acquire a GPS Signal above that.
It was not necessary for the experiment, as the critical
flight phase was below 35km and above only, an almost
frictionless free fall was expected due to the low speed of
the SpaceSeeds and the low atmospheric density at these
heights.
Figure 13. Velocity over time as predicted by the model
Figure 14. Velocity over time as measured during flight,
by means of GPS and to offer a rough control, via baro-
metric measurements
In figure 13 and 14 both simulated and real descent ve-
locity can be seen, they matched almost perfectly. The
final speed was roughly 25m/s. This is the result of the
model neglecting body drag and the resulting lower than
expected rotation rate, yet offering equal drag force and

creating a comparable flight envelope. Future simulations
will investigate these effects further to refine the accuracy
of the model. This is especially important for future de-
velopments of a landing control software. These readings
will help to develop such a system in the future.
6. CONCLUSION
It was determined that the SpaceSeed design principle is
a valid alternative for atmospheric descent mechanisms.
There are some design options to improve the problems
that were observed on the first flight. These options can
be implemented in the next iteration of the SpaceSeed
and are good indicators on where improvements are nec-
essary. Most importantly preventing flat-spin, reducing
the descent velocity even further. For GNC it is clear
that an option of avoiding trees needs to be evaluated as
it can not be guaranteed that a potential landing zone is
free of trees. The team has decided to continue work-
ing on the idea of the SpaceSeed as it showed to be very
promising. The next steps will be to improve the rotor
to be able to control the angle of attack of the wings to
ultimately perform a soft landing. Weight improvements,
especially with a focus on the centre of mass, will be per-
formed. The team will be applying for the next cycle
of the REXUS/BEXUS program, namely REXUS 29/30.
The platform offers a perfect way of validating the tech-
nology and helps the student team with professional ad-
vice as future opportunities are still evaluated where this
student project could be escalated to a non-educational
level.
7. ACKNOWLEDGEMENTS
The team wants to thank our advisor, Prof. Hakan Kayal,
Prof. Walter Baur, Dipl. Ing. J¨
urgen Fecher and Prof.
Georg Schitter. They helped our Team and Sub-Teams
where ever they could and believed that this untouched
technology can be developed by students. We also want
to thank EuroLaunch, the organizing committee of the
REXUS Program. The way the program is set up helped
us perfectly to reach our goals, thank you for that. Last,
but definitely not least, we want to thank all our spon-
sors, All-ahead-composites, Materialise, MTReuss, Preh,
FFG, BMViT, ANSYN, uBlox, CADSOFT, GLENPRO,
MOUSER, AST Systems, Digi-Key, blackit.de. Without
your generosity and support, we would have never been
able to achieve what we achieved. Thank you so much.
REFERENCES
[Hem18] Erik Hemmelmann. Fl¨
ugelauslegung eines
raumgleiter, 2018.
[Re18] Clemens Riegler et al. The Daedalus Project.
In L´
aszl´
o Bacs´
ardi, editor, Proceedings of the
2nd Symposium on Space Educational Activi-
ties, volume 1, pages 278–282, 2018.