Minimize PicSat

PicSat (Pictoris Satellite) mission

Spacecraft    Launch    Sensor Complement   References

PicSat is a nanosatellite designed to measure exoplanetary transits. The development of the 3U CubeSat is conducted within the High Angular Resolution Astronomy group at LESIA (Laboratoire d'Etudes Spatiales et d'Instrumentation en Astrophysique), Observatoire de Paris, along with partners: PSL Research University, Paris, CNRS,Sorbonne Universités, UPMC (University Pierre and Marie Curie), Paris. The PicSat satellite uses single-mode fiber filtering and single pixel avalanche photodiodes for high accuracy photometry. It is also a technology demonstrator for future interferometric missions. 1) 2)

The primary objective of this project is to observe the transit of the planet Beta Pictoris b as it passes in front of its star. The planet was first discovered by Anne-Marie Lagrange's team using ESO's VLT (Very Large Telescope) array in Chile. The team discovered a planet orbiting at about 10 AU from the star Beta Pictoris, thanks to the use of VLT's adaptive optics system (Lagrange et al., 2009). Measurements taken between 2003 and 2015 have refined its orbital parameters and this suggests that the planet (or at least its Hill sphere) passes in front of the star.

Moreover, these measurements are consistent with the November 1981 event, where important photometric variations were measured from the ground. If this planet has actually passed that year, the next transit would take place between July 2017 and March 2018 for an eccentric orbit of 0.12.

The ability to observe a transit of this type, a giant young planet, a few million years old, orbiting a bright star is a chance that must be seized. This requires continuous photometric monitoring of the star that only a space observatory can achieve while avoiding the atmospheric disturbances, and the day/night cycle.

The Beta Pictoris system (β Pic) is also known for its debris disk, typical of young star systems (20 million years). Precision photometry also allows us to characterize the dust tails of exo-comets (or comets in another solar system) and measure the structure of the debris disk.

Beta Pictoris is a bright star (A6V star of V magnitude 3.86) which, due to its proximity (63 light years from the Earth) and young age, has always been a most promising target for the study of circumstellar environment and planetary systems. In 2003, β Pictoris b, a young Jupiter like giant exoplanet, was directly imaged around this star. Its orbit was latter characterized using multiple astrometric position measurements acquired since 2009, is compatible with a transiting planet. It is now strongly suspected that its environment will transit during the third or fourth quarters of 2017. Because of its orbital period of 17 years, this represents a rare opportunity to finely characterize a young giant exoplanet and its close environment (Hill sphere) in front of a bright star.

PicSat_AutoB

Figure 1: Beta Pictoris at infrared wavelenghts (image credit: ESO, A.-M. Lagrange) 3)

Legend to Figure 1: This composite image represents the close environment of Beta Pictoris as seen in near infrared light. This very faint environment is revealed after a very careful subtraction of the much brighter stellar halo. The outer part of the image shows the reflected light on the dust disc, as observed in 1996 with the ADONIS instrument on ESO's 3.6 m telescope; the inner part is the innermost part of the system, as seen at 3.6 µm with NACO on the Very Large Telescope. The newly detected source is more than 1000 times fainter than Beta Pictoris, aligned with the disc, at a projected distance of 8 times the Earth-Sun distance. Both parts of the image were obtained on ESO telescopes equipped with adaptive optics.

 


 

Spacecraft:

PicSat is a nanosatellite designed to measure exoplanetary transits. The development of the 3U CubeSat is conducted within the High Angular Resolution Astronomy group at LESIA. The PicSat satellite uses interferometric instrumentation, integrated optics, and single-mode fiber filtering for the study of stellar environments.

PicSat_AutoA

Figure 2: The PicSat satellite CAD design shows the electronics unit comprising of the antennas, communication system, navigation computer, power electronics and batteries, a central unit for attitude control system (ADCS) including the inertia wheels and the scientific payload. The final unit houses the optomechanical payload and the stellar sensor (image credit: PicSat collaboration) 4)

The platform, including the bus and the ADCS, has been subcontracted as our philosophy was to focus on the design and the realization of the payload. The first step was to select the ADCS as the pointing performance is a key point of our mission. Then the 3U bus has been designed given the constraints of the payload and the ADCS.

ADCS (Attitude Determination and Control Subsystem): The ADCS selected for the PicSat mission is the iADCS100 provided by Hyperion Technologies of Delft, The Netherlands. The role of the ADCS will be to point the instrument to Beta Pictoris within an accuracy and stability of 30 arcsec.

PicSat_Auto9

Figure 3: Photo of the ADCS (image credit: Hyperion Thechnologies)

The iADCS100 matches the ST200 star tracker with Hyperion's RW200-series of reaction wheels, as well as the MTQ200 series of magnetorquers. Combined with Berlin Space Technologies' flight-proven control algorithms, it offers an entirely autonomous attitude control system, in the space of two standard CubeSat PCBs (Printed Circuit Boards). With the help of the RW200-series of reaction wheels, it is capable of precisely pointing and slewing a 3U CubeSat.

It features a host of operating modes, chief among which is the tracking mode, which allows users to enter target quaternion, after which the system will orient a pre-defined instrument-side towards that target, following it until it has passed the local horizon.

The ST200 star tracker is deported on the payload stack to have the same LOS (line of sight) with the telescope and a baffle was specially designed for the mission.

Bus: The bus, including the 3U structure, the communication, the command and data handling, and the power management, has been design by ISIS (Innovative Solutions In Space) BV, Delft, The Netherlands. The structure is a standard 3U structure that respects the CubeSat form factor. The structure has been slightly adapted to accommodate the payload.

PicSat_Auto8

Figure 4: Detailed view of the nanosatellite (image credit: ISIS)

RF communications: The communication is performed with the ISIS TR x VU transceiver. It transmits in UHF (435.525 MHz), and receives in VHF (145.910 MHz). We are allowed to use the radio-ham bands thank to a collaboration with the French radioamateur association (REF). This transceiver also includes a transponder that can be activated when the power budget is positive for the radio-ham. The two pairs of antennas are deployable with the HDRM (Hold Down and Release Mechanism) of ISIS.

OBC (On-Board Computer): The OBC is provided by ISIS. It is based around an ARM9 processor with a custom pluggable daughter to fit our needs. The OBC is equipped with two SD cards (32 GB each) to store the payload data.

EPS (Electrical Power Subsystem): The EPS is the GomSpace P31U and the battery pack GomSpace BP4 with a total capacity 37.4 Wh. The energy is produced by a total of 32 solar cells. But 8 of them are mounted on two deployable panels to give additional power.

Flight software: This was developed in-house as it is a key investment for future missions. — The general architecture and real time aspects are based on the Gericos framework. This framework has been created at LESIA for the development of the software of an instrument onboard Solar Orbiter. It is based on a RTOS (Real Time Operating System), namely FreeRTOS, which is used to share processor time between different high-level tasks. Gericos implements an AO (Active Objects) design pattern, in which different tasks corresponds to different "active objects". The processor time is shared between the different tasks by the RTOS. The framework, initially developed for LEON architectures, has then been ported for another family of microcontroller. The porting of Gericos has been done for the AT91SAM9G20 processor (on which the ISIS on-board computer is based) and STM32F303 (which is on the payload electronics).

Task

Description

ModeManager

Highest priority task, manages the state machine, process events.

EPSManager

Interface class with the electrical power system.

ComManager

Interface class with the communications systems. Receive and transmits telemetry and telecommands.

ADCSManager

Interface class with the ADCS.

PlManager

Interface class with the payload. Synchronize the payload modes with the OBC modes.

HouseKeeper

Collect housekeeping data, stores high frequency HK in a buffer.

DataRed

Lowest priority task, collect and process data from the payload.

Table 1: Tasks managed by the on-board computer software

The flight software of the on-board computer is split into two levels. The level 0 (L0) has to be fully stable and contains the vital tasks (EPSManager, EPSManager, ComManager), while the level 1(L1) takes care of higher tasks (ADCSManager, PIManager, DataRed, Mode Manager). The L1 software can be updated remotely during the mission. The software receives and transmits packet based on the CCSDS standard from ESA.

PicSat_Auto7

Figure 5: Flight model of PicSat with antennas and solar panels deployed (image credit: PicSat collaboration)

 

Launch: A launch of PicSat (~3.5 kg) as a secondary payload is scheduled for Q4 of 2017.

PicSat was initially scheduled to be lifted off by India's PSLV rocket in late June 2017, however it was put on hold as the satellite is not ready yet for the launch.

Orbit: Sun-synchronous target orbit with an altitude of 580 km.

 


 

Sensor complement: (PFP)

PFP (PicSat Fibered Photometer)

The science objectives are: 5) 6)

• The main and most important science objective of the PicSat mission is the monitoring of the transit of Beta Pictoris b's Hill sphere in front of its star, the measurement of its radius, and the detection of transiting material. Based on the light curve obtained in 1981, it can be estimated that this requires a photometric level of precision of at least 10-3 (7σ) to detect the event. A level of 10-4/hr would be better suited to characterize the transit.

• Studying exo-comets in the Beta Pictoris system: The giant planet Beta Pictoris b is not the only astronomical body which is expected to transit in front of the star. A number of smaller objects, identified as exo-comets 7), have also been detected But to this date, these exo-comets have only been seen by transit spectrometry, where absorption features are detected at certain wavelengths, revealing the gas tail of the comet. Lecavelier et al. 8) have shown, using a dedicated model of transiting exo-comets, that these objects could be observed in large band with a photometric level of precision of 10-4/hr. A secondary objective of the PicSat mission is to detect these exo-comets in the visible band, thus opening up the possibility of studying the dust tail of these objects.

• Disk inhomogeneities: The PicSat mission and its fibered instrument will provide a unique dataset of photometric measurements with the rare combination of precision, homogeneity, and long duration that only a space mission can achieve. This will help to study and hopefully to better understand the structure of this young debris disk, in which planet(s) have very recently been formed.

• Single mode fiber in space: Last but not least, the PicSat instrument has also been designed with a technical objective in mind: demonstrating our ability to properly inject star light into a single mode fiber in space. Single mode fibers are now broadly used to collect a d -guide light in astronomical instruments. They are especially useful for interferometric instrumentation, where their propensity to retain the coherence of the light is especially useful. But because of the very small diameter of the core of these fibers (3 µm), proper light injection is difficult, and requires advanced guiding/tracking techniques. The PFP (PicSat Fibered Photometer) will prove our ability to manage this injection, and pave the way to future projects, like the FIRST-S (Fibered Imager foR a Single Telescope) nanosatellite. 9)

 

Design of the PFP:

To tackle its science objective, the PicSat mission requires an instrument capable of simultaneously demonstrating the use of a single mode fiber in space, and achieving high-precision photometry. This lead to the design of a "fibered photometer", in which, contrary to all usual photometers, the 2D detector array is replaced by a SPAD (Single Pixel Avalanche Diode). The concept is shown in Figure 6.

The light coming from the stars is collected by a small optical telescope (effective diameter is 3.5 cm), injected in a SMF (Single Mode Fiber) placed in the focal plane of the telescope, and brought to the SPAD for photon counting. The fiber has a core diameter of 3 µm only, similar to the diameter of the star image on the focal plane of the instrument PSF (Point-Spread Function). Thus, to properly inject light into this fiber, a specific tracking mechanism is required, to ensure that the small fiber stays centered on the star, even in the presence of guidance system errors and/or jitter.

PicSat_Auto6

Figure 6: Schematic view of the PFP concept and control loop (image credit: PicSat collaboration)

To track the star in the focal plane of the telescope, the fiber is mounted on a two-axis piezoelectric actuator. This actuator can move the fiber in the focal plane, over a 450 µm x 450 µm area (about 1' x 1' in terms of field of view). The ADCS performance level ensures that the satellite is able to point Beta Pictoris with a sufficient precision so that its image will actually fall into the accessible range of the fiber. The piezo actuator will then move the fiber to "scan" the entire field, find the star, and track it.

To reach a level of precision of 100 ppm/hr on the photometry of the star, an excellent tracking is not enough. Some variations of the PSF of the instrument are expected (mainly due to thermal stress of the optics), and will result in instrumental variations of the photometry. To correct these, it is necessary that the "tracking" algorithm also regularly "scans" the PSF, and estimates some of its most critical parameters (mainly focus and astigmatism). To do so, the piezo actuator will constantly modulate the position of the fiber around the central position of the star. Different modulation patterns are still being studied, but so far, the pattern given in Figure 7 seems to give good results. The modulation will run at 100 Hz, with integrations of 1 ms (photometric data point are acquired at 1 kHz, and the modulation pattern contains 10 points).

All payload activities are controlled and managed by a dedicated electronic board, which embeds a 72 MHz STM32F303 microchip.

PicSat_Auto5

Figure 7: One of the modulation patterns for scanning the PSF of the star (image credit: PicSat collaboration)

An error budget for the photometric precision of this instrument is given in Table 2. The use of a Single Mode Fiber, with its very small acceptance angle, drastically reduces the scattered light noise. The use of a photon counting detector (the SPAD) removes any source of readout noise. The main sources of noise are then the injection stability (related to the tracking precision), the thermal regulation of the detector, and quantum photon noise.

Noise source

Assumption

Error (ppm/hr)

Comment

Photon noise

Mv = 3.86

60

Physical limit

Readout noise

No readout noise

0

 

Scattered light

150 e-/s

5

Moon and Earth. Filtered by SMF

SPAD bias voltage stability

150 µV

20

By design

Thermal stability

Regulated and corrected to 0.01ºC

40

By design and with data reduction

Injection stability

5% at 100 Hz

80

From simulations

Total error

110

Square-root of squared errors

Table 2: Photometric error budget

Optomechanical design: For proper injection into the fiber, the aperture ratio of the telescope must match the one of the fiber. Thus, the aperture ratio of the PicSat telescope is constrained to F/D = 4. The telescope must also fit into a single CubeSat unit. A compact 30° off-axis Newtonian design was selected, with an effective diameter of 3.7 cm, and a focal length of 14.8 cm. The primary off-axis parabola used is oversized (50 mm diameter) to ensure optimum optical quality on the edges, made of pure aluminum. The secondary mirror is a plane mirror, of 22 mm diameter. The overall optical design is shown in Figure 8.

PicSat_Auto4

Figure 8: Optical design of the PFP (image credit: PicSat collaboration)

The secondary mirror is mounted on a specific mechanical assembly which allows the fine tuning of its alignment with respect to the primary parabola thanks to the use of 3 peelable washers. A 3D rendering of this assembly is given in Figure 9. This assembly is mounted on top of the optical tube. The general mechanical assembly, integrated into a standard CubeSat unit is shown in Figure 10. The whole structure is made out of aluminum, so that thermal variations are expected to produce homogeneous deformations, minimizing their impact on optical quality of the system.

The fiber is mounted on the two-axis piezo actuator, and positioned in the focal plane of the telescope. The piezo actuator is based on space qualified components, and made by CEDRAT Technologies (Grenoble, France).

A space is left on the baseplate, next to the body of the telescope, to host the star tracker (part of the ADCS). This will ensure a rigid link between the two optical systems, and will help to maintain the relative position of their line-of-sights. This is of prime importance for the two-stage tracking system (ADCS + payload piezoelectric system).

PicSat_Auto3

Figure 9: A 3D view of the secondary mirror support (image credit: PicSat collaboration)

PicSat_Auto2

Figure 10: Mechanical assembly of the PFP (PicSat Fibered Photometer), integrated into a standard CubeSat unit (image credit: PicSat collaboration)

Electronics system of PFP: A dedicated electronic has been designed and realized for the PFP. This board includes all the required electronics to manage data acquisition, communications to and from the payload, and to run the tracking algorithm. Typical power consumption is 0.2 W in "Standby mode", and 2 W in "Science mode".

The electronic board embeds a 72 MHz STM32F303 microchip. This component, whereas not space qualified, combine the necessary computing performance to run a Kalman filter based algorithm at high frequency (1 kHz) with a low power consumption profile.

The two-axis piezo actuator is driven by 2 DACs (Digital-to-Analog Converters), and a high voltage (150 V) source. The typical displacement step is 0.01 µm, and the total accessible range is 450 µm on both axes.

The SPAD is connected to a 5 V power supply and to a -25 V bias voltage source which are regulated by the board itself to ensure an excellent level of stability (100 µV), as this is critical for the overall photometric precision. The output pin of the SPAD (electron cascade signal) is connected to a 4b-bit hardware counter. This counter is activated/deactivated by a 3.3 V signal outputted by one of the hardware timers of the STM microchip. This ensures a proper timing of the integration cycles.

The SPAD retained for the PFP has a built-in TEC (Thermo-Electric Cooling) system, which can be used to regulate the temperature of its single pixel. The TEC is managed by a dedicated TEC control system, on the payload board. A simple DAC is used to set the temperature set-point (between 2ºC and 16ºC). The system provides good thermal stability (down to 0.1ºC) over extended conditions.

Communications in both reception and transmission is achieved through a standard UART/RS422 4-wire interface.

Tracking algorithm: As stated above, one of the main role of the payload board of the PicSat Fibered Photometer to control 2-axis piezo actuator on which the SMF is mounted, to correct for pointing vibrations induced by the ADCS. The payload board has also in charge the initial scan of the focal plane of the telescope at first acquisition, or when the tracking is lost.

The algorithm used in the PicSat Fibered Photometer is based on Kalman filtering. To describe the state of the system, a standard position-velocity-acceleration representation could be used. However, in our case, the acceleration originates in the jitter created by the ADCS, and must be considered as a source noise. We thus use a simple position-velocity representation, in which we denote X the state of the system (the interested reader is referred to Ref. 5).

Environmental teats: Many of the elements of the Picsat Fibered Photometer (especially the electronics) are off-the-shelf industrial component, which are not space-qualified. To ensure that these components would net fail under the harsh condition of the launch/space environments, the project conducted vibration and thermal-vacuum test campaigns.

Vibrations: The full instrument (optomechanical + electronics) was integrated into a 3 unit CubeSat structure, and tested in vibrations. Quasi-static tests (5 – 100 Hz, at 9.7 g), sinusoidal vibrations (5.0 – 8.0 Hz at 10 mm amplitude, and 8 – 100 Hz at 4.5 g), and random vibrations (6.7 g during 120 s) were performed. The instrument systems showed no obvious sign of damage, and all subsystems responded well to a detail health-check. In particular, a complete scan of the optical response (instrumental PSF) before and after the tests proved that the optical alignment withstood the vibrations.

Thermal vacuum test: The full instrument, also integrated in a 3 units CubeSat structure, was placed in a thermal vacuum chamber, and was cycled between -20°C and +30°C. The electronics responded well under any of these conditions, but the secondary mirror, initially made of Zerodur glass ceramic and dielectric coating was damaged. The Zerodur-dielectric combination is not known to be resilient to the space environment8, and the mirror is currently being replaced by a space-qualified component. Otherwise, all other subsystems (including the primary parabola, the electronic board, the SPAD, the thermal regulation system, etc.) withstood the test very well, and we are now confident that the photometric performance of the PicSat Fibered Photometer will not be impacted by the space environment.

PicSat_Auto1

Figure 11: Results of a Matlab/Simulink simulation presenting the position of the star with respect to the SMF when the tracking algorithm is off (left panel) and on (right panel), image credit: PicSat collaboration

By combining an innovative approach to photometric observations using a single mode fiber and a single pixel avalanche detector, together with a more standard optical design (an off-axis Newtonian telescope), the PicSat Fibered Photometer achieve high-precision photometry while keeping an overall size, mass and power consumption extremely low. The final instrument fits in a standard CubeSat unit (10 cm x 10 cm x 10 cm), with a mass of only 1.3 kg, and is capable of achieving 100 ppm/hr photometric precision on stars with Mv < 5.

PicSat_Auto0

Figure 12: Payload qualification model. From bottom to top: electronic, piezo-stage and telescope (image credit: PicSat collaboration)

 


1) "PicSat - A nanosatellite to observe Beta Pictoris," May 2017, URL: https://picsat.obspm.fr/picsat/

2) Vicent Lapeyrere, Sylvestre Lacour, Lester David, Mathias Nowak, Antoine Crouzier, G. Schworer, P. Perrot, S. Rayane, "PicSat: a CubeSat mission for exoplanetary transit detection in 2017," Proceedings of the 31st Annual AIAA/USU Conference on Small Satellites, Logan UT, USA, Aug. 5-10, 2017, paper: SSC17-III-09, URL: http://digitalcommons.usu.edu/cgi/viewcontent.cgi?article=3617&context=smallsat

3) "Beta Pictoris as seen in infrared light - annotated,", ESO, 2008, URL: https://www.eso.org/public/images/eso0842b/

4) https://picsat.obspm.fr/picsat/the-satellite/

5) M. Nowak, S. Lacour, V. Lapeyrère, L. David, A. Crouzier, G. Schworer, P. Perrot, S. Rayane, "A Compact and Lightweight Fibered Photometer for the PicSat Mission," Proceedings of the 31st Annual AIAA/USU Conference on Small Satellites, Logan UT, USA, Aug. 5-10, 2017, paper: SSC17-VI-01, URL: http://digitalcommons.usu.edu/cgi/viewcontent.cgi?article=3634&context=smallsat

6) M. Nowak, S. Lacour, V. Lapeyrère, L. David, A. Crouzier, C. Dufoing, H. Faiz, T. Lemoult, P. Trebuchet, "Reaching sub-milimag photometric precision on Beta Pictoris with a nanosat: the PicSat mission," Proceedings of the SPIE, Volume 9904, 'Instrumentation and Methods for Astrophysics (astro-ph.IM),' id. 99044L 7 pp. (2016), DOI: 10.1117/12.2232508, URL: https://arxiv.org/pdf/1610.07802.pdf

7) Flavien Kiefer, Alain Lecavelier des Etangs, Jean-Charles Augereau, Alfred Vidal-Madjar, Anne-Marie Lagrange, Herve Beust, "Exocomets in the circumstellar gas disk of HD 172555," Astronomy & Astrophysics Letters, January 8, 2014, DOI: 10.1051/0004-6361/201323128, URL: https://arxiv.org/pdf/1401.1365.pdf

8) A. Lecavelier des Etangs, A. Vidal-Madjar, R. Ferlet, "Photometric stellar variation due to extra-solar comets," Astronomy & Astrophysics Letters, 1999, URL: https://arxiv.org/pdf/astro-ph/9812381.pdf

9) S. Lacour, V. Lapeyrère, L. Gauchet, S. Arroud, R. Gourgues, G. Martin, S. Heidmann, X. Haubois, G. Perrin, "CubeSats as pathfinders for planetary detection: the FIRST-S satellite," Proceedings of SPIE 2014, 'Astronomical telescopes and instrumentation - Montreal,' DOI: 10.1117/12.2057381, URL: https://arxiv.org/pdf/1407.4294.pdf
 


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

Spacecraft    Launch    Sensor Complement   References    Back to top