1 Introduction

Schiaparelli originated from the ExoMars programmatic objective for the development, in-flight and in situ demonstration of Entry, Descent and Landing (EDL) of a payload on the surface of Mars. In 2009, ExoMars was split into two missions: the first, for launch in 2016, to include a Mars orbiter for science and data relay; the second to include the ExoMars Rover with its Pasteur Payload. Since the landing system of the Rover mission was conceived to be led by an international partner (initially NASA, then later Roscosmos), the desire to retain the EDL demonstration objective for implementation by Europe led to the decision to embark an ‘EDL Demonstrator Module’ (EDM) on the 2016 Trace Gas Orbiter (TGO). Several elements of the EDM were derived from those already under development for the landing platform of the original ExoMars mission architecture. Emerging from industrial studies in mid-2009 were ‘large’ (550-600 kg entry mass) and ‘small’ (∼300 kg) options, from which the larger option was selected. The EDM was to be separated from the TGO 3 days prior to Mars arrival on a hyperbolic approach trajectory and (following wake-up from hibernation during the coasting phase) enter the martian atmosphere, transmitting essential, real-time telemetry via UHF to the TGO and (after touchdown) also to other Mars Orbiters for relay of its data to Earth.

Although for some time during development the possibility to include a Radioisotope Thermoelectric Generator (RTG) power source and Radioisotope Heater Unit (RHU)-supported thermal control (thus enabling long-duration surface operations) was studied in co-operation with Roscosmos, the initial concept—and indeed final configuration—was foreseen for operations on Mars of only a few sols, based on the electrical energy of on-board batteries. Long-duration surface operations (as might be afforded by an RTG or solar arrays) were not necessary for achieving the defined programmatic objectives of EDL demonstration. Note that solar array power would also have been vulnerable to the possibility of dust storms at the time of arrival, at a solar longitude of 244.7°, during the statistical dust storm season.

Agency reviews were held as part of the overall ExoMars 2016 mission system reviews as follows:

  • Preliminary Design Review (PDR): Oct-Dec 2010;

  • Critical Design Review (CDR): Oct 2013-Apr 2014;

  • Qualification and Acceptance Review (QAR), combining Qualification Review and Flight Acceptance Review: Aug-Oct 2015.

Lessons learned from Beagle 2 and recommendations from its commission of enquiry were applied at programmatic, mission system and technical levels, and the status with respect to the recommendations was assessed at each agency-level review.

The EDM was named Schiaparelli in November 2013 in honour of the Italian astronomer Giovanni Schiaparelli.

The Schiaparelli industrial consortium was led by Thales-Alenia Space Italy (TAS-I) in Turin and included the partners / suppliers listed in the Appendix.

In addition, several elements of payload were provided via ESA as ‘Customer-Furnished Items’ (CFI):

  • COMARS+ entry sensors (DLR and CNES);

  • DeCa descent camera (ESA);

  • DREAMS surface environmental sensors package (ASI), selected via an Announcement of Opportunity.

  • INRRI optical retroreflector (ASI).

2 Overall Configuration

Figure 1 shows the overall external configuration of Schiaparelli for Mars atmospheric entry (minus thermal blanket). The 2.4 m diameter module has a 70° sphere-cone Front Shield (FS) with 600 mm nose radius, and a truncated conical Back Shell (BSH) with 47° cone angle. The aeroshell shape was selected as the lowest risk option, based on the extensive proven heritage at Mars from Viking onwards, plus previous European experience including NetLander. The diameter was reduced from the 3.4 m foreseen for the original ExoMars configuration (landing platform + rover).

Fig. 1
figure 1

Overall dimensions and system reference frame of Schiaparelli (including Thermal Protection System (TPS) tiling but excluding Front Shield Multi-Layer Insulation (MLI) and Back Cover Single-Layer Insulation (SLI)). The MSA interface that remained on Schiaparelli after separation is visible in the recess (1 of 3 identical) on the back cover. Also visible just below this is the Sun Detection Sensor (SDS). Credit: TAS-I

Full size image

Schiaparelli was mounted on the TGO via the Main Separation Assembly (MSA), which interfaced to recessed brackets at three positions in the back cover, and on the TGO side on the top deck above the 591 mm-radius central tube of the primary structure (Fig. 2). The MSA (Fig. 3) supported Schiaparelli for launch and cruise, provided routing for electrical harness, and provided the necessary electrical and mechanical separation on approach to Mars. The total MSA mass was 39.3 kg, of which 30.3 kg remained on the TGO after separation.

Fig. 2
figure 2

Accommodation of Schiaparelli on the TGO (launch configuration). The mechanical loads of Schiaparelli are transferred via the MSA to the TGO central tube. Credit: TAS-I

Full size image
Fig. 3
figure 3

Main Separation Assembly (MSA), including the interface fittings that remained on the EDM after separation. The EDM is above and TGO is below in the view shown. The separation plane is between the cyan and magenta parts shown above. Each of the three separation mechanisms used a spring release, initiated via a Non-Explosive Actuator (NEA), and incorporated a 128-pin umbilical separation connector. For further details, see Caldirola et al. (2015). Credit: TAS-I/RUAG

Full size image

The nominal relative delta-V of separation was 0.32 m s−1, with a nominal rotation rate of 2.70 rpm for spin stabilisation. This was within a range specified between a minimum of 2.5 rpm (to guarantee exo-atmospheric stability) and a maximum of 3.0 rpm (driven by the need to allow aerodynamic torques to orient the probe to maintain a low angle of attack at peak heating). The spin was imparted by the angled MSA mechanism rather than spinning the whole composite.

Figure 4 shows an exploded view of the main elements of Schiaparelli.

Fig. 4
figure 4

Exploded view of the Schiaparelli EDM, attached to the Main Separation Assembly (MSA). Credit: TAS-I

Full size image

The conical element of the BSH, called the Back Cover (BCV), is made from aluminium honeycomb with CFRP skins. The BCV incorporates cut-outs for the three Main Separation Mechanism (MSM) brackets (i.e. interface to the MSA), access doors, fill & drain valves, venting, Sun Detection Sensor and UHF antenna. The BCV is connected to the FS structure via the 6-point Front Shield Separation Mechanism (FSSM). Inside the BCV, 12 struts (6 bipods) carry loads between the Surface Platform and Upper Ring (around the parachute cover), and between the Surface Platform and MSM brackets; six struts connect the Upper Ring with the six-point Surface Platform Separation Mechanism (SPSM) around the perimeter of the Surface Platform, while each of the three MSM brackets is connected with two shorter struts to adjacent pairs of SPSMs. There are two types of SPSM (arranged alternately around the perimeter); three with the spring-actuated separation mechanism and three carrying electrical connector interfaces (Prieto et al. 2013).

A cross-section of Schiaparelli is shown in Fig. 5. It should be noted that no redundancy was incorporated in the design, except where specified by the relevant standard (e.g. CAN Bus and pyro lines).

Fig. 5
figure 5

Schiaparelli cross-section. Credit: TAS-I

Full size image

The as-flown entry mass of Schiaparelli was 576.3 kg. Table 1 gives a mass breakdown.

Table 1 Schiaparelli mass breakdown (simplified w.r.t. original project documentation)
Full size table

The nominal EDL sequence for Schiaparelli is shown in Fig. 6.

Fig. 6
figure 6

Schiaparelli nominal EDL sequence. Copyright: ESA/ATG medialab

Full size image

3 Heatshield and Entry Parameters

A tiled Norcoat® Liège Thermal Protection System (TPS), bonded with silicone adhesive ESP495, covers both FS and BSH. The FS TPS comprises a central dome tile, surrounded by a ring tile, 3 rows of tiles on the conical section (of 6, 12 and 18 tiles, respectively) and finally a row of trailing edge tiles. The BSH TPS comprises a parachute break-out patch at the apex and 4 rows of tiles on the conical section.

The FS TPS incorporates:

  • 9 penetrations for MLI grounding;

  • 4 closing plugs at ground handling attachment points (clearly visible in Fig. 8a);

  • 7 thermal plugs for FS sensors, and

  • 4 pressure sensor inlet holes.

The BSH TPS incorporates:

  • 24 penetrations for SLI grounding;

  • 3 penetrations for attachment to the MSA;

  • one break-out patch for the parachute;

  • 6 access doors (one of which is closed by a patch holding the UHF antenna under an RF-transparent Norcoat® 4011 TPS patch);

  • 4 access doors for fill and drain of the RCS (propellant tanks and pressurant tank);

  • Venting box (a HEPA filter not being used, finally);

  • Sun Detection Sensor (see GNC section);

  • 3 thermal plugs;

  • 4 COMARS+ sensor locations.

Figures 7 and 8 show the BSH and FS layouts.

Fig. 7
figure 7

Back Shell layout with TPS tiles, showing the parachute break-out patch, MSA interface points, access doors, UHF antenna, Sun Detection Sensor and venting. Also visible in the lower left quadrant are the locations of the COMARS+ sensor heads, COMARS+ radiometer and the three Back Shell TPS thermocouple plugs. Credit: TAS-I

Full size image
Fig. 8
figure 8figure 8

Views of the Schiaparelli Front and Back Heat Shield TPS during assembly. Copyright: Airbus Defence and Space SAS 2014 A. Gilbert. The images are available on the ESA Exploration web site http://exploration.esa.int/

Full size image

Entry parameters are summarised in Table 2, and more information on the thermal aspects is available in Perotto et al. (2017).

Table 2 Summary of Schiaparelli entry parameters (format adapted from Davies and Arcadi 2006)
Full size table

4 GNC

The Schiaparelli GNC architecture was based around an inertial measurement unit (Honeywell MIMU), supplemented with a Sun Detection Sensor (a TNO Mini-FSS) in the Back Shell, and, following FS separation, a four-beam Radar Doppler Altimeter (RDA). During coasting hibernation the IMU was powered off; attitude reconstruction on exit from hibernation was performed based on SDS data.

The RDA operated at 35.76 GHz. Three beams 120° apart in azimuth were angled 20° off-nadir from a central beam. A field test campaign was performed for characterisation of the RDA performance.

Note that the final freefall would have been from 1.8 m following cut-off of the RCS; actual touchdown would thus have been open-loop, the landing deceleration being measured by both the IMU and landing accelerometers (4 s of sampling from RCS OFF).

See Amelio et al. (2016) for details of the avionics and GNC architecture, and Bombaci et al. (2016) for a description of the RDA.

5 Parachute System

The Schiaparelli Parachute System (PAS) comprised a nylon Disk-Gap-Band (DGB) parachute, its deployment bag, the Parachute Deployment Device (PDD) mortar and the PAS cover, referred to as the Break-out-Patch (BoP) when covered with TPS, and a 1.83 m diameter BoP drogue parachute. The PAS is shown in cross-section in Fig. 5. The main parachute has a diameter 12 m with a geometric porosity of 22.4%, scaled from the Huygens second stage parachute (see Fig. 9). The parachute was designed to decelerate Schiaparelli from up to Mach 2.1 (nominally 1.95 ± 0.15, 3\(\sigma \)) to a sub-sonic speed compatible with Surface Platform separation, and to achieve a flight path angle \(<-87\)° at a height above ground >1.4 km. The parachute was design to inflate under dynamic pressure at deployment between 632 and 928 Pa (in the probe’s wake). The design maximum load limit was 75 kN.

Fig. 9
figure 9

Schiaparelli parachute in deployed configuration. Credit: TAS-I

Full size image

The three parachute bridles are attached to the BCV Upper Ring.

Final functional and structural qualification of the Schiaparelli parachute was conducted at NASA Ames’ NFAC facility.

6 Surface Platform

The need to support EDL functions until touchdown, and then maintain operations on Mars for the relay of the full EDL dataset and science data, gave rise to a Surface Platform design based around a circular baseplate carrying both EDL elements (‘Propulsion Bay’) and a warm compartment (‘Central Bay’). Mounted underneath the baseplate are the RTPU, the crushable structure for touchdown, and the RDA (circular support panel carrying electronics unit, RF network and 4 antennas). Crushable structure was selected for the landing technology following a trade-off (in 2010 prior to System PDR) vs. landing legs, vented airbags and non-vented airbags. Although legs or vented airbags could be valid alternatives, crushable structure was found to be the simplest solution for the mission objectives. The core of the crushable structure is an aluminium honeycomb (Alcore 3/8-5052-0.0007 1.0) with crushing strength 0.17 MPa. The maximum axial quasi-static deceleration load experienced by equipment on the Surface Platform was expected to be approximately 48 \(g\).

After the touchdown phase only the Central Bay was needed to remain active to continue operations. Figures 10 and 11 show the layout of the Surface Platform.

Fig. 10
figure 10

Schiaparelli Surface Platform (viewed without thermal blanket). Credit: TAS-I

Full size image
Fig. 11
figure 11

Side view of the Surface Platform. Credit: TAS-I

Full size image

7 Reaction Control System

Hydrazine monopropellant was stored in three 17.5 l tanks (incorporating diaphragms for propellant management), supplying three clusters (120° apart) of three CHT 400N pulsed engines (nozzles canted 18° outboard from nadir, with the outer pair of each cluster tilted by 13° for roll control). The system was pressurized from a 15.6 l He tank. A schematic of the RCS is shown in Fig. 12.

Fig. 12
figure 12

Schiaparelli Reaction Control System overview. Credit: TAS-I

Full size image

With respect to the as-built configuration, pressure sensors were removed in Baikonur due to a late concern about welds. This represented a minor reduction in the data available for post-flight analysis rather than a performance loss.

8 Thermal Design

Schiaparelli’s thermal design had to cope with the variable thermal environments during the various phases of the mission from launch to the surface of Mars, including a 72 hours coasting phase between separation from TGO and atmospheric entry. See Perotto et al. (2017) for details of the design and of its performance in flight.

For cruise, passive thermal control employed SLI and MLI on the FS and BSH interior and exterior, and on the exterior of the surface platform. Active thermal control was used when needed by means of heaters, powered during cruise from TGO, then from the on-board batteries after separation (controlled by a mix of thermostats (for within the Central Bay) and the Thermal Control System (TCS) in the CTPU).

The Central Bay warm compartment employed passive thermal design elements in addition to the battery-powered heaters: polyimide foam insulation (sides and top), and a gas gap with standoffs below the support plate.

Thermal boosts were implemented before separation from TGO (to minimise electrical heating energy needed from the Schiaparelli batteries during coast) and another just before entry (mostly to warm up the RCS).

Thermal capacitors (using wax phase change material – see Gottero et al. 2014) were accommodated under the MIMU (97 kJ), CTPU (216 kJ) and UHF Transceiver (149 kJ) to prevent overheating from the heat dissipated from these units during cruise and EDL, and for the CTPU and Transceiver also during the surface phase; the stored dissipated heat from these units could be released during periods of low dissipation to support heating of the warm compartment.

EDM-level verification by thermal vacuum / thermal balance test was only for the cruise, coast and thermal boost (under vacuum), while for EDL and surface phase the verification was only by analysis. See Gottero et al. (2016) for more detail.

9 Planetary Protection

Schiaparelli was classified as Planetary Protection Category IVa, being a landed system neither carrying life-detection equipment nor accessing a ‘Special Region’ of Mars. The following bioburden requirements were thus applied (vs. values achieved at launch):

  • total (Encapsulated and Surface) bioburden of less than 5×105 spores; (achieved at launch: 291601 spores)

  • less than 3×105 spores on exposed internal and external surfaces; (achieved at launch: 151360 spores)

  • less than 300 spores/m2 average on exposed internal and external surfaces. (achieved at launch: 73 spores/m2)

DHMR was used on hardware shown to be compatible with that sterilization process, and sterile IPA 70% was used to maintain surface bioburden density below acceptable levels. Integration of Schiaparelli was performed in an ISO 7 Highly Controlled facility (ISO7HC).

An airtight bioseal ring was implemented between the BCV and FS to guard against contamination during integration activities with closed heatshield in facilities outside of ISO7HC. This bioseal had the additional design role to help prevent incursion of hot gases during entry.

The initial design included a HEPA filter for venting during launch but problems with its performance venting air with ambient humidity at low pressures led to its replacement with a simpler venting box design (HEPA filter Structural Model housing, with steel wool packed inside) while still maintaining acceptable bioburden levels and thermo-fluidic performance during ascent and entry.

10 Electrical Power

During cruise, Schiaparelli was powered from the TGO unregulated bus (routed via the MSA and RTPU). After separation from TGO, electrical power for Schiaparelli was provided by three batteries (charged during cruise):

  • CTPU (NL) battery mounted inside the Central Bay, using ABSL 18650NL Li-ion cells in 8s27p configuration (nominal capacity 62.1 Ah) to supply a 28V unregulated bus;

  • RTPU (NL) battery for EDL phase only, mounted externally, using ABSL 18650NL Li-ion cells in 8s38p configuration (nominal capacity 87.4 Ah) to supply a 28V unregulated bus;

  • RTPU (HR) battery for high rate needs during EDL (separation pyros, parachute deployment device, RCS valves, heaters), using ABSL 18650HR Li-Ion cells in 15s3p configuration (nominal capacity 3.3 Ah) to supply a 60 V unregulated bus.

The electrical power system is described in more detail by Ciancetta et al. (2017).

The DREAMS experiment package was equipped with its own battery (ABSL 18650NL Li-ion cells in 8s3p configuration), charged before launch.

11 CTPU and RTPU

Schiaparelli’s main computer was the Central Terminal Power Unit (CTPU), based around an Atmel AT697F Leon2-FT processor, including hibernation timer and 4 GBytes of mass memory, power distribution and thermal control for other units. During hibernation, the power consumption of the CTPU itself was <1 W, most of the electrical power being spent for thermal control.

The CTPU was housed inside the Central Bay. CAN (Controller Area Network) Bus interfaces linked the CTPU with the RTPU, RDA, UHF Transceiver and DREAMS. RS422 interfaces linked the CTPU with the UHF Transceiver for TM/TC, the MIMU and DeCa. The data interface with the TGO was via MIL-1553. For EDL functions such as pyro and thruster valve control, interfaces to the Sun Detection Sensor, entry sensors, COMARS+ and RDA, a separate Remote Terminal and Power Unit (RTPU) was mounted on the underside of the baseplate. The RTPU was not intended to be used after touchdown.

The avionics block diagram (Fig. 13) shows the interconnections between the CTPU, RTPU and other units of Schiaparelli.

Fig. 13
figure 13

Schiaparelli Avionics subsystem block diagram. UTCVR = UHF Transceiver; USW = UHF Switch; PBAT1 = CTPU (NL) Battery; PBAT2 = RTPU (NL) Battery; RBAT = RTPU (HR) Battery; RDE = RDA Drive Electronics. Credit: TAS-I

Full size image

12 Telecommunications

Communications with TGO and the other Mars orbiters was by means of a UHF Transceiver (developed as an evolution of the design used on Beagle 2) and RF switch, feeding a patch antenna on the back cover (under an RF-transparent TPS patch) for entry and descent until Surface Platform separation, and a quad-helix antenna (Fig. 14) on top of the Surface Platform Central Bay, for the remaining portion of the descent, touchdown and surface phase.

Fig. 14
figure 14

Surface Platform UHF antenna (height 211.5 mm, base diameter 236 mm). Credit: Rymsa

Full size image

Transmission of real-time essential telemetry at 8 kbps was performed during EDL (TM being available from ∼25 min before reaching the Entry Interface Point). This was received by the Electra telecommunications package on TGO. The low telemetry rate necessitated a selection of the most important parameters for transmission, sampled at 0.1 Hz, 1 Hz or 10 Hz depending on the parameter and mission sub-phase (higher rate data was stored on-board for transmission after touchdown). No interleaving was implemented to cope with the plasma black-out expected during entry, with the exception of an ‘event’ indication giving the maximum measured deceleration.

The carrier was also monitored by ESA’s Mars Express and the GMRT (Giant Metre-wave Radio Telescope) in Pune, India; see Esterhuizen et al. (2019).

Open-loop transmission was planned to continue during the touchdown phase until 15 min after landing, at which point the CTPU was planned to enter its first hibernation phase, having first activated DREAMS to begin its surface measurement phase.

For the surface relay phase, Mars Odyssey, Mars Express, MRO, MAVEN and TGO were all prepared to receive UHF transmission from Schiaparelli (see Edwards et al. 2017 for details), however TGO’s initial 4 sol, highly elliptical orbit after MOI limited its use for Schiaparelli post-landing to a single opportunity for open loop carrier monitoring. Passes were planned as shown in Table 3 until the estimated point that the DREAMS battery was to reach depletion; relay passes were defined thereafter according to orbiter availability until CTPU battery depletion including margin.

Table 3 Planned telemetry sessions with Schiaparelli during separation, EDL and surface phases. Passes of Mars orbiters and carrier DTE (Direct To Earth) opportunities over the Schiaparelli landing site. MRO = Mars Reconnaissance Orbiter, MEX = Mars Express, MAV = MAVEN, ODY = Mars Odyssey. ∗Carrier only for landing survival verification, i.e. no data relay to Mars orbiters (but data transfer from DREAMS to CTPU still expected). Between these sessions the CTPU was set to be in hibernation, with DREAMS following its own measurement sequence (see Bettanini et al. 2018)
Full size table

13 Payload Investigations

Although designed primarily as a technological demonstrator rather than as a platform for science, Schiaparelli carried a payload of experiments with total mass of approximately 15.2 kg (Table 4). The payload was a combination of: engineering sensors built into the vehicle, for heatshield monitoring during entry (Fig. 15) and for measuring the dynamics of the touchdown; sensors provided by ESA and national agencies, and experiments selected in response to an Announcement of Opportunity (released in November 2010) seeking proposals both for surface payload and for entry and descent science.

Fig. 15
figure 15

Schematic layout of sensors on Schiaparelli heat shield. Copyright: ESA/ATG medialab

Full size image
Table 4 Schiaparelli Payload overview and masses
Full size table

13.1 DREAMS

The DREAMS experiment package (‘Dust characterization, Risk assessment and Environment Analyser on the Martian Surface’) comprised six sensors plus associated electronics and battery (charged before launch). Two sensor masts were mounted on the top external surface of the Central Bay, for meteorological and electrical properties measurements, respectively. A further sensor for atmospheric pressure measurement was accommodated internally, with a tube protruding to the exterior to access the atmosphere. Communications with the CTPU were via a CAN Bus connection. DREAMS was planned to operate a timeline of measurement and data transfers to the CTPU during the surface phase, starting at touchdown. See Esposito et al. (2018) and Bettanini et al. (2018) for details.

13.2 DeCa

The Descent Camera (‘DeCa’) was supplied by ESA using a flight spare Visual Monitoring Camera from the Herschel-Planck project (VMC-HP), manufactured by OIP. Having been delivered to ESA in 2006 but not used for flight, it was available as an inventory item and selected for delta-qualification (using the VMC-HP EQM, which was also available and later incorporated into the EDM ATB), calibration, planetary protection cleaning and delivery for use on Schiaparelli.

The camera housing was \(97\times 97\times 83.5\) mm3 (excluding the optics and connectors), with an electrical interface compatible with Schiaparelli’s 28 V unregulated power supply. The Schiaparelli CTPU was equipped with a DS16 (RS422-like) serial data interface to receive the data from DeCa. The camera operated without telecommand, simply acquiring its sequence of images after power-on, the image data once acquired then being available for readout.

The objectives for DeCa were to obtain images of the martian surface in the vicinity of the landing area, and thus enable confirmation of the Surface Platform’s attitude and position. Mounted on a dedicated bracket at the edge of the surface platform (see Fig. 11) with its optical axis angled 20° from the platform’s +X axis, it was intended to acquire a sequence of 15 images of the martian terrain during descent, with 8-bit /pixel resolution. The time interval between images and the exposure time were pre-set to 1.5 s and 1 ms, respectively.

The image field of view through the fixed optics was 60° × 60°. Stored within the camera for as long as it is kept powered, the image data (one image cycle of 15 images corresponding to 3.76 Mb = 3947640 bytes of image data) was planned to be retrieved by the CTPU after touchdown for subsequent relay to ground. The detector is a Star-1000 APS sensor, of which the centre \(512\times 512\) pixels are used.

While the received housekeeping telemetry shows that image acquisition had started as planned, due to the EDL anomaly no images could be retrieved from the camera.

13.3 COMARS+

The COMARS+ sensor suite accommodated on the Back Shell operated during entry. See Gülhan et al. (2018, 2019) for details and results.

13.4 INRRI

A passive optical retroreflector was mounted on a bracket on top of the Surface Platform’s Central Bay. See Dell’Agnello et al. (2017) for details.

13.5 AMELIA

The AMELIA team was selected to perform scientific analysis of the EDL data. See Aboudan et al. (2018) and Ferri et al. (2019).

14 Landing Site

The landing site of Schiaparelli was chosen by ESA early in the project (2009) based on engineering constraints alone. Meridiani Planum was selected as the safest (in terms of elevation and surface characteristics) known site having ‘ground truth’ characterisation (from the Opportunity rover).

The targeted site was at 6.10°W / 2.05°S, at −1.45 km MOLA elevation. From mission analysis (including variation of arrival azimuth with the actual launch date within the window) and Monte Carlo simulation, a 3-\(\sigma \) ellipse of \(110\times 24\) km (major axes) was defined.

The atmospheric conditions were defined by the European Mars Climate Database (EMCD), a specific version of which (4.3.2) was generated for Schiaparelli. Knowledge of martian atmospheric conditions was also expected to be updated in the days prior to EDL with the help of data from other missions including MRO, Mars Express and Opportunity.

15 Touchdown & Surface Phases

The nominal duration of the surface phase was limited by the available energy from the on-board batteries, with some dependence on the thermal environment encountered by Schiaparelli at the landing site (with arrival at a Local Mean Solar Time (LMST) of 13:00). DREAMS’ own battery would likely have become depleted before that of the CTPU, in which case Schiaparelli would have completed relay transmission of any remaining DREAMS data transferred to the CTPU prior to expiry of DREAMS, together with housekeeping data from the platform. After that, any remaining opportunities for relay would have been used to attempt repeat transmission of the data accumulated in memory.

Although in principle forward commanding of Schiaparelli during the surface phase would have been technically possible, this was not needed within the scope of the (few sols) nominal mission, and thus not planned.

16 Schiaparelli Model Philosophy

The design and verification of Schiaparelli required several development models to be built. The principal models were the Structural Model, the Drop Test Model, the Avionics Test Bench (ATB), and the Flight Model. These models and several others were used for design, development and co-ordinated analysis and testing with TGO according to mission phase and spacecraft configuration, as described in Table 5.

Table 5 Verification activities carried out for the various Schiaparelli EDM functions and mission phases
Full size table

16.1 Structural Model

The main purpose of the Schiaparelli SM was to verify the mechanical properties of the vehicle’s main structural elements – back shell, front shield and surface platform, and the interfaces among them. The SM was subjected to sine testing, entry load static test and parachute deployment static load test. The SM was also used to perform leak testing of the bioseal and to verify DHMR compatibility of the main structural parts.

16.2 Surface Platform Drop Test Model

The purpose of this model was to test the platform’s robustness to landing scenarios (velocity and surface characteristics), in particular with the crushable material for damping the final touchdown (del Campo et al. 2016).

16.3 Avionics Test Bench

This bench-top model was used to verify the electrical and digital interfaces between the various elements of Schiaparelli, and to test the on-board software.

16.4 Flight Model

This was the only complete model of Schiaparelli assembled (Fig. 16). Flight spares of some elements were available in case of need.

Fig. 16
figure 16

View of Schiaparelli at Baikonur after mounting to the TGO, showing installation of the FS MLI (visible open flaps – closed soon after this photo was taken – were for access to the ground handling points). Copyright: ESA - B. Bethge

Full size image

17 Launch, Cruise, Commissioning and Check-Outs

During launch (and most of the cruise), Schiaparelli was passive except for thermal control from the TGO, via the RTPU (‘EDM Service Mode’).

Following launch on a Proton-M on 14 March 2016, ExoMars 2016 experienced a nominal 7-month cruise to Mars, including two Deep-Space Manoeuvres. Following these, two small Trajectory Correction Manoeuvres were performed. A possible third and final TCM was not performed due to the near-perfect navigation achieved. See Pellegrinetti et al. (2019) for details of the flight dynamics operations and targeting for EDL.

During cruise, Schiaparelli was commissioned and in-flight check-outs and battery charging performed, with total duration of 55.7 hours. Figure 17 shows the cumulative duration of CTPU operations. Commissioning and calibration for the pre-separation thermal boost were performed in the period 8-21 April 2016. Mid-cruise checkout was performed in the period 16-23 June 2016. A third set of checkouts was performed 8-21 Sept 2016, with preparations for separation beginning 28 Sept 2016.

Fig. 17
figure 17

Cumulative duration of Schiaparelli CTPU operation during cruise

Full size image

High disturbance torques were detected when Schiaparelli was exposed to the Sun during a slew early during cruise (triggering an autonomous TGO reaction wheel unloading). Following further outgassing characterisation slews (also needed for TGO), some of the torques were indeed attributed to outgassing from Schiaparelli (perhaps from absorbed moisture in the TPS). Baking operations (slews to illuminate different faces of the spacecraft) were performed during mid-cruise (26 May to 4 July 2016) until the torques had reduced to an insignificant level.

Separation of Schiaparelli occurred nominally on 16 October 2016, 3 days before Mars arrival (note: the date of Mars arrival was fixed before launch, independent of the actual launch date within the 2016 launch window). Telemetry for nearly 19 min around separation was received from Schiaparelli during this phase by the Electra UHF relay on TGO, operating in open-loop recording mode.

As reported by Bonetti et al. (2018), it was found from reconstruction of the separation velocity from IMU data that there was an over-performance of the spring-based separation system (delta absolute axial velocity of 335 mm s−1 vs. 320 mm s−1, corresponding to a 5.4 \(\sigma \) value), while the roll rate was slightly lower than expected (2.617 rpm vs. 2.70 rpm, a 1.5 \(\sigma \) value). Nevertheless the estimated Entry Interface Point (EIP) and landing point (Fig. 18) were well within the predictions.

Fig. 18
figure 18

Schiaparelli estimated landing point w.r.t. target, before and after separation from TGO. Entry was prograde, from west to east

Full size image

Nominal entry of Schiaparelli into its 3-day hibernation occurred 15 min after separation, with all on-board systems having operated nominally.

Wake-up from hibernation mode occurred nominally on 19 October 2016, 75 min before EIP. All on-board systems operated nominally when exiting the hibernation mode, and the thermal environment inside Schiaparelli during the hibernation phase was fully in line with expectations.

See Portigliotti et al. (2017) and Bonetti et al. (2018) for a post-flight analysis of the Schiaparelli coasting, entry and descent, and Pinaud et al. (2019) and Gülhan et al. (2019) for post-flight analyses of data from the heat shield sensors and COMARS+, respectively.

As is now well known, an anomaly occurred due to a mismatch between the IMU angular rate saturation behaviour at parachute deployment, and how such saturation was handled by the GNC algorithm once data from the RDA was brought into the control loop. This caused premature separation of the Surface Platform from the Back Shell, a short RCS propulsive phase (∼3 s priming phase to reduce the angular rates, vs. the expected nominal design duration of 28.5 s), and transition to the touchdown phase (including activation of DREAMS, as shown by receipt on ground of a first housekeeping packet from the experiment). Although sadly neither DeCa nor DREAMS were able to return science data, much was learned from the real-time telemetry about the functioning of Schiaparelli and the measurements made during entry and descent for analysis by the COMARS+ and AMELIA teams.

With respect to the nominal prediction of 2.8 Mbit EDL essential data transmitted, once the blackout duration (about 72 s) and shorter-than-nominal descent phase are taken into account, the telemetry actually received on ground amounts to some 2/3 of the prediction. The total amount of data expected in the nominal mission (for the full EDL data set plus science and engineering data from the coasting, EDL and nominal surface phases) was 207 Mbit.

See Tolker-Nielsen (2017) for the report of the anomaly enquiry.