Rosalind Franklin

ESA/S. Corvaja.

Overview

Rosalind Franklin is the ESA ExoMars rover, a solar-powered six-wheeled vehicle carrying a 2 m rotary drill and an internal analytical laboratory [1]. Rover mass is about 310 kg, of which 26 kg is instrument payload, a figure that excludes the drill and the sample processing mechanisms [1]. The vehicle is the only planetary surface machine designed to acquire cores from 2 m depth and deliver them, crushed, to onboard spectrometers and a mass spectrometer without exposing the material to the ambient environment [12], [13]. The 2004 Phase A concept run for ESA by an MDA-led team was a 240 kg rover with a different chassis and is not the flown configuration [3].

Launch was suspended in 2022 when cooperation with Roscosmos ended, removing the Russian descent module [10]. ESA is building a European entry, descent and landing module reusing the qualified onboard computer, the radar Doppler altimeter and the parachute system, with NASA providing the launch service, elements of the landing propulsion and radioisotope heater units for the rover [10].

Specifications

Parameter	Value	Source
Mass	about 310 kg	[1]
Instrument payload mass	26 kg, excluding drill and sample handling
Locomotion formula	6 x 6 x 6 + 6 (supported, driven, steered, deployed)
Suspension	triple bogie, no central differential
Wheel diameter	28.5 cm without grousers
Wheel width	12.0 cm
Wheel type	flexible, grousered
Average ground pressure	about 10 kPa
Steering modes	point turn, double Ackermann, crab
Wheel-walking gait	available on all six deployment drives	[1], [2]
Drill depth rating	0 to 2 m	[12], [4], [5]
Drill type	rotary, no percussion	[1]
Core sample size	about 3 cm long by 1 cm diameter	[12]
Rock strength demonstrated	up to 150 MPa unconfined compressive strength	[1]
Crushed particle size	few to about 500 um, median 250 um
MOMA single-use ovens	32	[6]
Power source	solar array	[1], [11]

Mission profile

Parameter	Value	Source
Earliest launch	October 2028	[10]
Transfer	two years, landing 2030
Launch service	NASA
Landing site	Oxia Planum	[7]
Nominal surface lifetime	218 sols, about 7 Earth months	[1]
Reference surface mission traverse	about 1.5 km
Experiment cycles	6
Vertical surveys	2
Samples per experiment cycle	2, one surface and one from 1.5 m
Vertical survey sample depths	0, 50, 100, 150, 200 cm
Assumed traverse rate	about 100 m per sol
Assumed drilling rate	50 cm per sol
Communication sessions	2 per sol via Trace Gas Orbiter

The reference surface mission is a contractual object as well as a science plan: ESA specified that industry deliver a rover able to execute it within the nominal mission duration, and demonstration by simulation and test is a requirement [1]. It begins with egress in either of two directions, a 10 sol commissioning phase near the platform, and a move of about 60 m out of the descent engine blast contamination zone before the drill and the analytical laboratory drawer may be opened. The first laboratory operation is a drill blank run for calibration and organic cleanliness assessment, and only after it completes does biosignature search begin. Experiment cycle spacing is assumed to grow in 100 m steps, 100 m to the first site, a further 200 m to the second, for a total of about 1.5 km [1]. The drill requirement set counts 7 experiment cycles and at least 2 vertical surveys to 2 m with four sample acquisitions each, a minimum of 17 samples [12].

Oxia Planum was selected for a wide exposure of Noachian-age Fe/Mg phyllosilicates overlain by a younger dark resistant unit, with the clay-bearing material accessible to a 2 m drill [7].

Mobility

The kinematic configuration is a six-wheel triple bogie with locomotion formula 6 x 6 x 6 + 6: six supporting wheels, six driven, six steered, plus six articulated deployment or knee drives [1]. The three bogies passively adapt to rough terrain and provide platform stability without a central differential. Available maneuvers are straight drive, turn on the spot, double-Ackermann steering and diagonal crabbing, the last used to translate sideways along an outcrop for imaging. Lander accommodation constrained the wheels to 28.5 cm diameter without grousers and 12.0 cm width [1]. Flexible wheels were adopted to offset the traction penalty of small diameter: their deformation enlarges the contact patch, brings average ground pressure to about 10 kPa, and absorbs impact loads [1]. That figure is above the 5.75 kPa average computed for the MER and MSL wheels, both of which nonetheless embedded in unconsolidated material, at Purgatory Ripple and in megaripple deposits respectively.

Wheel-walking is the mitigation for that margin. It is a coordinated rototranslational gait that drives the wheel motors and the deployment joints together, so each wheel train is advanced by the knee joint while the wheel is turned, producing forward motion by a different force path than pure rolling [2]. Tests reported for the ExoMars configuration showed improved dynamic stability during egress, better traction on loose soil where normal driving slips, and increased slope gradeability [1]. Simulation of the ExoMars 6 x 6 x 6 configuration gave 25 degrees climbed with wheel-walking against 18 degrees with normal driving. The cost is power: wheel-walking is less efficient in energy per meter than rolling, which makes locomotion mode selection a decision problem rather than a default. Slip ratio, the standard input to that decision, cannot be estimated during wheel-walking by the usual wheel-odometry difference because the wheels are not rolling continuously, and an estimator based on the acceleration of the wheel train was developed for this case.

The 2004 Phase A locomotion requirement set, for a 240 kg concept rover, specified uphill, downhill and crosshill driving on loose sandy 25 degree slopes, 0.3 m obstacle height, and at least 100 m/h maximum speed for short-duration recovery situations, over sampling sites 0.5 to 2 km apart [3]. Those are requirements on an earlier design, not measured capabilities of the flight vehicle.

Force-torque sensors at the wheel drives have been evaluated on ESA planetary rover testbeds as an input to traction control and terrain assessment, using the ExoMars Testing Rover, a reduced-scale platform with the same actuated-suspension kinematics [9].

Power and energy

The rover generates electrical power from a deployable solar array and carries batteries and heater units sized to survive the Martian night [11]. Regional mobility of several kilometers is stated to rely on solar array electrical power [1]. Array area, peak generation in watts and battery capacity in watt-hours are not published. NASA is supplying radioisotope heater units for the rebuilt mission [10].

Thermal

Preservation of organic and volatile content sets the thermal requirements on the sample path rather than component survival alone. The drill carries thermocouples close to the tip. Mars chamber tests through simulated stratigraphic columns containing ice lenses of 0 to 35 percent water content showed drilling-induced temperature rises of order 20 C in continuous cutting, reduced to 5 C or less using a variable cutting law that alternates short cutting periods with dwell time for thermal dissipation, at the cost of time [1]. Subsurface material at mid-latitudes oscillates between -30 and -80 C at 0.5 m depth and averages about -50 C deeper, so the drilling strategy can be set to keep the sample cold throughout acquisition. Crushing is scheduled very early in the morning, when the analytical laboratory drawer is at its coldest, to preserve the organic and volatile fractions, and the crushing station temperature is monitored before and throughout the operation. Drill string overnight parking was tested in a Mars atmospheric chamber at -110 C: the tool was driven to full penetration, stopped, left in the simulated subsurface overnight, and restarted and extracted the next morning.

Compute and avionics

Processor part, memory and data handling architecture for the rover are not published. The onboard computer of the descent module is identified only as a qualified item reused from the suspended configuration [10]. The rover computer performs onboard analysis of MicrOmega hyperspectral cubes, examining absorption bands per pixel to identify mineral grains and assign them as targets for Raman and MOMA laser desorption observations, which places band-fitting over a 250 x 256 pixel by 320 spectral step cube inside the rover’s own compute budget [1].

Autonomy

Ground control designates targets from compressed stereo imagery returned by the mast cameras, and the rover computes its own navigation solution over the resulting digital terrain maps, with inclinometers, gyroscopes and Sun sensors for attitude and heading [11]. The design travel figure quoted by the operator for autonomous driving is about 100 m per sol, matching the traverse rate assumed in the reference surface mission [1]. Onboard science autonomy is limited but real: MicrOmega cube analysis on the rover computer selects grain targets for the two spectrometers without a ground cycle.

Locomotion mode selection is a further decision the vehicle class exposes. Rolling and wheel-walking differ in traction and in energy per meter, so choosing between them needs an onboard estimate of slip and rolling resistance, both derivable from proprioceptive sensing [2]. Force-torque sensing at the drive units has been characterized in field trials on ESA rover testbeds as an input to that estimate [9].

Communications

The nominal link is a UHF proximity relay through the ExoMars Trace Gas Orbiter, which carries the NASA-supplied Electra UHF package for orbiter-to-surface proximity links [1]. Operations assume two sessions per sol, a morning uplink of the sol’s activities and an evening downlink of results, so all data needed to plan the next sol must reach the ground with the evening pass. TGO’s orbit is not Sun-synchronous, so session local time drifts about 30 minutes earlier per sol for three consecutive sols and then jumps about 2 hours forward, and overpass duration and data volume vary with the relative geometry [1]. Both effects constrain strategic planning: the required tasks cannot always be completed in time for the next pass, and the response is either to tailor activities to the time available or to skip a session. Infrequent communication windows are the stated reason for the rover’s autonomy level [11].

Payload and instruments

Nine Pasteur payload instruments are carried, split between mast-mounted remote sensing, subsurface sounding, drill-integrated sensing and the analytical laboratory drawer [1]. The combination is organized around the retrieval and analysis of samples from 0 to 2 m depth, which is the element that separates this payload from earlier Mars surface suites [4].

Instrument	Function	Key parameters	Source
PanCam	multispectral stereo panoramic and high-resolution imaging	two wide-angle cameras, 38.3 degree field of view, 11-position filter wheels, 150 mm stereo baseline; high resolution camera 4.8 degree field of view	[4]
ISEM	mast IR point spectrometer, coaligned with the PanCam high-resolution camera	records IR spectra of reflected solar light	[1]
WISDOM	ground-penetrating radar for subsurface structure and drill site selection	step-frequency, 0.5 to 3.0 GHz, 3 to 10 m depth range depending on regolith dielectric properties, vertical resolution to 3 cm	[5]
ADRON	neutron detector for subsurface hydrogen	supports drill site selection	[1]
CLUPI	close-up imager on the drill box	variable focus, images outcrops, drill fines and acquired cores
Ma_MISS	IR spectrometer integrated in the drill tool	images the borehole wall as drilling proceeds, one image column per rotational step of the tool
MicrOmega	near-IR hyperspectral microscope on crushed sample	250 x 256 pixels by 320 spectral steps, 0.95 to 3.65 um, 20 cm-1 sampling, 20 um per pixel
RLS	Raman spectrometer on crushed sample	532 nm continuous excitation, 50 um spot, about 150 to 3800 cm-1 shift, about 6 cm-1 resolution below 2000 cm-1
MOMA	mass spectrometry of organics, two modes	11.5 kg without margin; 32 single-use ovens 6 mm in diameter; pyrolysis to 850 C, derivatization to 600 C; 266 nm UV laser, 1 to 2 ns pulses; gas chromatograph subsystem 1.6 kg	[6]

MOMA operates either as gas chromatography mass spectrometry, in which powder is dosed into one single-use oven, sealed and heated stepwise, optionally with a derivatization agent, or as laser desorption mass spectrometry, in which the pulsed UV laser fires directly on powder held in the refillable container [1], [6]. The mass spectrometer is shared between the two modes and its low mass cutoff is set at about m/z 40.

WISDOM’s depth range spans the 2 m drill reach with margin, which is what makes it the site selection instrument for drilling rather than a standalone sounder [5].

Drill and sample chain

The drill assembly comprises a drill tool about 70 cm long carrying the sample acquisition device with shutter, movable piston, position and temperature sensors, and the Ma_MISS front optics of sapphire window, IR lamp, reflector and optical fiber; three extension rods of 50 cm each carrying optical and electrical contacts that route Ma_MISS signals to the spectrometer in the upper drill unit; a backup drill tool without spectrometer; and the rotation-translation group of sliding carriage motors, guides and sensors [12]. Operationally the drill box lies horizontally across the rover’s front face during traverse, is raised, rotated counterclockwise and lowered vertically to drill, then elevated, rotated clockwise and inclined to deliver the core to the laboratory inlet port [1]. The same positioning dexterity is used to aim CLUPI.

Sample transfer starts with the core sample transport mechanism, a hand that extends through a door in the rover’s front panel to receive the core [1]. Imaging by the PanCam high-resolution camera and CLUPI is limited to a window of a few minutes set by a sample contamination analysis of external rover sources. The mechanism then retracts into the analytical laboratory drawer, whose entire sample path is enclosed in an ultra clean zone sealed at positive pressure until opened on Mars. A jaw crusher produces particulate with an approximately Gaussian size distribution from a few micrometers to about 500 um, median 250 um [13]. A blank dispenser can feed individual blank samples to the crusher for contamination verification. Powder falls into one of two redundant dosing stations, both piezovibrated to improve granular flow, which meter material either onto a refillable container, a flat tray presented to the laboratory instruments, or into a single-use oven; a rotating carousel positions container and ovens under the dosing station. Two further mechanisms serve the refillable container, one flattening the powder to the correct height for the instruments and one emptying it for reuse. Off-nominal handling is built into the mechanisms. A spring-actuated hammer can shock the crusher’s fixed jaw to clear adhering material, a dedicated actuator can open the jaws to evacuate a jammed sample, both dosing stations can be bypassed, and an alternative transport container can drop the whole crushed charge at once onto the refillable container or into an oven without quantity control. If the drill string jams, counter rotation can disengage it at the last blocked element and recover the upper portion, with further drilling continued using the backup tool and any remaining rods; the positioning system carries an emergency ejection unit as a last resort, after which no further samples can be delivered.

Modes of operation

Published sol-level operating states follow the reference surface mission structure rather than a named mode set [1]. An experiment cycle runs seven sols: approach from about 20 m with panoramic imaging on sol 1; approach to 3 m and WISDOM and ADRON subsurface sounding on sol 2; close-range imaging of lithology and texture on sol 3; sample acquisition and delivery to the laboratory on sol 4; crushing followed by MicrOmega, RLS and MOMA laser desorption first-look analysis on sol 5; further analysis on sol 6; and a subsurface scanning and drilling sequence on sol 7. A vertical survey adds drilling in 50 cm increments over three sols to reach 1.5 m, then repeats the laboratory sequence [1]. Drill string handling has a planned progression of its own: initially all segments are disassembled and stored at the end of each sol, then the tip is left in the borehole to measure the torque needed to free it in the morning, and progressively longer sections are left downhole as confidence is established, the objective being to minimize dead assembly time and give Ma_MISS more borehole observing time. Instrument-level mode selection is constrained by consumables: MOMA ovens and carrier gas are finite and are conserved where a laser desorption run can answer the question instead [6].

Ground operations

The rover is operated from the Rover Operations Control Centre, on a tactical cycle bounded by the two TGO passes per sol, with operations on Mars time at least for the first months [1]. Planning products for the next sol must be derived from data arriving on the evening pass. The operational chain has been rehearsed in ExoMars-like Field Testing trials: in the Atacama Desert trial a rover emulator named Charlie was driven by a local control center near the Paranal Observatory while a remote control center in the United Kingdom planned each simulated sol using only the data the rover returned [8]. Instrument teams ran the sample analysis loop with a portable Raman technology demonstrator and, in the laboratory, with a Raman simulator coupled to a replica of the sample preparation and distribution system running the same multi-point analysis algorithms as the flight instrument.

Technologies developed

The 2 m rotary drill with in-tool spectroscopy is the program’s distinctive product: a string of one tool and three rods carrying optical and electrical contacts through every joint, so that a borehole-wall imaging spectrometer works at full depth [12], [1]. The sample chain contributes a sealed ultra clean zone spanning crushing, dosing and presentation, with blank injection as a routine contamination check rather than a pre-launch verification [13]. Wheel-walking on a triple-bogie chassis is the locomotion contribution, together with the associated body of work on locomotion mode selection and on slip estimation while walking, which is applicable to any actively articulated suspension [2]. The flexible wheel, adopted because lander accommodation capped wheel diameter, is a distinct approach to the small-wheel ground pressure problem. Force-torque sensing at the drive units, evaluated on ESA rover testbeds, follows from the same actuated-suspension architecture [9].