Sample Fetch Rover and Sample Recovery Helicopters

NASA/JPL-Caltech/AeroVironment. Public domain (NASA / US government work).

Overview

Two machines occupied the same role in successive Mars Sample Return architectures: the Sample Fetch Rover, an ESA and ASI funded rover developed by Airbus Defense and Space at Stevenage, and the pair of Sample Recovery Helicopters that replaced it in 2022 [2].

The Sample Fetch Rover was to launch on the Sample Retrieval Lander in 2028, arrive on the surface in mid 2030, egress from the platform, traverse to the depot where the Mars 2020 rover had cached sealed sample tubes, collect and store the tubes onboard, and carry them back to the Mars Ascent Vehicle [14]. It was designed for autonomous operation and fast mobility on Mars terrain. The mission was canceled before its thermal qualification testing was carried out.

The replacement decision was taken in 2022: given Ingenuity’s flight record and Perseverance’s health, two Ingenuity-sized helicopters were substituted for the European rover, saving mission mass and cost [2]. In that architecture Perseverance is the primary means of returning tubes to the lander and the two helicopters are backups, used if Perseverance cannot complete an extended mission. Both were subsequently overtaken by further MSR architecture changes; neither vehicle flew.

Specifications

Sample Recovery Helicopter, as designed in 2023 [2].

Parameter	Value
Configuration	coaxial, two counter-rotating rotors
Rotor radius	0.6 to 0.7 m trade range; baseline updated to 0.7 m in March 2023
Rotor solidity, 0.7 m blade	0.128, thrust-weighted chord
Peak hover figure of merit	0.48 at tip Mach 0.7, blade loading 0.13
Hover efficiency limit	reduced above tip Mach 0.75
Rotor speed range in test	2043 to 2500 RPM
Total mass	about 2.5 kg, against 1.8 kg for Ingenuity
Ground mobility	tank-like steering, four lightweight metal or composite wheels
Manipulation arm target mass	under 100 g
Sample tube handling capacity	up to 150 g per tube
Structure	carbon fiber mast, minimal payload frame, carbon fiber legs with Ingenuity leg suspension and deployment hinges
Power	solar panel for recharging after separation from the lander, high-capacity batteries
Flight computer	newer-model Snapdragon, replacing Ingenuity’s 801
Radio	improved unit based on the Ingenuity SiFlex design
Servos added over Ingenuity	8
Vehicles built as a set	2, identical, for redundancy

Values from [2].

Sample Fetch Rover parameters published are limited to the thermal design and the autonomy requirements; mass, dimensions, wheel count and power are not published.

Parameter	Value
Heating	electrical only, no radioisotope heater units
Night environment design point	-130 C
Primary heat leak path	the CO2 atmosphere
Insulation approach	gas-gap insulation around hot zones, gap sized below the convection onset

Thermal design from [14].

Mission profile

Parameter	Value	Source
SFR launch	2028 on the Sample Retrieval Lander	[14]
SFR arrival	mid 2030
Sample tubes to be collected	up to 35	[13]
SFR surface duration	150 sols
SFR traverse	up to 20 km in the first 150 sols
SRH deployment	from the top of the Sample Retrieval Lander	[2]
SRH role	backup to Perseverance for depot retrieval

Tube positions were to be tagged in orbital imagery, but that tagging is not precise enough to drive a final approach, so the rover had to localize each tube visually from its own cameras [13].

Sample Fetch Rover autonomy

The published autonomy work on the Sample Fetch Rover concentrates on the step the mission depended on: finding a 150 g metal tube lying on regolith and estimating its pose well enough to grasp it [13]. The proposed architecture detects tubes with a deep neural network trained by transfer learning from a synthetic dataset generated from photorealistic 3D simulations of Martian scenes, then estimates tube pose from the detected region using contour detection and line fitting [13]. Hardware acceleration was part of the proposal, because the onboard compute available to a rover of this class limits the solutions that can be considered. Laboratory validation used the ExoMars Testing Rover on a Mars-analog terrain with its LocCam and NavCam.

Sample Fetch Rover thermal design

The rover carried no radioisotope heater units and relied entirely on electrical heating to hold internal temperatures against a -130 C night, which makes minimizing heat leak the governing thermal requirement [14]. The dominant leak is conduction and convection through the CO2 atmosphere, so an efficient gas insulation design around the hot zones was required. Airbus reviewed the gas insulation used on previous Mars surface missions to set the design parameters, principally the maximum gap size that suppresses convection, and produced several candidate designs plus a test to determine empirically which insulates best. The comparison base includes the Mars Science Laboratory, the first Mars rover to use a gas gap as its primary insulation with a 25.4 mm CO2 gap around the internal equipment, and the Rosalind Franklin rover, which uses a baffle network across its external structure to create a 30 mm CO2 gap, adopted after a chamber test campaign of representative geometries in CO2 [14]. The Sample Fetch Rover was canceled before the test was conducted.

Sample Recovery Helicopter design

The helicopters keep Ingenuity’s architecture where they can, to reduce risk: two counter-rotating coaxial rotors, a solar panel for recharging after separation from the lander, cameras for navigation and imaging, a carbon fiber mast and minimal payload frame, and carbon fiber legs using the Ingenuity leg suspension and deployment hinges [2]. The coaxial layout was also what allowed them to fit the compact accommodation pockets available on the lander. Two identical vehicles are carried because a coaxial rotor system is a single failure point: unlike a multirotor, a failure anywhere in the rotor system grounds the aircraft, so redundancy is provided at vehicle level rather than at rotor level. The changes from Ingenuity follow from the new role. A ground mobility system with tank-like steering and four lightweight metal or composite wheels gives precise approach to a tube and precise delivery to the lander without consuming a flight. A sample manipulation arm, specified to grasp tubes of up to 150 g while itself massing under 100 g, gives the vehicle the ability to interact with the surface rather than observe it, and would be the smallest robotic arm flown on Mars [2]. Fitting a precise manipulator into that mass led to proposals such as deriving the arm servos from the rotor servos. Higher-capacity batteries, a larger solar array, eight additional servos, an additional camera and software extended to in-flight absolute localization and ground operations complete the change set. Two Ingenuity parts were unavailable in the condition and quantity required and had to be replaced: the SiFlex radio, superseded by an improved unit based on the same design, and the 801 Snapdragon flight computer, superseded by a newer model [2]. The result is a 2.5 kg vehicle against Ingenuity’s 1.8 kg, an increase of about 40 percent constrained by lander accommodation and by the wish to retain design heritage [2].

Rotor development

The mass growth had to be paid for in rotor performance. CAMRAD II was used to investigate rotor size, through blade chord or radius increase, and rotor shape through planform and twist, optimizing hover performance for a minimum-power design at each attainable blade loading for each candidate solidity [2]. The outcome was a risk-reduction rotor with increased blade radius, a similar dimensional chord, an updated planform and a twist distribution very close to Ingenuity’s, which lowers power required at a given thrust and raises the blade loading margin relative to reusing the Ingenuity rotor. Keeping the chord while increasing radius also avoids designing a blade from scratch and leads to smaller motors and less battery capacity, at the cost of attention to blade stiffness. The baseline was updated from the 0.6 m Ingenuity rotor to the 0.7 m optimized rotor in March 2023, giving greater capability and control margin at lower power [2]. The rotor performance database supporting that decision came from the 25-ft Space Simulator campaigns [3].

Rotor testing was done at NASA JPL 25-ft Space Simulator in three campaigns: the Ingenuity Engineering Design Model 1 with and without a cruciform box, a Transonic Rotor Test rig using the same blade geometry at much higher rotational speeds, and a Sample Recovery Helicopter Dual Rotor Test [3]. The cruciform box reproduces the Lift-off Adapter and Inverted Retention box, the takeoff platform on the lander, so the campaign also collected rotor-on-box interaction data. Measured performance of the Ingenuity blade geometry at high rotor speeds gives a peak figure of merit of 0.48 at tip Mach 0.7 and blade loading 0.13, with hovering efficiency reduced above tip Mach 0.75 and an abrupt fall at higher blade loading indicating the onset of blade loading stall [2].

Deployment from the lander

Ingenuity deployed from the underside of Perseverance and took off from the ground; the recovery helicopters deploy from the top of the lander and lift off from the Lift-off Adapter and Inverted Retention box, which introduces takeoff conditions that had no flight precedent [2]. Two problems were analyzed. Rotor outwash recirculating over the box can build with time, and if allowed to build too long it can alter flight dynamics and reduce the clean air available to the rotor; mid-fidelity RotCFD runs modeling the coaxial rotors as actuator disks show 15 m/s outwash developing over 8 to 20 seconds of a rapid collective ramp [2]. Separately, it must be shown that the blades cannot strike the lander at any point in the takeoff sequence. Experimental work in the 25-ft Space Simulator was to be combined with further computational fluid dynamics to close the takeoff flight dynamics case [3].

Technologies developed

The Sample Fetch Rover leaves a thermal result and an autonomy result. The thermal result is a documented review of Mars gas-gap insulation practice from Pathfinder through Curiosity and Rosalind Franklin, assembled to set gap sizing against convection onset for a rover carrying no radioisotope heaters [14]. The autonomy result is a tube detection and pose estimation pipeline trained on synthetic photorealistic Martian imagery and validated on a physical rover testbed, which addresses the general problem of acquiring a small known object whose orbital position fix is too coarse for a final approach [13].

The Sample Recovery Helicopter is the first Mars rotorcraft design to carry ground mobility and a manipulator, which converts the vehicle class from an observer into a retrieval machine [2]. The specific engineering products are the sub-100 g manipulator concept driven by rotorcraft mass budgets, the coaxial rotor optimization that buys a 40 percent mass increase through a radius increase at constant chord, and the characterization of lander-deck takeoff, where rotor outwash recirculation over the deployment box is a flight dynamics problem rather than a nuisance [3].