Chang'e-6 Sampling System

Overview

A far-side sample return has no direct-to-Earth link and a surface stay measured in hours, so a sampler built for that combination has to decide, on its own, where to dig and whether its own arm actually did what it was told. Chang’e-6 is that case: it reused the Chang’e-5 architecture, a rotary-percussive drill for a subsurface core plus a four-joint arm carrying a scooping sampler, but replaced Chang’e-5’s manual per-point confirmation and specialist pose fine-tuning with a vision pipeline that selects its own sample points and corrects its own manipulator pose [1], [7].

Autonomy of that kind is required, not merely convenient, when three conditions hold together: the environment or the spacecraft changes, the change is not predictable in advance, and the needed response is faster than the next communication cycle can supply [2]. Chang’e-6’s sampling window was about 48 hours between landing on 2 June and ascent on 4 June [11], against Chang’e-5’s 3.12 h of drilling plus 15.82 h of surface sampling on the near side [7], and the far-side location added a relay hop on top of that. Manual per-point analysis and manual pose fine-tuning both need several image-command round trips per sample point, and it is the long relay delay, not operator skill, that is identified as what made manual operation too slow for this budget.

Specifications

Parameter	Value	Source
Sampling mechanisms	rotary-percussive coring drill and a four-joint scooping arm	[1], [7]
Arm joints	4, sampler attached at joint 4	[1]
Arm reachable region	sector of 2.8 m radius and 120 degrees
Sampler attitude control	pitch adjustable, roll not adjustable
Sampler sensing	touching disc with force sensor, front shovel, narrow-field close-up camera
Binocular camera image size	2352 x 1728 pixels
Binocular baseline and focal length	200 mm, 15.4 mm
Binocular working imaging distance	2 to 5 m
Expected 3D reconstruction error	5 mm at 2.5 m
Sampler pose measurement precision	1.1 mm and 0.4 degrees
Sampler pose measurement time	1.6 s average

Mission profile

Parameter	Value	Source
Relay satellite	Queqiao-2, launched 20 March 2024 on a Long March 8	[6]
Launch	3 May 2024, Long March 5 from Wenchang	[11]
Circumlunar orbit insertion	8 May 2024	[11]
Lander-ascender separation	30 May 2024	[11]
Landing	2 June 2024, South Pole-Aitken basin	[11]
Landing point	approximately 154.0 W, 41.6 S, southern Apollo basin	[1]
Sampling	2 to 3 June 2024, drill and robotic arm	[11]
Ascent	4 June 2024	[11]
Rendezvous and sample transfer	6 June 2024	[11]
Earth return	25 June 2024, Siziwang Banner, Inner Mongolia	[11]
Surface window available for sampling	about 48 hours	[11]
Returned sample mass	1935.3 g	[12]

The program timeline page places sampling on 2 and 3 June and ascent on 4 June [11], which brackets the roughly 48 hour surface window the autonomy was designed against. The page states neither a sample mass nor landing coordinates; the 1935.3 g figure comes from the separate CNSA handover announcement [12], and the coordinates from the vision sampling paper.

Mobility

Not mobile. The mechanical arm has four joints with the sampler attached at joint 4, its reachable region on the terrain is approximately a sector of radius 2.8 m and 120 degrees, and it can adjust pitch but not roll [1]. Scooping allows more flexible site selection than drilling because the sampler can reach any point within that sector, whereas the drill samples only the column beneath it, 1.68 m of travel for about 1 m of core on Chang’e-5 [7]. A slope component in the roll direction under the shovel cannot be corrected by the arm and directly degrades control of scooping depth, so it is avoided by choosing where to sample, which is what the roll term in the site selection loss function below does.

Power and energy

No power figures are published for the sampling system. The lander operated for about two days inside a single lunar day [11] and was never required to survive a lunar night. The Chang’e-5 drilling chain, on which the Chang’e-6 drill is based, has a maximum single-machine power above 800 W [7].

Thermal

No thermal design is published for the Chang’e-6 sampling system. The Chang’e-5 drill, on which it is based, uses the helical blades of its rotating outer pipe both to discharge cuttings and to dissipate drilling heat, there being no convective path out of a borehole in vacuum [7]. At 45 degrees latitude, the site’s approximate band, an analytic fit to Diviner data gives a mean surface temperature of about 350 K at local noon falling to about 89 K before sunrise [4], compared with a NASA design envelope of 391 K to 96 K at the equator, where daytime insolation is highest [11].

Compute and avionics

The sampling system comprises the four-degree-of-freedom mechanical arm, the sampler, and a binocular camera. The sampler itself consists of a touching disc with a force sensor at its base, a front-end shovel, and a close-up camera used to monitor sampling and inspect the collected soil. The binocular camera, the primary sensor for the autonomy, returns 2352 by 1728 pixel images from a 200 mm baseline, a 15.4 mm focal length and a 2 to 5 m working distance, chosen against the arm’s 2.8 m by 120 degree reach to cover the whole sampling sector while achieving the expected 3D reconstruction error of 5 mm at 2.5 m [1].

The processing chain is CREStereo for stereo matching and depth estimation, the Point Cloud Library for point cloud processing, YOLOv5 to extract the local region of the sampler and so reduce reconstruction cost, SIFT and FPFH as feature descriptors, and TEASER followed by ICP for point cloud registration [1]. The source does not state where each stage executed, so the split between onboard and ground computation is not established here. Autonomy that must run inside a single communication cycle has to be resident on the spacecraft; where changes can be predicted and modeled instead, ground operators prescribe the actions and no autonomy is needed [2].

Autonomy

Every step of the sampling implementation is autonomous, with the operator only monitoring and making a simple decision if required [1]. Space autonomy is traded against non-autonomous approaches on risk and cost, and mission objectives are commonly scaled back toward non-autonomous practice where possible; autonomy becomes a requirement only where the response time is shorter than the next communication cycle [2].

Sample point selection

Terrain point clouds reconstructed from the binocular pairs are processed automatically to produce a slope map and to segment obstacles and pits, classifying points as collectable area, obstacle or unreachable, and pit, with obstacles below a size threshold classified instead as collectible rocks; candidate points are then scored by a multi-objective loss function over four factors, weighted equally for Chang’e-6, and for Chang’e-6 this produced ten sample points over a region whose average slope was 5.1 degrees, with a neighborhood radius parameter of 400 mm, of which points 1 to 3 and their neighboring areas were scooped [1]. The four factors are local slope, proximity to obstacles and pits, distance from the limits of the arm’s reachable region, and specifically the slope component in the roll direction, which is penalized because the arm cannot compensate it; a minimum point-spacing constraint prevents selected points from crowding, and simulated annealing is used to select the set with the lowest loss.

The landing zone within which this operates was itself selected against an 8 degree slope limit, and has an average slope of about 5.74 degrees, with 76 percent of its area below that maximum for safe landing, and contains 26,785 identified craters larger than 50 m, more than 96 percent of them between 100 m and 1 km across [9]. The three flat regions favored by engineering constraints have average slopes of 2.36 to 2.37 degrees at elevations of -5,197 and -5,277 m, with craters occupying 8.61 and 10.36 percent of their areas [9].

Pose measurement and fine-tuning

The arm cannot reach a commanded pose accurately. Flexible deformation and secondary effects produce end-localization errors of several centimeters, measured on Earth as about 30 mm and 2 degrees of absolute error even with a balloon used to compensate the gravity difference, and these errors are difficult to model and directly affect scooping depth [1]. Chang’e-6 measures the sampler’s actual pose visually instead: the sampler point cloud is reconstructed from a binocular pair, its region of interest located with YOLOv5, features matched against a template point cloud, and the transformation matrix estimated by TEASER followed by ICP, from which the sampler’s x, y, z and pitch angle follow, with a measured precision of 1.1 mm and 0.4 degrees, average relative distance error of 3.4 mm over movements of 300 to 450 mm and 2.0 mm over the 100 to 200 mm fine-tuning range, average measurement time 1.6 s, and correct registration with less than 30 percent of the valid region visible [1]. Because occlusion and specular highlights make the reconstructed region much smaller than the template, the method replaces the template with the current reconstruction once a correct registration is confirmed, so subsequent registrations are made against like data.

Two systematic errors are then compensated explicitly [1]:

• Soil sinking. When the touching disc contacts the surface it squeezes a small pit, so the sampler sits lower than the intended touching pose. The offset is recovered as the difference between the calculated and the vision-measured touching height, and ranges from 5 to 20 mm depending on soil density in terrestrial experiments [1]. Regolith in place is denser than a loose terrestrial simulant at the same depth, which is one reason a terrestrial calibration of this offset cannot be taken as the flight value outright [3].
• Support force. The force sensor has an activation threshold, so a support force is already acting on the touching disc when motion stops, altering the arm’s deformation. This cannot be seen in the touching pose alone, so it is recovered by comparing the displacement the arm reports from its four joint angles against the displacement the vision system measures between a position above the sample point and the touching pose.

The final target height is the touching height plus a lifting height set by the target scooping depth and the sampler geometry, plus both offsets, with an alternative direct formulation used for points with no available imagery or outside the binocular camera’s field of view [1]. Fine-tuning achieved an average scooping depth error of 3.8 mm in terrestrial experiments against a 22.0 mm target depth, and the failure modes are asymmetric: too high and the shovel does not touch the soil and collects nothing, too low and it takes excessive soil and can become stuck, which is a hazard [1]. In-situ lunar soil relative density is about 65 percent in the top 15 cm [3], which is the range the touching disc squeezes into, and where terrain was rough and points were widely spaced the per-point offsets differed from their mean by 2 to 5 mm, against under 3 mm on flat, closely spaced terrain.

Communications

Far-side operation means no direct-to-Earth path. All Chang’e-6 surface communications were relayed by the Queqiao-2 satellite, launched on a Long March 8 from Wenchang on 20 March 2024 specifically to support Chang’e-6, Chang’e-7 and Chang’e-8 [6]. A relay adds latency and restricts communication to windows when the satellite is in view of both lander and Earth. Chang’e-4 had already demonstrated autonomous terrain-relative navigation with hazard assessment and avoidance for far-side landing without radiometric data [7], and the same communication constraint that forced that navigation autonomy is what drove Chang’e-6’s sampling autonomy. Link parameters for Queqiao-2 are not published here.

Payload and instruments

The sampling hardware is the drill and the arm-mounted scooping sampler, the same two methods used on Chang’e-5: soil core drilling and soil and rock fragment scooping [1], [7]. Compared with drilling, scooping allows more flexible site selection because the sampler can reach any point within the arm’s sector.

The sampler assembly carries the touching disc with its force sensor, the front-end shovel that collects and transfers material to the container, and the close-up camera, and derives from the Chang’e-5 arm, which integrates shovelling, digging, suction, receiving and grasping, with repeated positioning accuracy better than 1 mm [1], [7].

Instrument-level parameters for the Chang’e-6 drill are not published in the sources consulted here; the mechanism is stated to be the same approach as Chang’e-5’s dual-pipe soft-bag corer [1], [7]. A separate Chinese concept, ICP-Coring, proposes preserving volatiles in a returned core with a sealed liner rather than a soft bag, but it remains a design study with no integrated coring test [8].

A Chinese national flag carried by the lander was unfurled after sampling, the first flag deployed on the lunar far side [11].

Modes of operation

The sequence is fixed rather than mode-driven, and closely follows Chang’e-5’s: unlock the drill and scoop after landing, drill and core and transfer the core to its primary package device, scoop surface samples and transfer them to their own package device, then move the package to the sealing capsule on the ascender and seal it with drilled and surface samples kept separate [7]. Within scooping, each sample point runs the same cycle: determine the target sampling pose and the corresponding touching-moon pose from the terrain analysis; move the arm to an approximate position, deliberately holding a large height margin during the midway movement for safety; touch down and measure; compute the sink and support-force offsets; fine-tune; and scoop [1].

Ground operations

Operated by CNSA, with the sampling system developed and tested by the China Academy of Space Technology, whose testing ground supplied the terrestrial evaluation data alongside the lunar images captured during the mission [1]. On Chang’e-5 an operator confirmed each sample point from imagery and a specialized operator performed pose fine-tuning, a procedure requiring complex training and carrying a high time cost, that campaign running 15.82 h for 12 surface sampling operations [7]; the tradeoff between full autonomy and this kind of supervised, high-training-cost operation is exactly the risk-and-cost calculation autonomy is weighed against [2]. On Chang’e-6 the operator monitors and intervenes only when required, and the reported result is a reduced time cost at maintained sampling quality, within the roughly 48 hour surface window [11].

Post-flight ground operations follow the Chang’e-5 pattern: the returner landed in Siziwang Banner and the sample container and certificate were handed over by CNSA to the Chinese Academy of Sciences on 28 June 2024, at which point the mass was announced as 1935.3 g [12].

Technologies developed

Four technologies come out of this mission: autonomous sample point selection from reconstructed terrain point clouds, with a loss function that encodes the manipulator’s own kinematic limitations, including the missing roll degree of freedom, as a terrain criterion; template-based 6D pose estimation of a manipulator’s end effector by point cloud registration, achieving 1.1 mm and 0.4 degrees with under 30 percent of the tool visible and no fiducial markers; explicit error decomposition for a scooping manipulator, separating arm flexible deformation, soil sinking under the contact disc at 5 to 20 mm depending on density, and force-sensor threshold support force, each measured with a different combination of arm kinematics and vision; and extension of the method by its authors to drilling and grabbing samplers, placing a drill away from obstacles and measuring pose for the vertical alignment drilling requires [1]. Insensitivity to occlusion, imaging distance and lighting direction in the pose estimation reduces the complexity required of the sampler mechanism itself.

The returned material is farside mare basalt with an eruption age of about 2.8 Ga [10], which is the sample the site selection and the sampling autonomy were built to obtain [9]. Bulk and true density measured on the returned scoop sample, and particle-size statistics from a single subsample, are broadly consistent with Chang’e-5 and Apollo values, though each rests on one sample or a small pair of aliquots [5].

What is not established

The vision pipeline’s accuracy numbers, including the 1.1 mm and 0.4 degree pose precision and the 3.8 mm scooping depth error, are all terrestrial testbed results using a balloon to offload gravity; no on-orbit accuracy verification, achieved lunar scooping depths, or in-flight failure and retry statistics have been published [1]. The claimed reduction in time cost relative to Chang’e-5 is asserted rather than quantified, with no before-and-after operator timings given for the two missions, and the division between onboard and ground computation in the processing chain is not stated in the sources consulted either. Regolith mechanical properties used here as design context, including relative density and bulk density, come from six Apollo and three Luna equatorial sites and from a single Chang’e-6 scoop sample respectively, and neither is a direct measurement at the Chang’e-6 site itself [3], [5].

References