cs.ROOct 29, 2026
SaveAI for Science in Space: Orbital Laboratories, Planetary Fieldwork, and Embodied Autonomy
Organizations: Astroup Lab, Beijing, China · Harbin Institute of Technology, Beijing Research Institute · Peking University
Abstract
Artificial intelligence for science (AI4S) is extending from interpretation of existing data to the design and execution of experiments. Space offers distinctive scientific opportunities while restricting physical support. Microgravity and extraterrestrial environments expose phenomena that terrestrial experiments cannot fully reproduce; finite samples, constrained resources, and limited opportunities for intervention complicate sustained investigation. This critical review examines orbital laboratories and planetary in situ science across biology, materials, fluid and thermal physics, and adaptive exploration. We distinguish learning from space-derived data, selecting scientific actions, executing them, and preserving experimental validity. Terrestrial self-driving laboratories provide complementary evidence on optimization, dedicated automation, robotic manipulation, and the support needed for unattended operation. The synthesis connects scientific gains already demonstrated by bounded AI systems to gaps in observation, executable actions, and continuity of valid evidence. Space robots can expand the attainable scope of autonomous science when necessary interventions exceed validated instrument capabilities and timely human support is unavailable. Their value must be established against dedicated mechanisms, redundancy, and human-supported operation. We develop a framework connecting sample history, scientific decisions, physical execution, and claim-specific validity. A worked orbital design, a planetary extension, and matched evaluation schemes separate the contributions of sensing, decision policies, and physical actions. The resulting agenda measures autonomy through valid scientific return, task coverage, intervention costs, and evidence integrity across complete campaigns.
Figures & tables
| Function | Scientific contribution | Evidence needed to establish the claim |
|---|---|---|
| Data interpretation | Estimate observables, predict responses, identify patterns or candidate mechanisms | Data provenance, appropriate held-out evaluation, uncertainty and scientific interpretation |
| Scientific decision | Choose a target, setting, intervention or measurement | Explicit objective; acquired evidence changes the choice; meaningful baseline |
| Physical execution | Obtain the selected observation or apply the intervention | Demonstrated action in a specified environment; known human contributions |
| Validity and recovery | Verify the result, diagnose interruption, preserve or restore useful evidence | Checks on sample and measurement state; recovery outcomes beyond cleared alarms |
Table 1: A functional taxonomy for interpreting space AI4S claims, organized by complementary capabilities and their evidence requirements.
| Related work | Established emphasis | Focus of this synthesis |
|---|---|---|
| Deep-space biological laboratories [ 10 ] | AI-supported biological research and self-driving laboratories in deep space | Compare orbital biology with physical sciences and planetary fieldwork using explicit deployment and support boundaries. |
| Terrestrial self-driving laboratories [ 4 , 11 ] | Experimental feedback, integrated hardware and software, scientific acceleration and access | Examine how limited intervention opportunities and experiment history constrain transfer to space. |
| Robotics in self-driving laboratories [ 12 ] | Robot functions, instrument interfaces, generality, maintenance and reconfiguration | Connect a missing action to a scientific acceptance condition and compare it with fixed mechanisms, redundancy and human support. |
Table 2: Position relative to closely related perspectives and reviews, comparing their established emphases with the focus of this synthesis. Record-level source-access limits are documented in the evidence supplement.
| Setting or constraint | Consequence for scientific work | Design implication to test |
|---|---|---|
| Pressurized orbital laboratory | Microgravity alters transport and loading within a controlled cabin or chamber | Validate mixing, sampling and measurement models; retain crew and ground roles |
| Exposed orbital payload | Vacuum and external exposure differ from cabin conditions; shielding and enclosures modify what reaches the sample | Verify containment, thermal control and calibration under the actual exposure |
| Lunar or martian surface | Partial gravity, destination-specific surface conditions and contact mechanics affect access and sampling | Validate instrument placement and acquisition; retain geological context |
| Small-body sampling | Weak gravity and contact disturbance couple acquisition to material retention | Confirm sample presence, containment and preservation after contact |
| Assistance timing | LEO contact gaps and interplanetary delays impose different limits relative to the decision window | Delegate expiring opportunities locally; retain consultation, pause and escalation options |
| Resources and history: all settings | Finite consumables, restricted access and prior exposure limit which comparisons remain feasible | Track sample history and resources; verify validity and provide feasible exchange or preservation actions |
Table 3: Setting-specific constraints and proposed design responses for validation. Conditions refer to the experiment’s actual enclosure and exposure history.
| Case / evidence | Method and scientific task | Demonstrated result | Baseline, support and boundary |
|---|---|---|---|
| GLARE [ 37 ] Ground analysis | Sparse autoencoders, transfer pretraining and ensemble clustering of plant transcriptomes. | Recovered cell-wall-associated patterns and identified a hypoxia-associated cluster for biological follow-up. | Alternative representations and clustering methods; researcher interpretation. Gene-record validation does not establish independent-mission prediction. |
| ISS crystal growth [ 13 ] Ground inference | Bayesian optimization of a transport model using existing GaSb/InSb growth observations. | Estimated a concentration-dependent diffusion coefficient. | Experiment–simulation matching; abstract-level access. Comparative predictive benefit and adaptive flight operation are not established by the accessed evidence. |
| ISS FBCE [ 59 ] Ground prediction | Separate feed-forward networks for heat transfer and critical heat flux. | Randomly held-out test-subset errors of 7.99% and 12.05%, respectively; lower than assessed correlations. | Consolidated ISS/Earth FC-72 observations; random 70/15/15 training/test/validation split. Transfer across held-out runs or geometry remains a separate test. |
| ISS wet foams [ 14 , 69 ] Post-acquisition analysis | Neural-network segmentation for bubble-size measurements. | Cross-method measurement comparison and bias assessment support physical observables. | Manual/scripted estimates and calibration. Computing location: ? (unreported, 2023); G(i) (inferred ground, 2024 software study). No apparatus feedback established. |
Table 4: Technical synthesis of representative space-science cases: analysis of acquired flight data. Each row retains its original validation scope and within-study baseline.
| Case / evidence | Method and scientific task | Demonstrated result | Baseline, support and boundary |
|---|---|---|---|
| MOBP [ 23 ] ISS automation comparator | Programmed fluid transfers and power-loss responses; no AI-guided experiment selection reported. | Microbial substrate conversion; only part of the transfer schedule executed. | No AI comparator. External power/thermal support and missing histories limited quantitative yield interpretation. |
| AEGIS [ 8 , 80 ] Mars execution | Image segmentation, preference ranking and autonomous instrument targeting. | More frequent selection of requested material than expected without intelligent targeting; successful geochemical acquisitions. | Scientist-defined preferences; blind-targeting expectation. Deployed on Curiosity/ChemCam and Perseverance/SuperCam. Preference agreement is distinct from discovery quality. |
| PIXL [ 9 ] Mars execution | Ground-built classifiers trigger longer spectral integrations onboard. | Refined composition estimates and rejected some spurious elemental detections. | Short versus combined dwells; benefits vary by mineral and rule. Finite dwell budget; ground rule maintenance. |
| Zoë [ 99 ] Ground field analogue | Spectral interpretation, hypothesis-map updates and adaptive exploration. | Increased data collection per command cycle; evolving geological interpretations. | Shared-team knowledge, handheld inputs and simulated rendezvous limit comparison. Omitted pedogenic kaolinite origin caused first-season hydrothermal misattribution; hypothesis-map completeness requires team revision. |
Table 5: Technical synthesis of representative space-science cases (continued): implemented experimental or field workflows. MOBP provides a dedicated-automation comparator alongside AI-supported workflows.
Figure 1: Representative scientific workflows and operational comparators, classified by function and demonstrated location. Locations and actors are coded independently. M denotes machine execution, including programmed automation; NR denotes a capability unreported in the selected evidence. Analysis location is distinguished from data origin. Interpret/assess denotes machine analysis. Columns show complementary functions across illustrative cases; an integrated loop requires evidence linking those functions. Fault responses distinguish inspection, restart, replacement and restored function. Sources correspond to GLARE [ 37 ] ; wet foams [ 14 , 69 ] ; FBCE [ 15 , 59 , 60 ] ; MOBP [ 23 ] ; CAL [ 73 , 74 ] ; AEGIS [ 8 , 80 ] ; PIXL [ 9 ] ; ISAAC [ 101 ] ; and Curiosity recovery [ 24 ] . For foams, ? denotes the unreported 2023 analysis location; G(i) denotes inferred ground reanalysis in the 2024 software study, based on its offline labeled-image workflow. MOBP resumed part of a predefined schedule; missing experimental history remained unresolved. CAL module replacement was performed by crew; fault recovery was not established. ISAAC inspected a mock vent after synthetic telemetry; it did not repair the obstruction. Curiosity’s workaround was developed on the ground. AEGIS preferences and PIXL rules were specified on Earth.
Figure 2: Complementary terrestrial routes toward experimental autonomy. Clio combined adaptive formulation with dedicated fluidics; its 42 physical experiments and simulated search-efficiency comparison are distinct results [ 110 ] . eVOLVER illustrates dedicated automation with residual vial replacement [ 111 ] . Ada reported different replenishment intervals in two configurations and workflows, not a controlled before–after comparison [ 112 , 113 ] ; Kanda et al. demonstrated a long robotic optimization campaign with specified human preparation, imaging and validation [ 114 ] . The lower panel is an original, proposed space-transfer strategy; the case annotations are not a common acceleration metric.
Figure 3: Different outcomes of attempted recovery. Curiosity recovered sample acquisition through ground-developed feed-extended drilling [ 24 ] . Arm assistance helped parts of InSight’s HP3 penetration attempts, but the mole did not achieve the required depth [ 145 ] . MOBP partially resumed programmed operations after interruptions, but uncertain liquid volumes and incomplete records prevented calculation of product yield [ 23 ] . These cases distinguish physical adaptability, bounded mechanical benefit, and loss of quantitative scientific validity.
| Experimental bottleneck | Dedicated or human-assisted alternative | Possible robotic contribution | Scientific acceptance condition |
|---|---|---|---|
| Policy or workflow limit. A useful synthesis treatment lies outside the implemented policy or workflow [ 126 ] | Expand the experiment policy or integrate additional steps using existing equipment. | Add mechanics only if the revised protocol requires an unavailable physical operation. | Establish whether policy, workflow or hardware is limiting; retain valid negatives. |
| Unreliable evidence. Ambiguous signal or phase assignment [ 128 , 127 ] | Use an appropriate sensor or independent analysis; audit interpretation. | Inspect or reposition only where physical geometry causes the ambiguity. | Establish why the observation or interpretation is now credible. |
| Routine physical support. Fouling or depleted consumables [ 111 , 113 ] | Programmed cleaning and fluid transfer; human vial replacement or replenishment. | Candidate module exchange where fixed fluid paths cannot provide the required action. | Preserve cleanliness and sample lineage; include replenishment and support effort. |
| Workflow access gap. Samples must cross instruments [ 86 , 125 ] | Dedicated carousel or transfer path; human handling and interpretation. | Transport samples among compatible synthesis and analytical stations. | Verify identity, transfer completion and analytical comparability. |
| Execution fault. Known sample-placement error [ 129 ] | Candidate fixtures and presence checks; human response to unsupported errors. | Bounded placement correction followed by reinspection. | Measure actual correction and subsequent experiment validity, not detection alone. |
| Physical capability limit. Failed actuation or unsuitable contact mechanics [ 24 , 145 ] | Built-in alternatives or redundancy where available; ground-developed procedures. | Exploit existing arm or tool capabilities; missing mechanics can still prevent recovery. | Demonstrate recovered science; retain failed attempts and the human support burden. |
Table 6: Choosing an intervention from the experimental bottleneck. Scientific negatives, invalid evidence, and physical limits require different responses. Cited cases establish the bottlenecks; candidate pathways and acceptance conditions are proposed here for evaluation. Dedicated automation and human assistance remain explicit comparators.
Figure 4: Proposed integration of scientific decisions, physical execution, and validity checks. Scientific planning selects an executable action, while physical-state estimation and observation establish the status of the experiment. Valid, unresolved and invalid outcomes lead to distinct responses before replanning. Invalidity is specific to the affected scientific claim; all outcomes and interventions remain in the record. Dashed returns update both accounts of state. Continuation, recovery, preservation, stopping and requests for support are explicit policy choices. Original schematic based on the synthesis in Sections 3.1 – 6 .
| Step | Complete history: continue | Unresolved delivery: preserve |
|---|---|---|
| Question and observations | Estimate the MYH1 flight–ground expression contrast at the assigned day-10 termination. Identity, temperature and an independent calibrated delivery record satisfy the predefined culture protocol. | Before day 10, flow telemetry becomes unreliable and a gap in the independent delivery record leaves actual perfusion unknown. Images and command acknowledgements cannot resolve the missing history. |
| Decision and action | Continue culture, then deliver RNAlater and transfer the sealed unit to qualified cold storage at the assigned time; terminate matched ground chips on schedule. | Trigger early preservation through an intact, separately verified reagent path, then transfer to qualified cold storage. If either step cannot be verified, keep preservation status unresolved and seek support. |
| Independent acceptance | An assessor blinded to controller type reconciles sample identity, delivery, thermal and termination records; an independent laboratory checks RNA integrity and sequencing quality before estimating the contrast and its uncertainty. | The same laboratory may accept an RNA measurement at the recorded early termination time if identity, preservation and assay checks pass. Unknown prior perfusion still excludes the planned exposure comparison, even with intact RNA. |
| Retained and lost claims | Retain the specified transcript comparison if all criteria pass; RNA quality alone does not establish valid exposure. | Retain a descriptive molecular endpoint with the interruption recorded. Do not relabel it a day-10 result or infer a flight effect from the unmatched history. Replacement starts a new unit and baseline. |
Table 7: A proposed muscle-chip decision branch. The reported study supplies the transcript endpoint, preservation precedent and RNA-quality assessment [ 16 ] ; the controller, independent delivery witness and exchange interface are new design requirements for validation.
| Criterion | What to report | Comparison or independent check |
|---|---|---|
| Scientific return and coverage | Task-specific parameter error or interval, hypothesis discrimination, or required geological coverage; valid outcomes and attempted, skipped and abandoned tasks | Preassigned scientific question and budgets; report distinct retained endpoints separately, without treating their counts as equivalent value |
| Scientific acceptance | Valid runs / all attempted runs; false acceptances / all invalid test cases | Endpoint-specific history and measurement criteria; independent or blinded assessment |
| Component benefit | Change attributable to sensing, policy or a specified physical action | Same-hardware policy comparison; matched sensing/policy for action comparison; human comparator |
| Recovery and preservation | Detection, diagnosis, intervention and valid restoration counts; preservation and stops separately | Common fault distribution, including misses; stratify by recoverability relative to declared sensing, actions and support |
| Support burden | Crew/expert effort, consumables, calibration and downtime | Entire campaign, including setup, replenishment and recovery |
| Generalization | Held-out performance across runs, samples, environments and fault combinations | Prevent leakage across related samples and repeated measurements |
Table 8: Evaluation priorities for sustained scientific campaigns. Predefine tasks, experimental units, at least one scientific information measure per task, fault cases and resource budgets. Report counts and denominators, including abstentions and failed attempts; thresholds are task-specific and must be justified before comparison.
Figure 5: Proposed validation matrix. Environmental setting and duration are independent dimensions; extended ground operation and a short flight trial answer different questions. Each cell specifies evidence needed for a prospective validation test, not a readiness score. Task scope, represented environmental factors, and available support must be reported throughout, with dedicated-automation and human-assisted comparators. Original synthesis.
More in this topic
AI for Science
AI for Science applies machine learning and AI agents to scientific data, models, experiments, and discovery workflows.
See all 161 papers