SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
The LBFD X-59 Propulsion Overview is prepared in support of a panel session at the 2026 AIAA Aviation Conference. The overview covers highlights of the X-59 propulsion system, a short history of propulsion testing at NASA GRC, A summary of X-59 ground testing, a summary of X-59 flight testing, and finally some lessons learned.
NASA's Orion spacecraft transports humans and cargo into cislunar space for the Artemis program. The European Service Module (ESM), supplied by ESA and its European industry partners, provides Orion with power and in-space propulsion. The Orion-ESM propulsion system is a bipropellant hypergolic propulsion system using monomethyl hydrazine (MMH) and nitrogen tetroxide (MON-3). Primary translational propulsion is provided by the Orbital Maneuvering System Engine(OMS-E), with backup translational propulsion provided by eight Auxiliary thrusters (AUX). Attitude control and small translational maneuvers are provided by twenty four Reaction Control System (RCS) engines. The 2022 Artemis I mission was the first integrated flight test of the Orion-ESM spacecraft and its propulsion system. The OMS-E used on Artemis I was a refurbished Space Shuttle OMS-E that previously flew on nineteen missions ranging from STS-41G in 1984 to STS-112 in 2002. The Auxiliary engines are modified Aerojet Rocketdyne R4D-11 engines produced specifically for the Orion program. The RCS engines are Ariane Group engines originally used for the Automated Transfer Vehicle (ATV) program. This paper will discuss the unique operational requirements for each engine on Orion and the development and qualification effort sat both the engine and system-level that were completed to enable a successful Artemis I mission. Next the paper will evaluate the in-flight performance of the engines during the Artemis I mission showing nominal performance as expected. Additionally, comparisons to models will be presented showing very good correlation. Finally, the paper will address the plan for the engines on future Orion missions and the evolution of the system operation.
NASA's Orion spacecraft transports humans and cargo into cislunar space for the Artemis program. The European Service Module (ESM), supplied by ESA and its European industry partners, provides Orion with power and in-space propulsion. The Orion-ESM propulsion system is a bipropellant hypergolic propulsion system using monomethyl hydrazine (MMH) and nitrogen tetroxide (MON-3). Primary translational propulsion is provided by the Orbital Maneuvering System Engine(OMS-E), with backup translational propulsion provided by eight Auxiliary thrusters (AUX). Attitude control and small translational maneuvers are provided by twenty four Reaction Control System (RCS) engines. The 2022 Artemis I mission was the first integrated flight test of the Orion-ESM spacecraft and its propulsion system. The OMS-E used on Artemis I was a refurbished Space Shuttle OMS-E that previously flew on nineteen missions ranging from STS-41G in 1984 to STS-112 in 2002. The Auxiliary engines are modified Aerojet Rocketdyne R4D-11 engines produced specifically for the Orion program. The RCS engines are Ariane Group engines originally used for the Automated Transfer Vehicle (ATV) program. This paper will discuss the unique operational requirements for each engine on Orion and the development and qualification effort sat both the engine and system-level that were completed to enable a successful Artemis I mission. Next the paper will evaluate the in-flight performance of the engines during the Artemis I mission showing nominal performance as expected. Additionally, comparisons to models will be presented showing very good correlation. Finally, the paper will address the plan for the engines on future Orion missions and the evolution of the system operation.
The thermal evolutions of the Moon, Mars, Venus, and Mercury are calculated theoretically starting from cosmochemical condensation models. An assortment of geological, geochemical, and geophysical data are used to constrain both the present-day temperatures and the thermal histories of the planets' interiors. Such data imply that the planets were heated during or shortly after formation and that all the terrestrial planets started their differentiations early in their history. Initial temperatures and core formation play the most important roles in the early differentiation. The size of the planet is the primary factor in determining its present-day thermal state. The Moon, smallest in size, is characterized as a differentiated body with a crust, a thick solid mantle, and an interior region which may be partially molten. It is presently cooling rapidly and is relatively inactive tectonically. Mercury, which probably has a large core, may have a 500-km-thick solid lithosphere and a partially molten core, if it is assumed that some heat sources exist in the core. If this is not the case, the planet's interior temperatures are everywhere below the melting curve for iron. The thermal evolution is dominated by the core separation and the high conductivity of iron that makes up the bulk of Mercury. Mars, intermediate in size, is assumed to have differentiated an Fe-FeS core. While the formation of an early crust is evident, large-scale melting and differentiation of the mantle silicates has occurred at least up until 1 b.y. ago. Present-day temperature profiles indicate moderate tectonic activity at the present time. Venus is characterized as a planet not unlike the Earth in many respects. Core formation has occurred probably during the first billion years after the formation. The present-day temperatures indicate a partially molten upper mantle overlain by a 100-km-thick lithosphere and a molten Fe-Ni core. We can expect that today Venus may have tectonic processes similar to the Earth's.
Since its inception in the early 2010s, the NASA High Efficiency Centrifugal Compressor (HECC) has been enigmatic for the propulsion research community: experimental data and numerical simulations of the stage have generally not aligned in their quantifications of performance metrics. Typically, the fault for these disagreements is assigned to the numerical simulations as the simulations are models, and models are inherently incomplete representations of the experiment. This “incompleteness” may manifest in assumptions regarding roughness or heat transfer, simplifications of the flow path (i.e., neglecting bleed flows), or the oft-scapegoated turbulence model. In the case of HECC, recent work showed unexpected discrepancies between the intended impeller geometry defined in the design report (termed Design-Intent) and the manufactured impeller used in the experimental campaigns (termed As-Manufactured). That work used numerical simulations to establish that the geometric differences between the Design-Intent and As-Manufactured impellers were significant enough to result in drastically different performance predictions for the HECC vaneless diffuser configuration. The Design-Intent impeller simulations over predicted the performance relative to the experiment, whereas the As-Manufactured simulations better represented the experimental data, both in terms of one-dimensional performance metrics and spanwise flow profiles. This effort expands on that work by examining in detail the geometry and aerodynamic performance of the HECC vaned diffuser configuration. Further differences between the Design-Intent and As-Manufactured geometries have been discovered in the vaned diffuser and exit guide vanes, and these differences are documented herein. The summations of the geometric differences for all of the components were used to create two numerical models of HECC vaned diffuser configuration: the Design-Intent simulations which are generated from the original geometry definitions given in the design report and the As-Manufactured simulations which are the best available representation of the manufactured compressor hardware used in the experimental test campaigns. In congruence with the earlier vaneless diffuser work, the numerical predictions of the Design-Intent choked mass flow rate, total pressure ratio, and efficiency were notably greater than that of the As-Manufactured simulations. To increase confidence in the experimental dataset, measurements from a recent test campaign conducted in 2024 are used to validate the original experimental data acquired from 2012 to 2014 with good repeatability overall, especially considering the passage of time and differences in the data acquisition systems between the test campaigns. Both numerical simulations were then extensively evaluated against the experimental data. The As-Manufactured simulations provided better estimates of the stage performance than the Design-Intent cases in terms of most performance metrics. Nonetheless, more detailed results still show opportunities for improvement. Despite a more accurate representation of the physical hardware, characterization of the impeller work input remains a challenge even for rigorously developed numerical models.
Since its inception in the early 2010s, the NASA High Efficiency Centrifugal Compressor (HECC) has been enigmatic for the propulsion research community: experimental data and numerical simulations of the stage have generally not aligned in their quantifications of performance metrics. Typically, the fault for these disagreements is assigned to the numerical simulations as the simulations are models, and models are inherently incomplete representations of the experiment. This “incompleteness” may manifest in assumptions regarding roughness or heat transfer, simplifications of the flow path (i.e., neglecting bleed flows), or the oft-scapegoated turbulence model. In the case of HECC, recent work showed unexpected discrepancies between the intended impeller geometry defined in the design report (termed Design-Intent) and the manufactured impeller used in the experimental campaigns (termed As-Manufactured). That work used numerical simulations to establish that the geometric differences between the Design-Intent and As-Manufactured impellers were significant enough to result in drastically different performance predictions for the HECC vaneless diffuser configuration. The Design-Intent impeller simulations over predicted the performance relative to the experiment, whereas the As-Manufactured simulations better represented the experimental data, both in terms of one-dimensional performance metrics and spanwise flow profiles. This effort expands on that work by examining in detail the geometry and aerodynamic performance of the HECC vaned diffuser configuration. Further differences between the Design-Intent and As-Manufactured geometries have been discovered in the vaned diffuser and exit guide vanes, and these differences are documented herein. The summations of the geometric differences for all of the components were used to create two numerical models of HECC vaned diffuser configuration: the Design-Intent simulations which are generated from the original geometry definitions given in the design report and the As-Manufactured simulations which are the best available representation of the manufactured compressor hardware used in the experimental test campaigns. In congruence with the earlier vaneless diffuser work, the numerical predictions of the Design-Intent choked mass flow rate, total pressure ratio, and efficiency were notably greater than that of the As-Manufactured simulations. To increase confidence in the experimental dataset, measurements from a recent test campaign conducted in 2024 are used to validate the original experimental data acquired from 2012 to 2014 with good repeatability overall, especially considering the passage of time and differences in the data acquisition systems between the test campaigns. Both numerical simulations were then extensively evaluated against the experimental data. The As-Manufactured simulations provided better estimates of the stage performance than the Design-Intent cases in terms of most performance metrics. Nonetheless, more detailed results still show opportunities for improvement. Despite a more accurate representation of the physical hardware, characterization of the impeller work input remains a challenge even for rigorously developed numerical models.
Time-resolved, high-speed self-aligned focusing schlieren images were acquired in the NASA Glenn Research Center Transonic Turbine Blade Cascade facility to help understand the aerodynamic behavior of high-pressure, thick trailing edge turbine blades. The trailing edge thickness of 9% of axial chord tested represents simulated ceramic matrix composite fabrication constraints, which was verified previously to possess a high-loss flow regime at high inlet turbulence conditions over a narrow range of Reynolds numbers and at a fixed design exit Mach number of 0.74. Our high-speed images, which were acquired at 10 distinct Reynolds numbers, show a significant increase in energy from flow oscillations due to transonic vortex shedding at Reynolds numbers corresponding to the high loss conditions. For those conditions, strong acoustic waves turn into shock waves. Spectral Proper Orthogonal Decomposition of the high-speed images shows acoustic waves from trailing edge vortex shedding at all conditions, with increased spectral energy at the high-loss conditions and slightly increasing frequency (about 6%) as a function of Reynolds number. Analysis of potential feedback timing is performed using velocity fields from a previous LES simulation, considering different feedback mechanisms. Most noteworthy is the acoustic/shock-boundary layer interaction mechanism on the suction surface at the blade geometric throat, which likely plays an important role in realistic curved blade passages.
The Surveyor Project planned and conducted seven unmanned lunar missions for which spacecraft were launched between May 1966 and January 1968. Each of the spacecraft was successfully launched with the then newly developed Atlas/Centaur vehicle which utilized for the first time a high-specific-impulse, liquid hydrogen/liquid oxygen fueled stage. Five of the spacecraft successfully soft-landed and returned a great quantity of engineering and scientific data on extensive postlanding operations, accomplishing all mission and project objectives. Four of the spacecraft soft-landed at selected mare sites to provide data which were required to support the Apollo Program. The final spacecraft was then successfully used for scientific investigation of a contrasting site in the rugged lunar highlands. Surveyor was a fully attitude-stabilized spacecraft designed to receive and execute a wide variety of earth commands, as well as to perform certain automatic functions including the critical terminal-descent and soft-landing sequences. Significant new and advanced subsystems that were developed and/or used in combination to enable Surveyor to execute the complex terminal phase of flight were: (1) a solid-propellant main retro motor, (2) throttlable liquid-propellant vernier engines (also used for midcourse velocity correction), (3) highly sensitive velocity- and altitude-sensing radars, and (4) an automatic closed-loop guidance and control system. The first Surveyor spacecraft carried a survey television camera which, together with other engineering instrumentation, obtained in-flight and postlanding data. The complement of instruments carried on later missions included various combinations of the following additional devices: (1) a soil mechanics/surface sampler instrument for picking, digging, and handling lunar surface material; (2) an alpha scattering instrument for performing a chemical analysis of the lunar surface material; and (3) magnets attached to the spacecraft for determining magnetic properties of the soil.
The Surveyor Project planned and conducted seven unmanned lunar missions for which spacecraft were launched between May 1966 and January 1968. Each of the spacecraft was successfully launched with the then newly developed Atlas/Centaur vehicle which utilized for the first time a high-specific-impulse, liquid hydrogen/liquid oxygen fueled stage. Five of the spacecraft successfully soft-landed and returned a great quantity of engineering and scientific data on extensive postlanding operations, accomplishing all mission and project objectives. Four of the spacecraft soft-landed at selected mare sites to provide data which were required to support the Apollo Program. The final spacecraft was then successfully used for scientific investigation of a contrasting site in the rugged lunar highlands. Surveyor was a fully attitude-stabilized spacecraft designed to receive and execute a wide variety of earth commands, as well as to perform certain automatic functions including the critical terminal-descent and soft-landing sequences. Significant new and advanced subsystems that were developed and/or used in combination to enable Surveyor to execute the complex terminal phase of flight were: (1) a solid-propellant main retro motor, (2) throttlable liquid-propellant vernier engines (also used for midcourse velocity correction), (3) highly sensitive velocity- and altitude-sensing radars, and (4) an automatic closed-loop guidance and control system. The first Surveyor spacecraft carried a survey television camera which, together with other engineering instrumentation, obtained in-flight and postlanding data. The complement of instruments carried on later missions included various combinations of the following additional devices: (1) a soil mechanics/surface sampler instrument for picking, digging, and handling lunar surface material; (2) an alpha scattering instrument for performing a chemical analysis of the lunar surface material; and (3) magnets attached to the spacecraft for determining magnetic properties of the soil.
This thesis developed a system-level optimization model of a regenerative fuel cell (RFC) system for long-duration, off-world energy storage applications. Prior RFC design studies have typically been limited to reduced parameter sets and simplified constraints due to computational limitations relative to the number of relevant degrees of freedom. As a result, important nonlinear interactions between subsystems have not been fully captured. This work began to address that gap by developing a higher-fidelity, nonlinear optimization framework that incorporates a broader set of design variables and coupled constraints, enabling a multidimensional model that captures the coupled behavior of RFC subsystems and demonstrates the feasibility of applying optimization to such systems. An expanded system-level optimization approach was established that captures interactions between electrochemical performance, structural requirements, and storage design. This enabled a more comprehensive evaluation of trade-offs than conventional formulations. The model integrates four coupled subsystems: a fuel cell, an electrolyzer, reactant gas, and high-pressure storage tanks, and was formulated to accommodate a wide range of mission parameters, including operational time and required output power. It incorporates constraints on available solar array power, reactant mass balance between production and consumption, and pressure-dependent storage requirements. To enable reliable convergence, the optimization problem was reformulated to reduce dimensionality and improve numerical stability, with subsystem models organized for efficient evaluation. Problem dimensionality was reduced by consolidating lower-level design variables into higher-level representative quantities, and subsystem behavior was evaluated within the optimization loop. A multi-start initialization strategy was employed to mitigate sensitivity to local minima and improve solution quality, while nonlinear relationships were solved using robust numerical methods. The results showed that convergence was achieved across a range of required output power values. Specific energy reached a maximum at a critical mission power level, where the electrolyzer power matched the available solar input and operated near its voltage and current density limits. Beyond this point, further increases in required power resulted in less mass-efficient operation, increasing total system mass and reducing overall performance. The developed model represents an advancement in RFC system-level optimization by enabling analysis of a broader and more tightly coupled design space than previous considerations. While convergence behavior and computational cost remain challenges, the methods introduced improve solvability and allow inclusion of additional design variables with minimal loss of physical fidelity. However, the numerical results should not be interpreted as definitive design recommendations, as the model includes simplifying assumptions and omits several higher-order effects. Future work should extend this framework by incorporating additional subsystems and loss mechanisms, such as thermal management, parasitic power consumption, and reactant losses, to improve fidelity and ensure more representative design conclusions.
A thermochemically-closed sonic-flow inversion framework (TSIF) is developed to infer bulk enthalpy and total temperature upstream of a choked nozzle in arc-jet flows. The formulation recasts a pressure-rise total enthalpy quantification technique as an inverse problem in characteristic-velocity c * space using measured mass flow rate, upstream total pressure, gas composition, and nozzle throat geometry as inputs. Unlike calorimetric energy-balance approaches or optical diagnostics, the method relies primarily on routinely measured facility quantities combined with explicit thermochemical closure. Thermochemical states are obtained using NASA’s open-source Chemical Equilibrium with Applications (CEA) code, enabling construction of a chemistry-consistent relation between characteristic velocity, total enthalpy, and total temperature under equilibrium or frozen assumptions. A discharge coefficient is self-calibrated using cold-flow (arc-off) operation data and applied to hot-flow (arc-on) measurements, enabling upstream losses to be accounted for without empirical correlations. The framework is applied to air, N 2 , and CO 2 –N 2 arc-jet flows and demonstrates expected trends for the inferred thermochemical states as function of arc power, specific energy input, mass-flow, heater configuration, and test gas. In the air limit, under equilibrium assumptions, the method recovers the classical high-enthalpy asymptotic correlation of Winovich with a mean residual of 4.4%, demonstrating compatibility with established sonic-flow scaling, while extending applicability to arbitrary multi-species mixtures and non-equilibrium chemistry. The framework provides a mixture-flexible methodology for determining bulk thermochemical states in modern arc-jet environments using routine facility pressure, mass-flow, gas-composition, and nozzle-geometry information together with a cold-flow calibration.
A thermochemically-closed sonic-flow inversion framework (TSIF) is developed to infer bulk enthalpy and total temperature upstream of a choked nozzle in arc-jet flows. The formulation recasts a pressure-rise total enthalpy quantification technique as an inverse problem in characteristic-velocity c * space using measured mass flow rate, upstream total pressure, gas composition, and nozzle throat geometry as inputs. Unlike calorimetric energy-balance approaches or optical diagnostics, the method relies primarily on routinely measured facility quantities combined with explicit thermochemical closure. Thermochemical states are obtained using NASA’s open-source Chemical Equilibrium with Applications (CEA) code, enabling construction of a chemistry-consistent relation between characteristic velocity, total enthalpy, and total temperature under equilibrium or frozen assumptions. A discharge coefficient is self-calibrated using cold-flow (arc-off) operation data and applied to hot-flow (arc-on) measurements, enabling upstream losses to be accounted for without empirical correlations. The framework is applied to air, N 2 , and CO 2 –N 2 arc-jet flows and demonstrates expected trends for the inferred thermochemical states as function of arc power, specific energy input, mass-flow, heater configuration, and test gas. In the air limit, under equilibrium assumptions, the method recovers the classical high-enthalpy asymptotic correlation of Winovich with a mean residual of 4.4%, demonstrating compatibility with established sonic-flow scaling, while extending applicability to arbitrary multi-species mixtures and non-equilibrium chemistry. The framework provides a mixture-flexible methodology for determining bulk thermochemical states in modern arc-jet environments using routine facility pressure, mass-flow, gas-composition, and nozzle-geometry information together with a cold-flow calibration.
Access to high-quality, high-resolution, near-real-time precipitation data is essential for hydrological and meteorological research and disaster mitigation. Traditional tools such as rain gauges and radar networks, though effective, have limitations, including sparse coverage in remote areas and high operational costs. Satellite data, with its global coverage and high spatial and temporal resolutions, mitigates limitations in coverage. Satellite precipitation products like Hydro Estimator (HE), Integrated Multi-satellitE Retrievals for Global Precipitation Measurement (IMERG), and Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks (PERSIANN) utilize both geosynchronous thermal infrared (IR) and passive microwave (PMW) data in their operation. PMW sensors offer detailed atmospheric profiles but suffer from higher latency, whereas IR sensors provide lower latency but only capture cloud-top information. Despite this constraint, IR data remains attractive for low-latency precipitation estimation. Recent advances in deep learning, particularly convolutional neural networks (CNNs), have further improved satellite precipitation retrievals. This study introduces PERSIANN-Unet (PUnet or PERSIANN V3), a quasi-global algorithm covering 60°N–60°S that combines IR data, monthly climatology, and the UNet architecture to produce half-hourly precipitation estimates at 0.04° resolution. The product is evaluated against HE, IMERG, and PDIR-Now for 2022–2023. Results show that PUnet closely matches its training target, IMERG V07 Final, at the global scale, and performance is further evaluated against Stage IV as a reference over CONUS. Training PUnet on IMERG (2016–2021) leverages a high-quality, integrated PMW IR-gauge precipitation product while developing an IR-based framework not reliant on PMW availability. By operating on a single global image, PUnet avoids tile partitioning and blending steps, reducing edge discontinuities, and produces more spatially consistent precipitation fields across hemispheres.
The failure of a nickel cadmium battery undergoing tests is discussed. Reasons for the complete destruction of the battery while undergoing preparation for thermal vacuum testing are given.
The Spartan Release Engage Mechanism (REM) is a system designed to restrain the Spartan spacecraft during Space Transportation System (STS) launch and landing. The mechanism is designed to allow deployment and retrieval of the Spartan free flyer spacecraft from the shuttle payload bay. Because current Spartan spacecraft payloads are much heavier than payloads intended for the original REM, an extensive redesign, analysis, and test program was necessary. Also, increased emphasis on safety in the post-Challenger era prompted a reevaluation of possible failures. Much of the design effort focused on improving the latch mechanism gearbox. Key concerns were effective gear lubrication, thermal gradients at the gearbox mounts, operation at thermal extremes, and gear-train failure contingencies. Increased concern for reliability led to the design of an Extra Vehicular Activity (EVA) backup latch system.
An analysis was performed to determine the effectiveness of solar reflectors in reducing heat transfer through the insulation of cryogenic fuel tanks located in deep space. A solar reflector, as used herein, is defined as a surface having a ratio of absorptance with respect to solar radiation to hemispherical emittance (α/ϵ) less than 1.0. This analysis considers the effect of surface coatings with α/ϵ values varying from 0.1 to 1.0 when applied to the external surfaces of bulk insulation or multifoil insulation. Results were obtained in terms of heat leak to two cryogenic propellants, liquid hydrogen and liquid oxygen, when stored at Mars, Earth, and Mercury orbital distances from the sun. The results show that, with sufficiently good insulation and low heat leaks to the propellant, α/ϵ is a reasonable criterion of merit for these coatings, but that, at larger heat leaks, consideration of ϵ (or α) is necessary. It was observed that selective finishes of α/ϵ = 0.1 on a given amount of bulk insulation material can reduce the heat leak to 40 to 55 percent of the heat leak obtained when gray finishes are considered. Or, for equivalent heat leaks, a reduction in bulk insulation material thickness of about one-half is possible with high-quality selective finishes (α/ϵ = 0.1) compared with flat reflectors. Larger benefits may be achieved from selective finishes painted on foils; the analysis shows that heat leak may be reduced tenfold for surfaces with an α/ϵ of 0.1 compared with gray surfaces. Again, consideration of equivalent heat leaks indicates that the use of approximately 10 times fewer foils of emissivity 0.05 is possible when spectrally selective surfaces are employed. The present state of the art in achieving such surfaces is also briefly reviewed.
Achieving larger benefits from electrified propulsion in single aisle aircraft (the largest commercial aircraft market segment) necessitates that a significant fraction of their 20 MW or greater propulsion system be electrified. 5 to possibly 20 of the MW-scale electric machines investigated by NASA would be required to meet this need. Approximately 10-20 of NASA’s MW-scale electric machines would be required to reach the necessary power levels. Therefore, electric machines of 5 MW or greater are of interest. However, conventionally cooled high power density MW-scale electric machines still produce a significant amount of waste heat, which makes obtaining these power levels problematic. Highly efficient fully superconducting electric machines (or those with a cryogenically cooled, high purity copper or aluminum stator) are a possible solution but have multiple barriers to be overcome. One such barrier is the durability of superconducting windings. Superconductors are typically brittle, and encapsulants used to hold windings in place are a significant source of stress because of the mismatches in coefficient of thermal expansion at temperatures between room temperature and operational temperatures (20 to 77 K). These encapsulants also form a thermal barrier that can be detrimental to cooling the superconductors. This paper describes an initial study of commercially available encapsulants as it relates to superconducting machine applications along with initial modeling of thermal stress in superconducting windings for a superconducting or cryogenically cooled stator in a 5 MW electric machine.
Maintaining optimal team functioning is critical for long-duration space exploration missions, yet traditional monitoring methods, such as self-reports and wearable sensors, often impose operational burdens or suffer from bias. This paper investigates a non-intrusive speech-based artificial intelligence (AI) framework to predict degradations in team functioning using data from the Human Exploration Research Analog (HERA) of the U.S. National Aeronautics and Space Administration (NASA). Using acoustic features, linguistic descriptors, and semantic embeddings, we evaluate static non-linear and temporal machine learning models to predict both objective (task accuracy) and subjective (self-reported efficacy and cohesion) team functioning outcomes. Results indicate that temporal models outperform static approaches, with prediction of objective task accuracy in Team Interaction Battery (TIB) improving from near chance to 71%. Self-reported outcomes, including team efficacy and cohesion, are predicted more reliably than task performance, achieving balanced accuracies of up to 85.56% and 78.12%, respectively, and are found to be most strongly associated with acoustic features. In a second interdependent task, the MMSEV–EVA, accuracies of up to 78% are achieved using temporal models with acoustic features. Furthermore, incorporating just 1–2 days of team-specific historical data systematically improved performance, and acoustic markers from informal pre-task interactions provided modest predictive gains. Finally, while automated preprocessing yielded viable accuracy, humancorrected data provided moderate performance gains, though transcription error rates did not significantly correlate with model performance. These findings highlight the potential of speech as a passive, high-fidelity monitoring tool for autonomous habitats.