Search NASA⌕ Search

SEARCH · Search NASA

Results for “Continual Learning”

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.

At least 667 records · Page 37
All sourcesNASA NTRSDOE OSTI
NASA NTRSDate not supplied

Biological Research and Space Health Enabled by Machine Learning to Support Deep Space Missions

A key science goal of the NASA “Moon to Mars” campaign is to understand how biology responds to the Lunar, Martian, and deep space environments in order to advance fundamental knowledge, reduce risk, and support safe, productive human space missions. Through the powerful emerging approaches of artificial intelligence (AI) and machine learning (ML), a paradigm shift has begun in biomedical science and engineered astronaut health systems, to enable Earth independence and autonomy of mission operations. Here we present an overview of AI/ML architecture to support deep space mission goals, developed with leaders in the field. First, we focus on the fundamental biological research that supports our understanding of physiological responses to spaceflight, and we describe current efforts to support AI/ML research including data standardization and data engineering through maximally open and FAIR (findable, accessible, interoperable, reusable) databases and the generation of AI-ready datasets for reuse and analysis. We also discuss remote data management frameworks for research data as well as environmental and health data that are generated during deep space missions. We highlight several research projects that leverage data standardization and management for fundamental biological discovery to uncover the complex effects of space travel on living systems. Next, we provide an overview of cutting-edge AI/ML approaches that can be integrated to support remote monitoring and analysis during deep space missions, including generative models and large language models to learn the underlying biomedical patterns and predict outcomes or answer questions during off world medical scenarios. We also describe current AI/ML methods to support this research and monitoring through automated cloud-based labs which enable limited human intervention and closed-loop experimentation in remote settings. These labs could support mission autonomy by analyzing environmental data streams, and would be facilitated through in situ analytics capabilities to avoid sending large raw data files through low bandwidth communications. Finally, in the context of deep space missions with limited communications or access to medical advice from Earth, we describe a solution for integrated, real-time mission biomonitoring across hierarchical levels from continuous environmental monitoring, to wearables and point-of-care devices, to molecular and physiological monitoring. We introduce a precision space health system that will ensure that the future of space health is predictive, preventative, participatory and personalized.

artificial intelligence↗
NASA NTRSDate not supplied

Machine Learning Approaches for Rare-Earth Silicate Environmental Barrier Coating Thermochemical and Thermomechanical Property Predictions

Environmental barrier coatings (EBCs) are a necessary enabling technology for the transition from superalloys to silicon carbide (SiC) ceramic matrix composites (CMCs) in gas turbine engines for increased efficiency and decreased fuel costs. SiC-based CMCs are prone to oxidation-based degradation in the engine hot section, and rare-earth (RE) silicates are promising candidates for EBCs due to their close thermal expansion match to the composite substrate and oxidation resistance. However, the design of EBCs is hindered by the large chemical space of candidate materials and the difficulty in obtaining material properties for engineering optimization. This is especially difficult as research continues into mixed-cation or “high-entropy” RE silicates. First-principles computational methods such as density functional theory (DFT) are highly effective at calculating material properties to guide coating design but are limited by their computational cost. Atomistic simulations have the potential to both accelerate property calculations and expand the properties able to be calculated due to their lower computational compared to DFT. However, they require interatomic potentials (IAPs) specific to the material system of interest, and, to our knowledge, there are no suitable IAPs for RE silicates. Machine learning (ML) is a promising technique to accelerate material property predictions indirectly by generating IAPs for atomistic simulations or via direct prediction. In this work, we present two ML approaches to accelerate the calculation of RE silicate properties relevant to EBC design: 1) a ML-derived interatomic potential (IAP) for atomistic simulations of yttrium disilicate (Y2Si2O7) from DFT training data, and 2) a neural network (NN) model to directly predict thermochemical properties of RE silicates and oxides directly from easily obtainable unit cell parameters. Classical MD simulations using the IAP yield lattice properties and bond lengths in good agreement with both DFT and experimental results from x-ray diffraction. Thermodynamic properties calculated using the finite-displacement phonon method and quasi-harmonic approximation were orders of magnitude faster than DFT with good agreement to the DFT results. The IAP was also used to calculate properties such as coefficient of thermal expansion (CTE) that require large simulation supercells and are therefore difficult with DFT. The IAP correctly predicted the anisotropic nature of the CTE in three different phases of Y2Si2O7. The NN model predicts constant pressure heat capacity, Cp, orders of magnitude faster than DFT calculations, which can enable its use as a surrogate model for multiscale simulations. The two methods presented in this work demonstrate the utility of ML for accelerating the prediction of RE silicate properties, which can in turn accelerate EBC design and optimization.

machine learning↗
DOE OSTI2024

Trust Not Verify? The Critical Need for Data Curation Standards in Materials Informatics

The importance of data curation has been recognized in multiple areas of research; however, the discussion of this important issue is only beginning to emerge in materials science. In this Perspective, we highlight the benefits of using the standardized data curation protocols in materials science and discuss current gaps in accurate and reproducible data reporting using case studies drawn from high-impact materials science papers and well-known databases such as the Crystallography Open Database (COD) and the Cambridge Structural Database (CSD). We argue that both experimental and computational materials scientists need to embrace a culture of rigorous data curation as part of modern research data management. We propose a sample data curation pipeline for materials chemistry and illustrate its use by creating two new materials chemistry databases. Here, we hope that this perspective will serve to catalyze further discussion and promote the continuous development of rigorous data curation practices within the materials science research community. We posit that adherence to best practices of data curation will promote and enhance the reliability, reproducibility, and integrity of materials research and enable the development of reliable AI and machine learning models that critically depend on the use of quality data.

Chemical structure↗
DOE OSTI2025

Scaling Uintah on the Aurora Exascale System up to 122,880 Intel Ponte Vecchio Xe Stacks

The challenge of being able to scale application codes based on the Asynchronous Many-Task (AMT) Uintah framework on the Department of Energy (DOE) Aurora exascale system is addressed in this work by considering a challenging Reverse Monte Carlo Ray Tracing radiation benchmark calculation. This benchmark involves potentially global all-to-all communication and uses adaptive mesh refinement and ray tracing to achieve scalability. This benchmark has been used as part of previous scalability studies on a number of pre-exascale systems and on the DOE Frontier exascale system. This paper describes steps taken to enable this benchmark to run successfully on up to 10,240 nodes and 122,880 Intel® Ponte Vecchio Xe stacks on the DOE Aurora exascale system. This scalability was achieved through a limited number of experiments on Aurora, given machine loads and its uniqueness. These experiments constitute valuable lessons learned to achieve scalability at this level. The resulting scalability runs, while few in number, demonstrate relatively good strong-scaling characteristics. A detailed analysis of these results provides important indications about the path to scalability on Aurora for future work. Overall, these results continue the remarkable ability of this AMT approach to produce scalable solutions for challenging problems at extreme scale on heterogeneous architectures.

Garcia, Marta [Argonne National Laboratory (ANL)]↗
DOE OSTI2024

PhyML

Python scripts for physics informed machine learning for time-dependent problems. The code implements hard constrained sequential physics-informed neural networks (HCSPINNs) using JAX library. The details of the method and implementation can be found in the following paper: Roy, P., & Castonguay, S. (2024). Exact Enforcement of Temporal Continuity in Sequential Physics- Informed Neural Networks. arXiv preprint arXiv:2403.03223. (https://arxiv.org/abs/2403.03223)

Roy, Pratanu↗
DOE OSTI2025

CSP Plant Optimization Study for the California Power Market (“CalCSP”) (Final Technical Report)

Concentrating Solar Power (CSP) with thermal energy storage offers a unique and strategic opportunity to support California’s clean energy transition. Unlike photovoltaic (PV) systems, CSP with thermal storage can generate electricity after sunset and during periods of high demand, making it a valuable complement to intermittent renewable resources. CSP also provides synchronous, inertia-contributing generation, long-duration storage, and flexible dispatch—capabilities increasingly important as thermal plants retire. This report summarizes the findings of the CSP Plant Optimization Study for the California Power Market or “CalCSP study,” which evaluated the technical, economic, environmental, and policy factors that influence the deployment of CSP technologies in California. The CalCSP study was conducted to assess how CSP can contribute to California’s long-term decarbonization goals while enhancing grid reliability, supporting local economic development, and making efficient use of land and transmission resources. It draws on detailed modeling of CSP performance and costs, site suitability analysis, policy reviews, and stakeholder engagement across utilities, regulators, developers, and community organizations. The analysis focuses on mature molten-salt tower technology and incorporates lessons learned from the global CSP fleet, distinguishing today’s CSP from earlier first-of-a-kind projects in the U.S. The findings support a more prominent role for CSP in California’s evolving clean energy landscape. With strategic planning, targeted policy support, and continued cost improvements, CSP can complement PV and batteries to deliver reliable, around-the-clock clean electricity—especially in areas with high solar resource and constrained grid capacity.

14 SOLAR ENERGY↗
NASA NTRS1995

Reconstruction of an infrared band of meteorological satellite imagery with abductive networks

As the current fleet of meteorological satellites age, the accuracy of the imagery sensed on a spectral channel of the image scanning system is continually and progressively degraded by noise. In time, that data may even become unusable. We describe a novel approach to the reconstruction of the noisy satellite imagery according to empirical functional relationships that tie the spectral channels together. Abductive networks are applied to automatically learn the empirical functional relationships between the data sensed on the other spectral channels to calculate the data that should have been sensed on the corrupted channel. Using imagery unaffected by noise, it is demonstrated that abductive networks correctly predict the noise-free observed data.

Singer, Harvey A.↗
DOE OSTI2026

Project Development of an Electrochemical Denitration and Caustic Generation System for HLW Pretreatment at Hanford - 26350

An engineering-scale electrochemical processing skid is proposed to perform the denitration of Hanford tank waste, which would help to mitigate a key process concern with the direct feed processing of the Hanford Tank Waste Treatment and Immobilization Plant (WTP). The reduction of nitrates and organic compounds in the waste feed will directly reduce hazardous NOx and ammonia gases generated during the vitrification process, which in turn will aid in addressing potential regulatory and safety challenges associated with processing large volumes of tank waste. This paper highlights the past legacy work, project layout, accomplishments from Phase 1 and research and development envisioned for Phase 2. An innovative electrochemical denitration and caustic generation (EDCGe) process was demonstrated for the pretreatment of tank waste at the Savannah River Site (SRS) in the early 2000s. The denitration electrolyzer, off-gas abatement system, and caustic generator electrolyzer are being developed with the intent that the denitration electrolyzer will convert nitrate and nitrite anions to nitrogen gas while also yielding other gaseous byproducts, which may include N2O, NH3, VOCs, and H2. The gaseous byproducts will be managed via a tandem off-gas catalyst-bed treatment system. The caustic generation electrolyzer will recycle NaOH from the feed to produce a clean caustic stream for use within the batching tanks at Hanford, aiding in the preparation of waste for WTP. The reduction in hazardous emissions and improved waste treatment processes provides a robust solution for nuclear waste management, contributing to environmental safety and regulatory compliance. The EDCGe technology is being adapted, modified, and updated for the preparation of the Direct Feed-High Level Waste (DF-HLW) flowsheet at Hanford. Phase 1 demonstrated a bench-scale proof-of-concept for reactions involving the denitration electrolyzer and gas phase abatement of ammonia. The electrochemical technology is drawing on the scientific outcomes that were reported in the legacy work. The results from Phase 1 demonstrated the viability of the EDCGe system in reducing the nitrogen species of simple non-radioactive waste simulants. Commercially available alloys used as electrode materials and membranes are being studied for the denitration and caustic generation electrolyzers. The continuation of this project holds promise for broader applications, such as energy-efficient ammonia production, and contributes significant advancements in nuclear waste management. Additional material discovery has been investigated into ceramic Na super ion conductive (NaSICON) materials and off-gas abatement catalyst discovery. NaSICON is of interest for selective transport of Na within the electrolyzers to make a clean caustic stream. Future integration and optimization efforts, informed by Phase 1 results and ongoing research, will continue to drive advancements in nuclear waste management technology. The technology developed for the EDCGe treatment of tank waste will also have broader potential to inform other fields, such as energy-efficient ammonia production, as well as ammonia abatement catalysis through the lessons learned in electrochemical nitrate reduction. The applications and benefits of this research extend beyond Hanford and the Savannah River Site, supported by a collaborative team of scientists and engineers from national labs, academia, and industry, ensuring a comprehensive approach to solving complex waste treatment challenges. The team is leveraging advanced electrochemical technologies, machine learning, novel catalysts tailored for gaseous nitrogen species, and cutting-edge reactor systems to enhance the process efficiency and effectiveness of the denitration process.

Rodene, Dylan [Savannah River National Laboratory ↗
DOE OSTI2024

Denoising Seismic Waveforms Using a Wavelet-Transform-Based Machine-Learning Method

Seismic waveform data recorded at stations can be thought of as a superposition of the signal from a source of interest and noise from other sources. Frequency‐based filtering methods for waveform denoising do not result in desired outcomes when the targeted signal and noise occupy similar frequency bands. Recently, denoising techniques based on deep‐learning convolutional neural networks (CNNs), in which a recorded waveform is decomposed into signal and noise components, have led to improved results. These CNN methods, which use short‐time Fourier transform representations of the time series, provide signal and noise masks for the input waveform. These masks are used to create denoised signal and designaled noise waveforms, respectively. However, advancements in the field of image denoising have shown the benefits of incorporating discrete wavelet transforms (DWTs) into CNN architectures to create multilevel wavelet CNN (MWCNN) models. The MWCNN model preserves the details of the input due to the good time–frequency localization of the DWT. In this report we use a data set of over 382,000 constructed seismograms recorded by the University of Utah Seismograph Stations network to compare the performance of CNN and MWCNN‐based denoising models. Evaluation of both models on constructed test data shows that the MWCNN model outperforms the CNN model in the ability to recover the ground‐truth signal component in terms of both waveform similarity and preservation of amplitude information. Model evaluation of real‐world data shows that both the CNN and MWCNN models outperform standard band‐pass filtering (BPF; average improvement in signal‐to‐noise ratio of 9.6 and 19.7 dB, respectively, with respect to BPF). Evaluation of continuous data suggests the MWCNN denoiser can improve both signal detection capabilities and phase arrival time estimates.

58 GEOSCIENCES↗
NASA NTRS2006

Open Microphone Speech Understanding: Correct Discrimination Of In Domain Speech

An ideal spoken dialogue system listens continually and determines which utterances were spoken to it, understands them and responds appropriately while ignoring the rest This paper outlines a simple method for achieving this goal which involves trading a slightly higher false rejection rate of in domain utterances for a higher correct rejection rate of Out of Domain (OOD) utterances. The system recognizes semantic entities specified by a unification grammar which is specialized by Explanation Based Learning (EBL). so that it only uses rules which are seen in the training data. The resulting grammar has probabilities assigned to each construct so that overgeneralizations are not a problem. The resulting system only recognizes utterances which reduce to a valid logical form which has meaning for the system and rejects the rest. A class N-gram grammar has been trained on the same training data. This system gives good recognition performance and offers good Out of Domain discrimination when combined with the semantic analysis. The resulting systems were tested on a Space Station Robot Dialogue Speech Database and a subset of the OGI conversational speech database. Both systems run in real time on a PC laptop and the present performance allows continuous listening with an acceptably low false acceptance rate. This type of open microphone system has been used in the Clarissa procedure reading and navigation spoken dialogue system which is being tested on the International Space Station.

Hieronymus, James↗
NASA NTRS1970

Analysis of a Decision-Directed Receiver with Unknown Priors

A decision-directed receiver (DDR) uses previous outputs (decisions) to estimate unknown parameters and, on the basis of these estimates, modifies the detector structure for subsequent decisions. Although the DDR is less complex to instrument than other adaptive schemes, inherent in the decision-directed approach is the possibility of a runaway. This occurs when the detector commits a sequence of decision errors resulting in a degradation of parameter estimates, which, in turn, results in a further deterioration of detector performance. Because of the dependencies introduced by the learning process, runaway is difficult to analyze. In this paper, a DDR with unknown a priori probabilities is considered. The priors are estimated by the relative frequency of decisions of that event. For binary detection, it is shown that there is a positive probability of a runaway (the estimates converge to 1 or 0), which equals 1 if the signal-to-noise ratio is below a critical value. A tight bound on the probability of a runaway is obtained by approximating the learning process by a random walk with independent increments. The analysis demonstrates that a runaway is quite improbable even for moderate signal-to-noise ratios. The analysis is extended to multiple signals and to the situation where the estimates of the priors are updated continuously through exponential weighting rather than allowed to converge.

Davisson, Lee D.↗
NASA NTRS1993

MEDUSA - An overset grid flow solver for network-based parallel computer systems

Continuing improvement in processing speed has made it feasible to solve the Reynolds-Averaged Navier-Stokes equations for simple three-dimensional flows on advanced workstations. Combining multiple workstations into a network-based heterogeneous parallel computer allows the application of programming principles learned on MIMD (Multiple Instruction Multiple Data) distributed memory parallel computers to the solution of larger problems. An overset-grid flow solution code has been developed which uses a cluster of workstations as a network-based parallel computer. Inter-process communication is provided by the Parallel Virtual Machine (PVM) software. Solution speed equivalent to one-third of a Cray-YMP processor has been achieved from a cluster of nine commonly used engineering workstation processors. Load imbalance and communication overhead are the principal impediments to parallel efficiency in this application.

Smith, Merritt H.↗
NASA NTRS1977

Voice technology and BBN

The following research was discussed: (1) speech signal processing; (2) automatic speech recognition; (3) continuous speech understanding; (4) speaker recognition; (5) speech compression; (6) subjective and objective evaluation of speech communication system; (7) measurement of the intelligibility and quality of speech when degraded by noise or other masking stimuli; (8) speech synthesis; (9) instructional aids for second-language learning and for training of the deaf; and (10) investigation of speech correlates of psychological stress. Experimental psychology, control systems, and human factors engineering, which are often relevant to the proper design and operation of speech systems are described.

Wolf, Jared J.↗
DOE OSTI2025

gRASPA

GPU Monte Carlo Simulation Code with a taste of RASPA We present enhancements in Monte Carlo simulation speed and functionality within an open-source code, gRASPA, which uses graphical processing units (GPUs) to achieve significant performance improvements compared to serial, CPU implementations of Monte Carlo. The code supports a wide range of Monte Carlo simulations, including canonical ensemble (NVT), grand canonical, NVT Gibbs, Widom test particle insertions, and continuous-fractional component Monte Carlo. Implementation of grand canonical transition matrix Monte Carlo (GC-TMMC) and a novel feature to allow different moves for the different components of metal-organic framework (MOF) structures exemplify the capabilities of gRASPA for precise free energy calculations and enhanced adsorption studies, respectively. The introduction of a High-Throughput Computing (HTC) mode permits many Monte Carlo simulations on a single GPU device for accelerated materials discovery. The code can incorporate machine learning (ML) potentials. The open-source nature of gRASPA promotes reproducibility and openness in science, and users may add features to the code and optimize it for their own purposes. The code is written in CUDA/C++ and SYCL/C++ to support different GPU vendors. The gRASPA code is publicly available at https://github.com/snurr-group/gRASPA.

Li, Zhao [Purdue/Northwestern/Notre Dame Universit↗
NASA NTRSDate not supplied

BioSentinel: Leading the Way for Deep Space CubeSat Missions

Flagship science missions are not alone in Deep Space thanks to BioSentinel, a 6U spacecraft launched on Artemis-1. BioSentinel is one of the longest operating CubeSats beyond cislunar space. The subsystems and COTS components of the BioSentinel bus are a template for future deep space missions, and the lessons learned from over a year of operations will enable improved performance for the next missions. BioSentinel achieved its unprecedented performance for an SLS secondary payload due to preparation, planning, and a robust design. Pre-launch antenna and interface testing with both DSN and ESA confirmed command and data pathways and allowed for operational flexibility in the critical early hours post-deployment. Mission Operations simulations prior to launch identified potential risks and primed operators to respond in flight, preparing the team to react quickly to successfully detumble the spacecraft and enter a power-positive state. The spacecraft would not have survived without the inclusion of the trailblazing 3D-printed composite cold gas propulsion system. The non-standard tank geometry enabled efficient use of the limited space available in the CubeSat, as well as the capability to detumble the spacecraft and manage momentum, while providing sufficient margin to execute potential delta-V maneuvers. The Iris radio has operated for over 18 months with no significant issues. Initial Iris performance estimates have been accurate throughout the mission. BioSentinel continues to collect data on thermal conditions and to validate our performance models with real-world knowledge. We have received exemplary support from our DSN partners. Following the conclusion of the primary science mission, the Linear Energy Transfer (LET) Spectrometer continued to collect solar and galactic radiation data from its location in heliocentric orbit. The free space dataset offered by the BioSentinel LET is a valuable source of data for model validation and future mission planning. As the spacecraft travels farther from Earth it is poised to provide longitudinally distributed measurements of solar particle events during solar maximum. The lessons learned from BioSentinel suggest key areas to enhance performance. The ability to upload modified flight software can increase the stability of memory management. Additional heaters in the propulsion system design have already proven successful on the Starling mission. Streamlining mission operations can reduce costs, increase data return, and better utilize DSN time. Enhancements such as these will facilitate reliable, long-duration deep space exploration using the proven BioSentinel 6U CubeSat bus.

BioSentinel↗
DOE OSTI2024

Physics-constrained machine learning for electrodynamics without gauge ambiguity based on Fourier transformed Maxwell’s equations

We utilize a Fourier transformation-based representation of Maxwell’s equations to develop physics-constrained neural networks for electrodynamics without gauge ambiguity, which we label the Fourier–Helmholtz–Maxwell neural operator method. In this approach, both of Gauss’s laws and Faraday’s law are built in as hard constraints, as well as the longitudinal component of Ampère–Maxwell in Fourier space, assuming the continuity equation. An encoder–decoder network acts as a solution operator for the transverse components of the Fourier transformed vector potential, $\hat{A}_⟂(k,t)$, whose two degrees of freedom are used to predict the electromagnetic fields. This method was tested on two electron beam simulations. Among the models investigated, it was found that a U-Net architecture exhibited the best performance as it trained quicker, was more accurate and generalized better than the other architectures examined. We demonstrate that our approach is useful for solving Maxwell’s equations for the electromagnetic fields generated by intense relativistic charged particle beams and that it generalizes well to unseen test data, while being orders of magnitude quicker than conventional simulations. We show that the model can be re-trained to make highly accurate predictions in as few as 20 epochs on a previously unseen data set.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Accurate and Fast Anomaly Detection in Additive Composite-Based Manufacturing using Thermal Cameras

Today, large-scale additive manufacturing with plastics and composite materials requires continuous monitoring by experienced staff to prevent, detect and correct anomalous events affecting the performance of the printed part. We address the complexity of this demanding task by designing a camera-based anomaly detection system utilizing probabilistic principal component analysis (PPCA). This is a machine learning technique is trained with thermal images collected during normal operation of the large-scale printer (Cincinnati BAAM). This technique is advantageous for practical applications as there is no need to artificially introduce anomalous conditions into model training. During deployment, we challenge this model by introducing deliberate variations of the extruder speed. We reduce extrusion speed to a lower level, between 70 and 95% of the nominal value to collected test images. Our results show that images are easily identified as anomalous for extruder speeds at or below 85% of the nominal speed, meaning that an anomalous reduction of the material deposition rate can be detected within seconds of its onset. We show that our results are robust to (a) camera-to-camera variability and (b) print-to-print variability.

Pike, John [ORNL]↗
NASA NTRS1987

The 3-D world modeling with updating capability based on combinatorial geometry

A 3-D world modeling technique using range data is discribed. Range data quantify the distances from the sensor focal plane to the object surface, i.e., the 3-D coordinates of discrete points on the object surface are known. The approach proposed herein for 3-D world modeling is based on the Combinatorial Geometry (CG) method which is widely used in Monte Carlo particle transport calculations. First, each measured point on the object surface is surrounded by a small sphere with a radius determined by the range to that point. Then, the 3-D shapes of the visible surfaces are obtained by taking the (Boolean) union of all the spheres. The result is an unambiguous representation of the object's boundary surfaces. The pre-learned partial knowledge of the environment can be also represented using the CG Method with a relatively small amount of data. Using the CG type of representation, distances in desired directions to boundary surfaces of various objects are efficiently calculated. This feature is particularly useful for continuously verifying the world model against the data provided by a range finder, and for integrating range data from successive locations of the robot during motion. The efficiency of the proposed approach is illustrated by simulations of a spherical robot in a 3-D room in the presence of moving obstacles and inadequate prelearned partial knowledge of the environment.

Goldstein, M.↗