Search NASA⌕ Search

SEARCH · Search NASA

Results for “controller tuning”

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 361 records · Page 20
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Amine-Thiol/Selenol Chemistry for Efficient and Stable Perovskite Solar Cells

Controlling the crystallization of perovskites is imperative to reduce defect densities in perovskite thin films and extend device lifetimes. In this work, combinations of amine and chalcogenide ligands were introduced in the sequential deposition method to fabricate highly crystalline and orientated formamidinium lead iodide thin films with reduced defect densities and increased charge carrier lifetimes. The dual additives can tune the perovskite intermediate state and control the crystallization leading to devices with improved efficiencies and stabilities. While thiophenol failed to prevent the amine ligand from degrading the perovskite precursors, benzene selenol combinations with amine-ligands drastically changed the solution chemistry to increase the PbI 2 conversion to highly crystalline and orientated α-FAPbI 3 films with lower defect densities. X-ray photoelectron spectroscopy studies reveal benzene selenol evaporates from the thin film leaving behind a modified surface, which is associated with the amine additive. These results indicate the amine selection can be used to tune the surface properties. Lastly, we propose a highly tunable I 2 reduction strategy using chalcogenide chemistry to help enable the realization of perovskite solar cells with high performance and stability.

defects↗
DOE OSTI2023

Tuning magnetoelectricity in a mixed-anisotropy antiferromagnet

Control of magnetization and electric polarization is attractive in relation to tailoring materials for data storage and devices such as sensors or antennae. In magnetoelectric materials, these degrees of freedom are closely coupled, allowing polarization to be controlled by a magnetic field, and magnetization by an electric field, but the magnitude of the effect remains a challenge in the case of single-phase magnetoelectrics for applications. We demonstrate that the magnetoelectric properties of the mixed-anisotropy antiferromagnet LiNi 1-x Fe x PO 4 are profoundly affected by partial substitution of Ni 2+ ions with Fe 2+ on the transition metal site. This introduces random site-dependent single-ion anisotropy energies and causes a lowering of the magnetic symmetry of the system. In turn, magnetoelectric couplings that are symmetry-forbidden in the parent compounds, LiNiPO 4 and LiFePO 4 , are unlocked and the dominant coupling is enhanced by almost two orders of magnitude. Our results demonstrate the potential of mixed-anisotropy magnets for tuning magnetoelectric properties.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2021

Tunable Stochasticity in an Artificial Spin Network

Metamaterials present the possibility of artificially generating advanced functionalities through engineering of their internal structure. Artificial spin networks, in which a large number of nanoscale magnetic elements are coupled together, are promising metamaterial candidates that enable the control of collective magnetic behavior through tuning of the local interaction between elements. In this work, the motion of magnetic domain-walls in an artificial spin network leads to a tunable stochastic response of the metamaterial, which can be tailored through an external magnetic field and local lattice modifications. This type of tunable stochastic network produces a controllable random response exploiting intrinsic stochasticity within magnetic domain-wall motion at the nanoscale. An iconic demonstration used to illustrate the control of randomness is the Galton board. In this system, multiple balls fall into an array of pegs to generate a bell-shaped curve that can be modified via the array spacing or the tilt of the board. A nanoscale recreation of this experiment using an artificial spin network is employed to demonstrate tunable stochasticity. Furthermore, this type of tunable stochastic network opens new paths toward post-Von Neumann computing architectures such as Bayesian sensing or random neural networks, in which stochasticity is harnessed to efficiently perform complex computational tasks.

Artificial spin network↗
DOE OSTI2025

Tuning magnetism in Ising-type van der Waals magnet FePS 3 by lithium intercalation

Recently, layered transition metal thiophosphate MPX 3 (M = transition metals, X = S or Se) have gained significant attention because of their rich magnetic, optical, and electronic properties. Specifically, the diverse magnetic structures and the robustness of magnetism in the two-dimensional (2D) limit have made them prominent candidates to study 2D magnetism. Numerous efforts such as substitutions and interlayer intercalations have been adopted to tune the magnetic properties of these materials, which has greatly deepened the understanding of the underlying mechanisms that govern the properties. In this work, we focus on modifying the magnetism of Ising-type antiferromagnet FePS 3 using electrochemical lithium intercalation. Furthermore, our work demonstrate the effectiveness of electrochemical intercalation as a controllable tool to modulating magnetism, including tuning magnetic ordering temperature and inducing low temperature spin-glass state, offering an approach for implementing this material into applications.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2024

Transient Covalent Polymers through Carbodiimide‐Driven Assembly

Biochemical systems make use of out-of-equilibrium polymers generated under kinetic control. Inspired by these systems, many abiotic supramolecular polymers driven by chemical fuel reactions have been reported. Conversely, polymers based on transient covalent bonds have received little attention, even though they have the potential to complement supramolecular systems by generating transient structures based on stronger bonds and by offering a straightforward tuning of reaction kinetics. In this study, we show that simple aqueous dicarboxylic acids give poly(anhydrides) when treated with the carbodiimide EDC. Transient covalent polymers with molecular weights exceeding 15,000 are generated which then decompose over the course of hours to weeks. Disassembly kinetics can be controlled using simple substituent effects in the monomer design. The impact of solvent polarity, carbodiimide concentration, temperature, pyridine concentration, and monomer concentration on polymer properties and lifetimes has been investigated. The results reveal substantial control over polymer assembly and disassembly kinetics, highlighting the potential for fine-tuned kinetic control in nonequilibrium polymerization systems.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Continuous Encodable Reshaping of Gold Nanocrystals through Facet Modulation

Shape control of nanocrystals (NCs) is crucial for tuning their assembly behavior and functional properties, yet the precise manipulation of facet composition remains challenging. Here, we present a nanocrystal reshaping strategy to control and modulate the facets of gold (Au) NCs. Our one-pot approach, conducted at room temperature, requires only initial Au NCs, Au 3+ ions, and surfactants, distinguishing it from conventional reduction-mediated “etching-and-regrowth” methods. Detailed structural studies using electron microscopy, small-angle X-ray scattering (SAXS), and UV−vis spectroscopy reveal the surfactant-encoded pathway for NC transformation from shaped particles to spheres and then into various polyhedral shapes while preserving the individual particles' volume. The proposed reshaping mechanism involves the dissolution of surface Au atoms into Au + complexes in the presence of Au 3+ and surfactant, followed by surfactant-guided redeposition and formation of facets with different atomic planes. Using the ethanol oxidation reaction (EOR) as a probe, we observe a quasi-linear decrease in onset potential and an increase in activity with increasing {100} facet exposure. This work broadens synthetic strategies by offering precise NC reshaping and facet control.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Controlled Growth of Ta Oxide With Reactive Sputtering for Overlap Josephson Junction Application

Josephson junctions $\($JJs$\)$ are indispensable elements for superconducting quantum electronics. There is a growing need to develop processes for high quality JJs. Here we explore the growth of Ta oxide as a dielectric barrier for Metal-Oxide-Metal overlap JJ implementations. We study films of Ta oxides prepared with reactive magnetron sputtering. TaOx is grown on silicon and Ta films with varying RF power and oxygen concentrations. Tuning these parameters allows us to control the rate of deposition as well as composition of polycrystalline Ta oxides needed to meet our dielectric requirements for Josephson Junctions.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

Aero‐servo‐elastic co‐optimization of large wind turbine blades with distributed aerodynamic control devices

Abstract This work introduces automated wind turbine optimization techniques based on full aero‐servo‐elastic models and investigates the potential of trailing edge flaps to reduce the levelized cost of energy (LCOE) of wind turbines. The Wind Energy with Integrated Servo‐control (WEIS) framework is improved to conduct the presented research. Novel methods for the generic implementation and tuning of trailing edge flap devices and their controller are also introduced. Primary flap and controller parameters are optimized to demonstrate potential maximum blade tip deflection reductions of 21 % . Concurrent design optimization (i.e., co‐design) of a novel segmented wind turbine blade with trailing edge flaps and its controller is then conducted to demonstrate blade cost savings of 5 % . Additionally, rotor diameter co‐design optimization is demonstrated to reduce the LCOE by 1.3 % without significant load increases to the tower. These results demonstrate the efficacy of control co‐design optimization using trailing edge flaps, and the entirety of this work provides a foundation for numerous control co‐design‐oriented studies for distributed aerodynamic control devices.

17 WIND ENERGY↗
DOE OSTI2024

Performance Demonstration of an Occupancy Sensor-enabled Integrated Solution for Commercial Buildings

Traditionally, a single-loop fixed-gain controller is applied to supply fan (SF) and cooling coil (CC) valve controls while a fixed-damper position control is applied to outdoor air (OA) damper control at air handling units (AHUs) in commercial buildings. With the increasing application of occupancy sensors, the information generated by occupancy sensors is applied to not only reduce the electricity loads from lighting and controllable plug loads, but also reset OA intake and minimum supply airflow setpoints. Meanwhile, these intermittent operation actions greatly elevate the dynamics of AHU systems, which may introduce unstable SF and CC valve operations and inaccurate OA flow control at AHUs and consequently degrade maximum energy efficiency gains. With virtual fan and valve flow meter technologies, two advanced controls, including cascade control and gain scheduling control, can be implemented on both the SF and CC valve, and an advanced control using a virtual OA flow meter can be implemented on OA damper integrated with occupancy sensors. The goal of this project is to demonstrate the savings, cost, and performance of an integrated solution that integrates the three advanced HVAC controls with occupancy sensors to allow accurate and stable AHU operations in real buildings. The project objectives are to: 1) develop and validate an advance SF control algorithm; 2) develop and validate an advanced CC valve control algorithm; 3) validate an algorithm to implement a virtual OA flow meter; and 4) demonstrate the savings, cost, and performance of the integrated solution in real buildings. The technical approaches are to: 1) select a test system at the University of Oklahoma; 2) develop and implement the algorithms of advanced SF and CC valve controls and validate the performance; 3) develop and implement the advanced OA control using a virtual OA flow meter and validate the performance; 4) demonstrate the savings, cost, and performance of the proposed integrated solution with and without three advanced HVAC controls; and 5) disseminate the project results through publications and presentations. For the SF control, both the gain scheduling and cascade controls can improve the fan energy performance by reducing the fan power during the transient period and the fan control performance at lower speeds by reducing fan speed variation. Moreover, the gain scheduling control provides a simple and low-cost solution and is recommended. The fan power savings can reach 30% in a transient period. For the CC valve control, the gain scheduling control can considerably reduce the supply air temperature oscillation range and frequency under both higher and lower load conditions and the control valve response is much more stable. As a result, the gain scheduling control is recommended. The projected pump energy consumption can be reduced by 68.5%. With the developed virtual OA flow meter, the OA can be accurately controlled at its setpoint, which is determined based on the actual number of occupants in the building provided by occupancy sensors. The RMSE of the proposed OA control is 15.9 L/s. The energy data shows that the fan power and CC cooling energy were significantly reduced. On the other hand, the energy savings majorly results from the occupancy sensors and the energy savings by the advanced HVAC controls is minimal because that the controllers in the test AHU were tuned with very slow response. An annual technical savings potential is estimated as 0.5 quads in the commercial sector.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2024

Phase control of spin waves in the van der Waals antiferromagnet NiPS 3

Here we demonstrate phase control of magnons in the van der Waals antiferromagnet NiPS3 using optical excitation by polarized light. The sign of the coherent precession of spin amplitude changes upon (i) reversing the helicity of a circularly polarized pump beam, or (ii) rotating the polarization of a linearly polarized pump by π/2. Because these two excitation pathways have comparable generation efficiency, the phase of spin precession can be continuously tuned from 0 to 2π by controlling the polarization state of the pump pulse. The ability to excite magnons with a desired phase has potential applications in the design of a spin-wave phased array and ultrafast spin information processing.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2024

Optical tuning of the diamond Fermi level measured by correlated scanning probe microscopy and quantum defect spectroscopy

Quantum technologies based on quantum point defects in crystals require control over the defect charge state. Here we tune the charge state of shallow nitrogen-vacancy and silicon-vacancy centers by locally oxidizing a hydrogenated surface with moderate optical excitation and simultaneous spectral monitoring. The loss of conductivity and change in work function due to oxidation are measured in atmosphere using conductive atomic force microscopy and Kelvin probe force microscopy (KPFM). We correlate these scanning probe measurements with optical spectroscopy of the nitrogen-vacancy and silicon-vacancy centers created via implantation 15–25 nm beneath the diamond surface and annealing. The observed charge state of the defects as a function of optical exposure demonstrates that laser oxidation provides a way to precisely tune the Fermi level over a range of at least 2.00 eV. We also observe a significantly larger oxidation rate for implanted surfaces compared to unimplanted surfaces under ambient conditions. Here, combined with knowledge of the electron affinity of a surface, these results suggest KPFM is a powerful, high-spatial-resolution technique to advance surface Fermi level engineering for charge stabilization of quantum defects.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Bi-directional tuning of thermal transport in SrCoO x with electrochemically induced phase transitions

Unlike the wide-ranging dynamic control of electrical conductivity, there does not exist an analogous ability to tune thermal conductivity by means of electric potential. The traditional picture assumes that atoms inserted into a material’s lattice act purely as a source of scattering for thermal carriers, which can only reduce thermal conductivity. In contrast, here we show that the electrochemical control of oxygen and proton concentration in an oxide provides a new ability to bi-directionally control thermal conductivity. On electrochemically oxygenating the brownmillerite SrCoO 2.5 to the perovskite SrCoO 3–δ , the thermal conductivity increases by a factor of 2.5, whereas protonating it to form hydrogenated SrCoO 2.5 effectively reduces the thermal conductivity by a factor of four. This bi-directional tuning of thermal conductivity across a nearly 10 ± 4-fold range at room temperature is achieved by using ionic liquid gating to trigger the ‘tri-state’ phase transitions in a single device. We elucidated the effects of these anionic and cationic species, and the resultant changes in lattice constants and lattice symmetry on thermal conductivity by combining chemical and structural information from X-ray absorption spectroscopy with thermoreflectance thermal conductivity measurements and ab initio calculations. This ability to control multiple ion types, multiple phase transitions and electronic conductivity that spans metallic through to insulating behaviour in oxides by electrical means provides a new framework for tuning thermal transport over a wide range.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Modulating the Surface Ligand Orientation for Stabilized Anionic Redox in Li‐Rich Oxide Cathodes

Abstract Anionic redox chemistry is emerging as a key concept in the development of high‐energy lithium‐ion batteries, as it enables a nearly doubled charge storage capacity, aiding the development of high‐capacity batteries. However, the anionic reactivity is frequently irreversible from charge to discharge, leading to rapid decay of the capacity and voltage of batteries over long‐term cycling. Although the possibility of controlling the anionic redox reactions by tuning the geometric and electronic structures has recently been proposed, the implementation of this strategy is still a critical challenge. Herein, a strategy is proposed to improve the anionic redox reversibility of a model anionic redox active cathode material, Li 1.2 Ni 0.13 Co 0.13 Mn 0.54 O 2 , by tuning the surface ligand geometry via the growth of a lattice‐compatible spinel LiCoO 2 coating layer on the particle surface. Detailed local structure and first principles investigations reveal that the shape and orientation of the octahedral layer in the host lattice are modified. Accordingly, a two‐band oxygen redox behavior is triggered in the ligand‐orientation‐regulated Li‐rich cathode, leading to enhanced reversibility, and thus, remarkably improved capacity and voltage retention over cycling. This study highlights the importance of controllable ligand orientation, carving a new path for the development and design of Li‐rich cathodes in the future.

Liu, Yanchen↗
DOE OSTI2023

Simple Monomers for Precise Polymer Functionalization During Ring-Opening Metathesis Polymerization

Controlling the monomer sequence of synthetic polymers is a grand challenge in polymer science. Conventional sequence control has been achieved in dispersed polymers by exploiting the kinetic tendencies of monomers and their order of addition. While the sequence of blocks in multiblock copolymers can be readily tuned using sequential addition of monomers (SAM), control over the sequence distribution is eroded as the targeted block size approaches a single monomer unit (i.e., X n ~ 1) due to the stochastic nature of chain-growth reactions. Thus, unique monomers are needed to ensure precise single additions. Herein, we investigate common classes of cyclic olefin monomers for ring-opening metathesis polymerization (ROMP) to identify monomers for single unit addition during sequential monomer addition synthesis. Through careful analysis of polymerization kinetics, we find that easily synthesized oxanorbornene imide monomers are suitable for single-addition reactions. With the identified monomers, we demonstrate the synthesis of multiblock copolymers containing up to three precise functionalization sites and singly cross-linked four-armed star copolymers. In conclusion, we envision that expanded kinetic analyses of monomer reactivities in ROMP reactions will enable novel polymer synthesis capabilities such as the autonomous synthesis of sequence-defined polymers or one-shot multiblock copolymer syntheses.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Robust highly durable solid oxide fuel cell cathodes – Improved materials compatibility & self-regulating surface chemistry

Solid oxide fuel cells (SOFCs) are electrochemical conversion devices that directly transform hydrogen or hydrocarbon fuels to electricity, with energy efficiencies as high as 90%, coupled with reduced emissions. Several factors, however, remain to be addressed when considering scale-up of SOFC technology, including the need to overcome decreased performance due to sluggish rates of the oxygen reduction reaction (ORR) at the cathode under reduced temperatures and susceptibility to degradation in performance from surface poisoning e.g. from chromia, while limiting the use of critical raw materials (lanthanides and transition metals) present in high performing mixed ionic electronic conducting electrodes like (La,Sr)CoO 3 (LSC). In this project we explored the key descriptors for determining ORR activity in SOFC electrodes and tried to recover performance degradation by applying them to SOFC electrodes. In order to do this, we first selected a model mixed ionic electronic conducting (MIEC) oxide, Pr-doped CeO 2 (Pr 0.1 Ce 0.9 O 2-δ , PCO), which is a chemically stable fluorite and free of inherent poison sources (e.g. Sr segregation in LSC) that potentially react with external impurities such as Cr-species vaporized from the interconnect. The three approaches originally planned in this project are as follows: 1) evaluation of scavenger exsolution characteristics, 2) study of scavengers gettering efficacy following Cr and Si poisoning and 3) integration of new compositions into porous electrodes. Among them, exceptional progress has been made in 2) and 3), especially understanding the role of surface infiltrants in impacting electrode performance and degradation of PCO materials. We found that the Smith acidity scale for binary oxides serves as a powerful descriptor for tuning and predicting the oxygen exchange kinetics on MIEC PCO surfaces. As a result, with infiltration with binary oxides, ranging from strongly basic (Li 2 O) to strongly acidic (SiO 2 ) onto the surface of porous PCO, it was possible to systematically vary the chemical surface exchange coefficient (k chem ) by 6 orders of magnitude! L i2 O increased k chem by nearly 1,000 times over that of pristine PCO, while SiO 2 decreased k chem by nearly the same factor. Strikingly, although the pre-exponential of k chem scales linearly with the acidity of the infiltrated binary oxide, there is nearly no change in the activation energy. With this insight, we attributed the origin of these dramatic changes in k chem values to the systematic increase and decrease in the surface electron density induced by infiltrated binary oxides. More interestingly, although both Cr 2 O 3 and SiO 2 were determined to be acidic by Smith, suggesting that this feature could likely be the primary reason that these compounds serve to poison the ORR on SOFC cathodes, the effect of poisoning could be subsequently tuned by adding multiple infiltrants and controlling their relative surface acidities. We also systematically examined the effect of serial infiltration of basic and acidic oxides. It turned out that serial infiltration of Li not only recovers approximately 20-fold degraded k chem of PCO by acidic Cr 2 O 3 but its k chem is enhanced even beyond that of the non-infiltrated PCO by more than three orders of magnitude. This was further verified with a screen-printing PCO symmetric cell in terms of the electrode performance (area-specific resistance, ASR) related to approach 3). These observations point to acidity as a key descriptor not only in tuning and predicting the ORR activity of SOFC cathodes that largely determines the overall performance of SOFC, but in mitigating and reactivating poisoned electrode performance. This work provides novel guidelines for making the electrode performance much more active and robust in SOFCs, which can further be applied to all applications requiring oxygen exchange reaction, such as electrolyzers, permeation membranes and gas sensors.

30 DIRECT ENERGY CONVERSION↗
DOE OSTI2023

Gradient-Descent-Based Learning Gain for Backstepping Controller and Disturbance Observer of Nonlinear Systems

This paper proposes a gradient-descent-based learning (GL) gain for backstepping controller and disturbance observer (DOB) of nonlinear system. The proposed method consists of the GL gain update law, controller, and DOB. The GL gain update law is proposed to adapt the control gain and DOB gain according to the direction that minimizes the cost function. The mathematical analysis reveals that the GL gain always has a positive sign and upper bound. The controller is designed via a backstepping procedure to track the desired output with GL control gain. The DOB is designed to estimate the unknown external disturbance with the GL DOB gain. Because the control and DOB gains are simultaneously tuned to achieve improved performance, the time consumption for tuning can be reduced. In addition, the peaking phenomenon can be avoided initially by a small initial value of GL gains. The stability of the closed-loop system is guaranteed using the input-to-state stability property. The performance of the proposed method was validated via simulations and experiments using a DC motor.

42 ENGINEERING↗
DOE OSTI2026

Reaching quantum critical point by adding nonmagnetic disorder in single crystals of superconductor (Ca 𝑥 ⁢Sr 1−𝑥 ) 3 ⁢Rh 4 ⁢Sn 13

The Remeika series superconductor, (Ca 𝑥 ⁢Sr 1−𝑥 ) 3 ⁢Rh 4 ⁢Sn 13 , shows a rare nonmagnetic quantum critical point (QCP) associated with the continuous charge-density wave (CDW) and structural transition under the “dome” of superconductivity achieved by tuning composition and applying pressure. Here, we use a nonmagnetic pointlike disorder induced by 2.5 MeV electron irradiation to suppress the CDW and drive the system to and even beyond the QCP. This conclusion is based on a clear evolution of temperature-dependent resistivity, 𝜌⁡(𝑇), from the Fermi liquid to the non-Fermi liquid regime with increasing amount of disorder. Starting on the CDW side, below the suggested QCP concentration of 𝑥 𝑐 =0.9, added disorder resulted in a progressively larger linear term and a reduced quadratic term in 𝜌⁡(𝑇). Nearly perfect 𝑇−linear dependence is observed at the dose at which long-range CDW order is suppressed to 𝑇=0, consistent with the expectations. We refine the QCP location in this system and place it in the interval between 𝑥=0.75 and 0.85. Our results strongly support the concept that the disorder can tune the system to the quantum critical regime and even beyond. It follows from the argument by Imry and Ma that any ordered phase is unstable toward quenched disorder. Introduced in a controlled way, this disorder becomes a nonthermal tuning parameter likely applicable to a variety of different systems.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2020

How Low Can You Go? Using Synthetic 3D Imagery to Drastically Reduce Real-World Training Data for Object Detection

Deep convolutional neural networks (DCNNs) currently provide state-of-the-art performance on image classification and object detection tasks, and there are many global security mission areas where such models could be extremely useful. Crucially, the success of these models is driven in large part by the widespread availability of high-quality open source data sets such as Image Net, Common Objects in Context (COCO), and KITTI, which contain millions of images with thousands of unique labels. However, global security relevant objects-of-interest can be difficult to obtain: relevant events are low frequency and high consequence; the content of relevant images is sensitive; and adversaries and proliferators seek to obscure their activities. For these cases where exemplar data is hard to come-by, even fine-tuning an existing model with available data can be effectively impossible. Recent work demonstrated that models can be trained using a combination of real-world and synthetic images generated from 3D representations; that such models can exceed the performance of models trained using real-world data alone; and that the generated images need not be perfectly realistic (Tremblay, et al., 2018). However, this approach still required hundreds to thousands of real-world images for training and fine tuning, which for sparse, global security-relevant datasets can be an unrealistic hurdle. In this research, we validate the performance and behavior of DCNN models as we drive the number of real-world images used for training object detection tasks down to a minimal set. We perform multiple experiments to identify the best approach to train DCNNs from an extremely small set of real-world images. In doing so, we: Develop state-of-the-art, parameterized 3D models based on real-world images and sample from their parameters to increase the variance in synthetic image training data; Use machine learning explainability techniques to highlight and correct through targeted training the biases that result from training using completely synthetic images; and Validate our results by comparing the performance of the models trained on synthetic data to one another, and to a control model created by fine-tuning an existing ImageNet-trained model with a limited number (hundreds) of real-world images.

97 MATHEMATICS AND COMPUTING↗