Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Autonomous Systems”

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 181 records · Page 10

Determination of elementary first integrals of a generalized Raychaudhuri equation by the Darboux integrability method

The Darboux integrability method is particularly useful to determine first integrals of nonplanar autonomous systems of ordinary differential equations, whose associated vector fields are polynomials. In particular, we obtain first integrals for a variant of the generalized Raychaudhuri equation, which has appeared in string inspired modern cosmology.

97 MATHEMATICS AND COMPUTING↗

Periodicity, chaos, and multiple attractors in a memristor-based Shinriki's circuit

In this contribution, a novel memristor-based oscillator, obtained from Shinriki's circuit by substituting the nonlinear positive conductance with a first order memristive diode bridge, is introduced. The model is described by a continuous time four-dimensional autonomous system with smooth nonlinearities. The basic dynamical properties of the system are investigated including equilibria and stability, phase portraits, frequency spectra, bifurcation diagrams, and Lyapunov exponents' spectrum. It is found that in addition to the classical period-doubling and symmetry restoring crisis scenarios reported in the original circuit, the memristor-based oscillator experiences the unusual and striking feature of multiple attractors (i.e., coexistence of a pair of asymmetric periodic attractors with a pair of asymmetric chaotic ones) over a broad range of circuit parameters. Results of theoretical analyses are verified by laboratory experimental measurements.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Are humans still necessary? Expanding the discussion

The rise of automation, artificial intelligence (AI), and autonomous systems raises important questions about the future role of humans and the field of human factors/ergonomics in workplaces. This paper builds on Dr. Peter Hancock’s 2023 ‘Are Humans Still Necessary?’ article published in the Ergonomics journal. Using a multi-method approach that included a debate, opinion polling, roundtable discussions, and AI queries, the current effort examined the necessity of human involvement in future work environments. Debate team members presented arguments for and against the need for human workers, considering human factors, technology, and socioeconomic factors. Observations indicate that while AI may handle routine tasks, humans will likely remain essential for complex decision making, creativity, and ethical considerations. The paper advocates for viewing workplace dynamics as collaborative human-AI partnerships rather than competition, highlighting the need for a transdisciplinary approach in which human factors/ergonomics professionals play a vital role in enhancing these relationships.

99 - GENERAL AND MISCELLANEOUS↗

Application of Artificial Intelligence in Detection and Mitigation of Human Factor Errors in Nuclear Power Plants: A Review

Human factors and ergonomics have played an essential role in increasing the safety and performance of operators in the nuclear energy industry. In this critical review, we examine how artificial intelligence (AI) technologies can be leveraged to mitigate human errors, thereby improving the safety and performance of operators in nuclear power plants (NPPs). First, we discuss the various causes of human errors in NPPs. Next, we examine the ways in which AI has been introduced to and incorporated into different types of operator support systems to mitigate these human errors. We specifically examine (1) operator support systems, including decision support systems, (2) sensor fault detection systems, (3) operation validation systems, (4) operator monitoring systems, (5) autonomous control systems, (6) predictive maintenance systems, (7) automated text analysis systems, and (8) safety assessment systems. Finally, we provide some of the shortcomings of the existing AI technologies and discuss the challenges still ahead for their further adoption and implementation to provide future research directions.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

de Sitter gauge theory of gravity: an alternative torsion cosmology

A new cosmological model based on the de Sitter gauge theory (dSGT) is studied in this paper. By some transformations, we find, in the dust universe, the cosmological equations of dSGT could form an autonomous system. We conduct dynamics analysis to this system, and find 9 critical points, among which there exist one positive attractor and one negative attractor. The positive attractor shows us that our universe will enter a exponential expansion phase in the end, which is similar to the conclusion of ΛCDM. We also carry out some numerical calculations, which confirms the conclusion of dynamics analysis. Finally, we fit the model parameter and initial values to the Union 2 SNIa dataset, present the confidence contour of parameters and obtain the best-fit values of parameters of dSGT.

79 ASTRONOMY AND ASTROPHYSICS↗

It's a dark, dark world: background evolution of interacting φCDM models beyond simple exponential potentials

We study the background cosmological dynamics with a three component source content: a radiation fluid, a barotropic fluid to mimic the matter sector and a single scalar field which can act as dark energy giving rise to the late-time accelerated phase. Using the well-known dimensionless variables, we cast the dynamical equations into an autonomous system of ordinary differential equations (ASODE), which are studied by computing the fixed points and the conditions for their stability. The matter fluid and the scalar field are taken to be uncoupled at first and later, we consider a coupling between the two of the form Q = √(2/3)κβρ{sub m} φ̇ where ρ{sub m} is the barotropic fluid density. The key point of our analysis is that for the closure of ASODE, we only demand that the jerk, Γ = V V ''/ V '{sup 2} is a function of acceleration, z = −M{sub p}V'/V, that is, Γ = 1+ f ( z ). In this way, we are able to accommodate a large class of potentials that goes beyond the simple exponential potentials. The analysis is completely generic and independent of the form of the potential for the scalar field. As an illustration and confirmation of the analysis, we consider f ( z ) of the forms μ/ z {sup 2}, μ/ z , (μ− z )/ z {sup 2} and (μ− z ) to numerically compute the evolution of cosmological parameters with and without coupling. Implications of the approach and the results are discussed.

79 ASTRONOMY AND ASTROPHYSICS↗

Probe microscopy is all you need *

We pose that microscopy offers an ideal real-world experimental environment for the development and deployment of active Bayesian and reinforcement learning methods. Indeed, the tremendous progress achieved by machine learning (ML) and artificial intelligence over the last decade has been largely achieved via the utilization of static data sets, from the paradigmatic MNIST to the bespoke corpora of text and image data used to train large models such as GPT3, DALL·E and others. However, it is now recognized that continuous, minute improvements to state-of-the-art do not necessarily translate to advances in real-world applications. We argue that a promising pathway for the development of ML methods is via the route of domain-specific deployable algorithms in areas such as electron and scanning probe microscopy and chemical imaging. This will benefit both fundamental physical studies and serve as a test bed for more complex autonomous systems such as robotics and manufacturing. Favorable environment characteristics of scanning and electron microscopy include low risk, extensive availability of domain-specific priors and rewards, relatively small effects of exogenous variables, and often the presence of both upstream first principles as well as downstream learnable physical models for both statics and dynamics. Recent developments in programmable interfaces, edge computing, and access to application programming interfaces (APIs) facilitating microscope control, all render the deployment of ML codes on operational microscopes straightforward. We discuss these considerations and hope that these arguments will lead to create novel set of development targets for the ML community by accelerating both real world ML applications and scientific progress.

47 OTHER INSTRUMENTATION↗

Asymptotic behavior of dynamical variables and naked singularity formation in spherically symmetric gravitational collapse

In considering the gravitational collapse of matter, it is an important problem to clarify what kind of conditions leads to the formation of naked singularity. For this purpose, we apply the 1+3 orthonormal frame formalism introduced by Uggla et al. to the spherically symmetric gravitational collapse of a perfect fluid. This formalism allows us to construct an autonomous system of evolution and constraint equations for scale-invariant dynamical variables normalized by the volume expansion rate of the timelike orthonormal frame vector. We investigate the asymptotic evolution of such dynamical variables towards the formation of a central singularity and present a conjecture that the steep spatial gradient for the normalized density function is a characteristic of the naked singularity formation.

79 ASTRONOMY AND ASTROPHYSICS↗

Unified dark energy and dark matter from a scalar field different from quintessence

We explore unification of dark matter and dark energy in a theory containing a scalar field of non-Lagrangian type, obtained by direct insertion of a kinetic term into the energy-momentum tensor. This scalar is different from quintessence, having an equation of state between -1 and 0 and a zero sound speed in its rest frame. We solve the equations of motion for an exponential potential via a rewriting as an autonomous system, and demonstrate the observational viability of the scenario, for sufficiently small exponential potential parameter {lambda}, by comparison to a compilation of kinematical cosmological data.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Background cosmological dynamics in f(R) gravity and observational constraints

In this paper, we carry out a study of viable cosmological models in f(R) gravity at the background level. We use observable parameters like {Omega} and {gamma} to form an autonomous system of equations and show that the models under consideration exhibit two different regimes in their time evolution, namely, a phantom phase followed by a quintessencelike behavior. We employ state finder parameters to emphasize a characteristic discriminative signature of these models.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Cosmological dynamics of a Dirac-Born-Infeld field

We analyze the dynamics of a Dirac-Born-Infeld (DBI) field in a cosmological setup which includes a perfect fluid. Introducing convenient dynamical variables, we show that the evolution equations form an autonomous system when the potential and the brane tension of the DBI field are arbitrary power law or exponential functions of the DBI field. In particular we find scaling solutions can exist when powers of the field in the potential and warp factor satisfy specific relations. A new class of fixed-point solutions are obtained corresponding to points which initially appear singular in the evolution equations, but on closer inspection are actually well defined. In all cases, we perform a phase-space analysis and obtain the late-time attractor structure of the system. Of particular note when considering cosmological perturbations in DBI inflation is a fixed-point solution where the Lorentz factor is a finite large constant and the equation of state parameter of the DBI field is w=-1. Since in this case the speed of sound c{sub s} becomes constant, the solution can be thought to serve as a good background to perturb about.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Asynchronous and Distributed Tracking of Time-Varying Fixed Points

NREL researcher Andrey Bernstein discusses asynchronous and distributed tracking of time-varying fixed points as part of the Innovative Optimization and Control Methods for Highly Distributed Autonomous Systems workshop. NREL hosted the workshop to exchange ideas on state-of-the-art control and optimization strategies as well as get feedback on work being done at NREL.

97 MATHEMATICS AND COMPUTING↗

Evaluation of Autonomous Vehicle Sensing and Compute Load on a Chassis Dynamometer

The sensing and compute load auxiliary energy consumption in autonomous vehicles may be significant due to the large number of sensors and the high compute load from sensor processing and route planning. To understand this issue, this study investigates the top-down energy usage of an electric 2015 Kia Soul fully instrumented with state sensors and a state-specific computer for path planning and sensor processing. A chassis dynamometer was then used to evaluate the cases of (1) no sensors or computation, (2) only sensors operating, and (3) sensors plus compute load. The vehicle was operated autonomously on the dynamometer using a PolySync drive-kit with drive-by-wire longitudinal control. The DynoJet model 224xLC was used to adapt the eddy current dynamometer's road load parameters to comply with an Environmental Protection Agency drive schedule and to evaluate performance against the Argonne National Laboratory Digital Dynamometer Dataset. On the UDDS-HWFET combined driving cycle, the stock battery's range was reduced by 5.6% for sensors alone and 12.2% for sensors and compute load. These results show that the added sensing and compute auxiliary load from automated and autonomous systems is significant and that research efforts need to be spent investigating new energy efficient systems.

Brown, Nicholas E.↗

Meta-optic accelerators for object classifiers

Rapid advances in deep learning have led to paradigm shifts in a number of fields, from medical image analysis to autonomous systems. These advances, however, have resulted in digital neural networks with large computational requirements, resulting in high energy consumption and limitations in real-time decision-making when computation resources are limited. Here, we demonstrate a meta-optic–based neural network accelerator that can off-load computationally expensive convolution operations into high-speed and low-power optics. In this architecture, metasurfaces enable both spatial multiplexing and additional information channels, such as polarization, in object classification. End-to-end design is used to co-optimize the optical and digital systems, resulting in a robust classifier that achieves 93.1% accurate classification of handwriting digits and 93.8% accuracy in classifying both the digit and its polarization state. This approach could enable compact, high-speed, and low-power image and information processing systems for a wide range of applications in machine vision and artificial intelligence.

42 ENGINEERING↗

On the nonlinear dissipative dynamics of weakly overdamped oscillators

We consider the motion of weakly overdamped linear oscillators. Weak overdamping of an oscillator is defined as a slight excess of the damping decrement over its natural frequency. Exact solutions are obtained for a certain relation between the decrement and the natural frequency and qualitatively different regimes of motion are analyzed. The threshold conditions corresponding to changes of regimes are established; one-component models with an arbitrary degree of nonlinearity are analyzed, and quadratic and cubic nonlinearities are considered in detail. If the nonlinearity in a multicomponent model is determined by a homogeneous function, transformations of the Kummer-Liouville type can be reduced to an autonomous system of second-order differential equations in the case when the relation between the decrement and the natural frequency has been established. Some integrable multicomponent models with quadratic and cubic nonlinearities are analyzed.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Demonstrating Cross-Facility Data Processing at Scale with Laue Microdiffraction

In February and April 2023 live, at-scale data processing demonstrations were conducted between the Advanced Photon Source (APS), a synchrotron light source, and the Argonne Leadership Computing Facility (ALCF). These tests were run as part of a novel beamline technique: coded aperture laue micro-diffraction. This technique requires a significant amount of compute to decode appeture patterns embedded in the detector stream. An autonomous system was able to send data to ALCF during an experiment, utilize 50 nodes of the Polaris supercomputer to process 6-12 hour scans, and return the data back to the APS within 12-15 minutes behind the detector. With scan points arriving every 72 seconds, the system kept up with the beamline, potentially enabling in-experiment analysis. The data processing system utilizes Globus infrastructure and an on-demand queue to dynamically acquire nodes on Polaris. The underlying reconstruction algorithms were parallelized via MPI and accelerated with custom CUDA kernels.

Prince, Michael↗

Autonomy for Hypersonics Mission Campaign (A4H)

A4H is developing autonomous hypersonic flight vehicles that can intelligently navigate, guide, and control themselves and home-in on targets. Autonomous systems are characterized by the use of closed-loop SENSE-THINK-ACT operations to achieve their desired goals. Closing the SENSE-THINK-ACT loop onboard these systems will provide an improved ability to engage diverse targets in contested environments.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

Gaussian process for calibration and control of GlueX Central Drift Chamber

We have developed and implemented a machine learning based system to calibrate and control the GlueX Central Drift Chamber at Jefferson Lab, VA, in near real-time. The system monitors environmental and experimental conditions during data taking and uses those as inputs to a Gaussian process (GP) with learned prior. The GP predicts calibration constants in order to recommend a high voltage (HV) setting for the detector that maintains consistent detector performance (gain and resolution) throughout data taking. This approach is in stark contrast to traditional detector operations in which the detector operates at fixed HV and its calibration parameters vary quite considerably with time. Additionally, the ML based system utilizes uncertainty quantification to correct the recommended control parameters when appropriate. We will present results from the ML system autonomously during the Charged Pion Polarizability (CPP) experiment conducted in Hall D at Jefferson Lab.

McSpadden, Helen↗