SEARCH · Search NASA

Results for “brain circuits”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

81 records · Page 5

All sources NASA NTRS DOE OSTI

NASA NTRS2005

Insect-Inspired Optical-Flow Navigation Sensors

Integrated circuits that exploit optical flow to sense motions of computer mice on or near surfaces ( optical mouse chips ) are used as navigation sensors in a class of small flying robots now undergoing development for potential use in such applications as exploration, search, and surveillance. The basic principles of these robots were described briefly in Insect-Inspired Flight Control for Small Flying Robots (NPO-30545), NASA Tech Briefs, Vol. 29, No. 1 (January 2005), page 61. To recapitulate from the cited prior article: The concept of optical flow can be defined, loosely, as the use of texture in images as a source of motion cues. The flight-control and navigation systems of these robots are inspired largely by the designs and functions of the vision systems and brains of insects, which have been demonstrated to utilize optical flow (as detected by their eyes and brains) resulting from their own motions in the environment. Optical flow has been shown to be very effective as a means of avoiding obstacles and controlling speeds and altitudes in robotic navigation. Prior systems used in experiments on navigating by means of optical flow have involved the use of panoramic optics, high-resolution image sensors, and programmable imagedata- processing computers.

Thakoor, Sarita↗

DOE OSTI2022

Complex Oxides for Brain–Inspired Computing: A Review

The fields of brain-inspired computing, robotics, and, more broadly, artificial intelligence (AI) seek to implement knowledge gleaned from the natural world into human-designed electronics and machines. In this review, the opportunities presented by complex oxides, a class of electronic ceramic materials whose properties can be elegantly tuned by doping, electron interactions, and a variety of external stimuli near room temperature, are discussed. The review begins with a discussion of natural intelligence at the elementary level in the nervous system, followed by collective intelligence and learning at the animal colony level mediated by social interactions. An important aspect highlighted is the vast spatial and temporal scales involved in learning and memory. The focus then turns to collective phenomena, such as metal-to-insulator transitions (MITs), ferroelectricity, and related examples, to highlight recent demonstrations of artificial neurons, synapses, and circuits and their learning. First-principles theoretical treatments of the electronic structure, and in situ synchrotron spectroscopy of operating devices are then discussed. The implementation of the experimental characteristics into neural networks and algorithm design is then revewed. Finally, outstanding materials challenges that require a microscopic understanding of the physical mechanisms, which will be essential for advancing the frontiers of neuromorphic computing, are highlighted.

36 MATERIALS SCIENCE↗

DOE OSTI2025

Design of Hopfield Networks Based on Superconducting Coupled Oscillators

The global energy shortage has driven the development of many energy-efficient computational platforms beyond Moore's law, among which brain-inspired neuromorphic computing is one of the promising solutions. Associative memory and pattern recognition are important computations solved by brain-inspired Hopfield networks. Classical Hopfield networks store memories via fixed point attractors of their dynamics. In oscillatory Hopfield networks, these attractors are replaced by periodic orbits. Here, we design an oscillatory Hopfield network based on coupled superconducting oscillators. We first employ a mathematical phase reduction approach to map networks of coupled superconducting rapid single flux quantum (RSFQ) ring oscillators to coupled Kuramoto phase-oscillator networks. We use this theory to numerically optimize the hardware's mutual inductances in order to directly match the phase-reduced superconducting oscillators to a model of phase-oscillator-based Hopfield networks. The resulting network can store multiple oscillatory phase-locked memory patterns and recover the patterns based on the initial phase conditions. As different pattern recognition tasks, or learning, require tunable connectivity strengths between the oscillatory nodes, we further employ a coupler circuit that enables tuning the coupling strength between two oscillators by applying an external flux. We demonstrate the functionality of our design through numerical simulations of a small example network with oscillators operating at 86 GHz and recognizing patterns within 10 ns. Our approach enables the learning and retrieval of dynamical memory patterns with a wide range of applications where rhythmic dynamic output is beneficial.

Cheng, Ran↗

DOE OSTI2022

Neuromorphic Computing is Turing-Complete

Neuromorphic computing is a non-von Neumann computing paradigm that performs computation by emulating the human brain. Neuromorphic systems are extremely energy-efficient and known to consume thousands of times less power than CPUs and GPUs. They have the potential to drive critical use cases such as autonomous vehicles, edge computing and internet of things in the future. For this reason, they are sought to be an indispensable part of the future computing landscape. Neuromorphic systems are mainly used for spike-based machine learning applications, although there are some non-machine learning applications in graph theory, differential equations, and spike-based simulations. These applications suggest that neuromorphic computing might be capable of general-purpose computing. However, general-purpose computability of neuromorphic computing has not been established yet. In this work, we prove that neuromorphic computing is Turing-complete and therefore capable of general-purpose computing. Specifically, we present a model of neuromorphic computing, with just two neuron parameters (threshold and leak), and two synaptic parameters (weight and delay). We devise neuromorphic circuits for computing all the μ-recursive functions (i.e., constant, successor and projection functions) and all the μ-recursive operators (i.e., composition, primitive recursion and minimization operators). Given that the μ-recursive functions and operators are precisely the ones that can be computed using a Turing machine, this work establishes the Turing-completeness of neuromorphic computing.

Date, Prasanna↗

NASA NTRS2003

Organization of projections from the raphe nuclei to the vestibular nuclei in rats

Previous anatomic and electrophysiological evidence suggests that serotonin modulates processing in the vestibular nuclei. This study examined the organization of projections from serotonergic raphe nuclei to the vestibular nuclei in rats. The distribution of serotonergic axons in the vestibular nuclei was visualized immunohistochemically in rat brain slices using antisera directed against the serotonin transporter. The density of serotonin transporter-immunopositive fibers is greatest in the superior vestibular nucleus and the medial vestibular nucleus, especially along the border of the fourth ventricle; it declines in more lateral and caudal regions of the vestibular nuclear complex. After unilateral iontophoretic injections of Fluoro-Gold into the vestibular nuclei, retrogradely labeled neurons were found in the dorsal raphe nucleus (including the dorsomedial, ventromedial and lateral subdivisions) and nucleus raphe obscurus, and to a minor extent in nucleus raphe pallidus and nucleus raphe magnus. The combination of retrograde tracing with serotonin immunohistofluorescence in additional experiments revealed that the vestibular nuclei receive both serotonergic and non-serotonergic projections from raphe nuclei. Tracer injections in densely innervated regions (especially the medial and superior vestibular nuclei) were associated with the largest numbers of Fluoro-Gold-labeled cells. Differences were observed in the termination patterns of projections from the individual raphe nuclei. Thus, the dorsal raphe nucleus sends projections that terminate predominantly in the rostral and medial aspects of the vestibular nuclear complex, while nucleus raphe obscurus projects relatively uniformly throughout the vestibular nuclei. Based on the topographical organization of raphe input to the vestibular nuclei, it appears that dense projections from raphe nuclei are colocalized with terminal fields of flocculo-nodular lobe and uvula Purkinje cells. It is hypothesized that raphe-vestibular connections are organized to selectively modulate processing in regions of the vestibular nuclear complex that receive input from specific cerebellar zones. This represents a potential mechanism whereby motor activity and behavioral arousal could influence the activity of cerebellovestibular circuits.

NASA Discipline Neuroscience↗

NASA NTRS2008

Hand-Held Units for Short-Range Wireless Biotelemetry

Special-purpose hand-held radiotransceiver units have been proposed as means of short-range radio powering and interrogation of surgically implanted microelectromechanical sensors and actuators. These units are based partly on the same principles as those of the units described in "Printed Multi- Turn Loop Antennas for RF Biotelemetry" (LEW-17879-1), NASA Tech Briefs, Vol. 31, No. 6 (June 2007), page 48. Like the previously reported units, these units would make it unnecessary to have wire connections between the implanted devices and the external equipment used to activate and interrogate them. Like a unit of the previously reported type, a unit of the type now proposed would include a printed-circuit antenna on a dielectric substrate. The antenna circuitry would include integrated surface-mount inductors for impedance tuning. Circuits for processing the signals transmitted and received by the antenna would be included on the substrate. During operation, the unit would be positioned near (but not in electrical contact with) a human subject, in proximity to a microelectromechanical sensor or actuator that has been surgically implanted in the subject. It has been demonstrated that significant electromagnetic coupling with an implanted device could be established at a distance of as much as 4 in. (.10 cm). During operation in the interrogation mode, the antenna of the unit would receive a radio telemetry signal transmitted by the surgically implanted device. The antenna substrate would have dimensions of approximately 3.25 by 3.75 inches (approximately 8.3 by 9.5 cm). The substrate would have a thickness of the order of 30 mils (of the order of a somewhat less than a millimeter). The substrate would be made of low-radiofrequency- loss dielectric material that could be, for example, fused quartz, alumina, or any of a number of commercially available radio-frequency dielectric composite materials. The antenna conductors would typically be made of copper or a combination of chromium and gold. The choice of metal and the thickness of the metal layer(s) would depend on the choice of substrate material. For example, on a quartz or alumina substrate, one would typically use a layer of chromium 150 A thick and a layer of gold 2 m thick. The proposed units and the implanted devices that they would interrogate or activate would be inherently safe to use. They would operate at low radiated-power levels for short interrogation times (typically, milliseconds). Hence, there would be little local heating of tissues surrounding the implanted devices and little absorption of radio energy by such sensitive body parts as the eyes and the brain. Because the implanted devices would not depend on battery power and would be activated only during short interrogation intervals and would otherwise be in the goff h state most of the time, the useful lifetimes of the implanted devices would be greater than those of comparable battery-powered implanted devices. The compactness of the hand-held transceiver units would facilitate transport and storage and would facilitate self-diagnosis by patients able to handle the units while away from medical facilities.

Miranda, Felix A.↗

DOE OSTI2022

Lithium-Ion Batteries with Safer Current Collectors

The US Department of Energy’s Oak Ridge National Laboratory and Soteria Battery Innovation Group collaborated to develop a metallized polymer film as a current collector for lithium-ion batteries and design slit patterns for electrodes to improve battery safety. The metallized polymer film serves as a fuse that will break under high temperature resulting from a short circuit. Consequently, the short circuit will be broken and the heat generated from the short circuit will be alleviated to avoid thermal runaway. The electrodes with slit patterns are expected to be broken into small segments upon mechanical impact, which can isolate the damaged electrodes from the rest. Thus, the heat generation associated with the mechanical impact will be reduced, along with the likelihood of thermal runaway. This effort aimed to improve lithium-ion battery safety by replacing traditional metal foils with metallized polymer films as current collector and introducing slit patterns to battery electrodes. The metallized polymer films were polyethylene terephthalate with a thin metal (aluminum for cathodes and copper for anodes) coating on both sides. Slit patterns with various geometries and dimensions were designed via simulation. Temperature distribution on cells with electrodes with and without slit patterns was also simulated via nemerical modeling. Electrodes on traditional metal foils and metallized polymer films were coated via a pilot-scale slot-die coater at the US Department of Energy Battery Manufacturing Facility at Oak Ridge National Laboratory. The electrodes were calendered to 35% porosity and punched out with and without slit patterns. Pouch cells were assembed with the electrodes inside a dry room for electrochemical and safety testing. Electrodes coated on metallized polymer films exhibited comparable electrochemical performance to their counterparts coated on traditional metal foils. They also showed improved safety in nail penetration and indentation tests.

25 ENERGY STORAGE↗

DOE OSTI2026

Harnessing the Second-Order Metal−Insulator Transition for Neuromorphic Computing

Vanadium oxides are widely studied phase change materials for brain-inspired computing architectures. Systems like VO 2 and V 2 O 3 exhibit first-order metal−insulator transitions (MITs) with hysteresis and percolative switching, increasing stochasticity and device variability. Here, we focus on the less explored Magnéli phase V 4 O 7 , which undergoes a continuous, non-hysteretic, second-order MIT. This surprisingly enables highly reproducible volatile resistive switching in spiking-neuron-type devices. We synthesize V 4 O 7 films, characterize their structural and transport properties, and demonstrate voltage and current-driven threshold switching with electrothermal feedback. In a Pearson–Anson oscillator, V 4 O 7 devices produce stable, tunable spiking across 20–200 kHz, with consistent operation among multiple devices. We introduce a numerical analog leaky-integrate-and-fire (aLIF) model that captures waveform shapes and their dependence on load resistance, temperature, and voltage. Furthermore, these findings suggest that second-order MIT materials like V 4 O 7 are promising for deterministic, scalable spiking neuron arrays for neuromorphic computing.

V4O7↗

DOE OSTI2025

Introduction: Neuromorphic Materials

The explosive growth in data collection and the need to process it efficiently, as well as the desire to automate increasingly complex tasks in transportation, medical care, manufacturing, security and many other fields have motivated a growing interest in neuromorphic computing. Unlike the binary, transistorbased ON/OFF logic gates and separate logic and memory functionalities employed in digital computing, neuromorphic computing is inspired by animal brains that use interconnected synapses and neurons to perform processing, storage and transmission of information at the same location, while only consuming ~20 W or less of power. Motivated by the brain’s efficiency, adaptability, self-learning and resiliency qualities, neuromorphic computing can be broadly defined as an approach to processing and storing information using hardware and algorithms inspired by models of biological neural systems. Present research in neuromorphic computing encompasses approaches that vary significantly in their degree of neuro-inspiration, from systems that only incorporate features such as asynchronous, event-driven operation or use crossbar arrays of non-volatile memory (NVM) elements to accelerate deep neural networks (DNNs), to designs that embrace the extreme parallelism, sparsity, reconfigurability, adaptability, complexity and stochasticity observed in nervous systems. The term ‘neuromorphic’ computing is often credited to Carver Mead, who in the 1980s investigated Si-based analog electronics to replicate functions of the animal retina. Earlier important advances in this field include the work of Frank Rosenblatt, who proposed the concept of the perceptron, Bernard Widrow, who used this concept to build one of the first analog neural networks, the Adaline and many other researchers (see ref. 6 for an historical perspective on neuromorphic computing). With the recent increase in the use of artificial intelligence and large language models, and rising concerns over the associated energy costs, interest in neuromorphic hardware has expanded rapidly. According to some estimates, driven largely by the drastic growth in the training use of artificial intelligence (AI) models using the current computing architectures, the energy cost of computing is projected to reach the energy supply worldwide by 2045. Furthermore, while this is not a realistic outcome, it means that, if more efficient computing technologies are not developed -- soon -- the world will soon become one where demand for energy and market constraints limit the continued increase of societal access to AI and cloud services from data centers. Data centers used for training and use of these models consume hundreds of terawatt hours of electricity, already past 4% of the US electricity demand.

Circuits↗