SEARCH · Engineering Papers
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.
An adaptive neural network for auto-associative memories operating in continuous time is considered. A new learning algorithm for the weight matrix defined by explicit locations of desirable equilibrium points is introduced. The approach is based upon the minimum 'strength energy' of the weight matrix for each prescribed performance of the neural network.
Two neural network architectures are compared with a classical spline interpolation technique for the approximation of functions useful in a nonlinear control system. A standard back-propagation feedforward neural network and a cerebellar model articulation controller (CMAC) neural network are presented, and their results are compared with a B-spline interpolation procedure that is updated using recursive least-squares parameter identification. Each method is able to accurately represent a one-dimensional test function. Tradeoffs between size requirements, speed of operation, and speed of learning indicate that neural networks may be practical for identification and adaptation in a nonlinear control environment.
A prototype system employing a genetic algorithm (GA) has been developed to support the scheduling of the Hubble Space Telescope. A non-standard knowledge structure is used and appropriate genetic operators have been created. Several different crossover styles (random point selection, evolving points, and smart point selection) are tested and the best GA is compared with a neural network (NN) based optimizer. The smart crossover operator produces the best results and the GA system is able to evolve complete schedules using it. The GA is not as time-efficient as the NN system and the NN solutions tend to be better.
Increased complexity of space missions during the 1980s led to the introduction of expert systems and advanced automation techniques in mission operations. This paper describes several technologies in operational use or under development at the National Aeronautics and Space Administration's Goddard Space Flight Center. Several expert systems are described that diagnose faults, analyze spacecraft operations and onboard subsystem performance (in conjunction with neural networks), and perform data quality and data accounting functions. The design of customized user interfaces is discussed, with examples of their application to space missions. Displays, which allow mission operators to see the spacecraft position, orientation, and configuration under a variety of operating conditions, are described. Automated systems for scheduling are discussed, and a testbed that allows tests and demonstrations of the associated architectures, interface protocols, and operations concepts is described. Lessons learned are summarized.
Treatment of neurological anomalies, places stringent demands on device functionality and size. A micro-spectrometer has been developed for use as an implantable neural probe to monitor neuro-chemistry in synapses. The microspectrometer, based on a NASA-invented miniature Fresnel grating, is capable of differentiating the emission spectra from various brain tissues. The micro-spectrometer meets the size requirements, and is able to probe the neuro-chemistry and suppression voltage typically associated with a neural anomaly. This neural probe-pin device (PPD) is equipped with wireless power technology (WPT) enabling operation in a continuous manner without requiring an implanted battery. The implanted neural PPD, together with a neural electronics interface and WPT, allow real-time measurement and control/feedback for remediation of neural anomalies. The design and performance of the combined PPD/WPT device for monitoring dopamine in a rat brain will be presented to demonstrate the current level of development. Future work on this device will involve the addition of an embedded expert system capable of performing semi-autonomous management of neural functions through a routine of sensing, processing, and control.
A closed-loop optimal neural-network controller technique was developed to optimize rotorcraft aeromechanical behaviour. This technique utilities a neural-network scheme to provide a general non-linear model of the rotorcraft. A modem constrained optimisation method is used to determine and update the constants in the neural-network plant model as well as to determine the optimal control vector. Current data is read, weighted, and added to a sliding data window. When the specified maximum number of data sets allowed in the data window is exceeded, the oldest data set is and the remaining data sets are re-weighted. This procedure provides at least four additional degrees-of-freedom in addition to the size and geometry of the neural-network itself with which to optimize the overall operation of the controller. These additional degrees-of-freedom are: 1. the maximum length of the sliding data window, 2. the frequency of neural-network updates, 3. the weighting of the individual data sets within the sliding window, and 4. the maximum number of optimisation iterations used for the neural-network updates.
The objectives of this project were to demonstrate applicability and advantages of a neural network approach for evaluating the performance of an electro-mechanical actuator (EMA). The EMA in question was intended for the X-37 Advanced Technology Vehicle. It will have redundant components for safety and reliability. The neural networks for this application are to monitor the operation of the redundant electronics that control the actuator in real time and decide on the operating configuration. The system we proposed consists of the actuator, sensors, control circuitry and dedicated (embedded) processors. The main purpose of the study was to develop suitable hardware and neural network capable of allowing real time reconfiguration decisions to be made. This approach was to be compared to other methods such as fuzzy logic and knowledge based systems considered for the same application. Over the course of the project a more general objective was the identification of the other neural network applications and the education of interested NASA personnel on the topic of Neural Networks.
A new methodology for faster supervised temporal learning in nonlinear neural networks is presented which builds upon the concept of adjoint operators to allow fast computation of the gradients of an error functional with respect to all parameters of the neural architecture, and exploits the concept of teacher forcing to incorporate information on the desired output into the activation dynamics. The importance of the initial or final time conditions for the adjoint equations is discussed. A new algorithm is presented in which the adjoint equations are solved simultaneously (i.e., forward in time) with the activation dynamics of the neural network. We also indicate how teacher forcing can be modulated in time as learning proceeds. The results obtained show that the learning time is reduced by one to two orders of magnitude with respect to previously published results, while trajectory tracking is significantly improved. The proposed methodology makes hardware implementation of temporal learning attractive for real-time applications.
ANTS (Autonomous Nano-Technology Swarm), a large (1000 member) swarm of nano to picoclass (10 to 1 kg) totally autonomous spacecraft, are being developed as a NASA advanced mission concept. ANTS, based on a hierarchical insect social order, use an evolvable, self-similar, hierarchical neural system in which individual spacecraft represent the highest level nodes. ANTS uses swarm intelligence attained through collective, cooperative interactions of the nodes at all levels of the system. At the highest levels this can take the form of cooperative, collective behavior among the individual spacecraft in a very large constellation. The ANTS neural architecture is designed for totally autonomous operation of complex systems including spacecraft constellations. The ANTS (Autonomous Nano Technology Swarm) concept has a number of possible applications. A version of ANTS designed for surveying and determining the resource potential of the asteroid belt, called PAM (Prospecting ANTS Mission), is examined here.
A method for modeling error-driven adaptive control of an aircraft. Normal aircraft plant dynamics is modeled, using an original plant description in which a controller responds to a tracking error e(k) to drive the component to a normal reference value according to an asymptote curve. Where the system senses that (1) at least one aircraft plant component is experiencing an excursion and (2) the return of this component value toward its reference value is not proceeding according to the expected controller characteristics, neural network (NN) modeling of aircraft plant operation may be changed. However, if (1) is satisfied but the error component is returning toward its reference value according to expected controller characteristics, the NN will continue to model operation of the aircraft plant according to an original description.
Since the 1970s, most airlines have incorporated computerized support for managing disruptions during flight schedule execution. However, existing platforms for airline disruption management (ADM) employ monolithic system design methods that rely on the creation of specific rules and requirements through explicit optimization routines, before a system that meets the specifications is designed. Thus, current platforms for ADM are unable to readily accommodate additional system complexities resulting from the introduction of new capabilities, such as the introduction of unmanned aerial systems (UAS), operations and infrastructure, to the system. To this end, we use historical data on airline scheduling and operations recovery to develop a system of artificial neural networks (ANNs), which describe a predictive transfer function model (PTFM) for promptly estimating the recovery impact of disruption resolutions at separate phases of flight schedule execution during ADM. Furthermore, we provide a modular approach for assessing and executing the PTFM by employing a parallel ensemble method to develop generative routines that amalgamate the system of ANNs. Our modular approach ensures that current industry standards for tardiness in flight schedule execution during ADM are satisfied, while accurately estimating appropriate time-based performance metrics for the separate phases of flight schedule execution.
Automatic docking of orbiting spacecraft is a crucial operation involving the identification of vehicle orientation as well as complex approach dynamics. The chaser spacecraft must be able to recognize the target spacecraft within a scene and achieve accurate closing maneuvers. In a video-based system, a target scene must be captured and transformed into a pattern of pixels. Successful recognition lies in the interpretation of this pattern. Due to their powerful pattern recognition capabilities, artificial neural networks offer a potential role in interpretation and automatic docking processes. Neural networks can reduce the computational time required by existing image processing and control software. In addition, neural networks are capable of recognizing and adapting to changes in their dynamic environment, enabling enhanced performance, redundancy, and fault tolerance. Most neural networks are robust to failure, capable of continued operation with a slight degradation in performance after minor failures. This paper discusses the particular automatic docking tasks neural networks can perform as viable alternatives to conventional techniques.
An artificial neural network is disclosed that processes holography generated characteristic pattern of vibrating structures along with finite-element models. The present invention provides for a folding operation for conditioning training sets for optimally training forward-neural networks to process characteristic fringe pattern. The folding pattern increases the sensitivity of the feed-forward network for detecting changes in the characteristic pattern The folding routine manipulates input pixels so as to be scaled according to the location in an intensity range rather than the position in the characteristic pattern.
Two optical implementations of production systems are advanced. The production systems operate on a knowledge base where facts and rules are encoded as formulas in propositional calculus. The first implementation is a binary neural network. An analog neural network is used to include reasoning with uncertainties. The second implementation uses a new optical symbolic substitution correlator. This implementation is useful when a set of similar situations has to be handled in parallel on one processor.
Geostationary satellite imagers, such as those of the Geostationary Operational Environmental Satellite (GOES) series, have been observing severe convection at 15–60-minute intervals for over 40 years. When properly assessed, such a data record can be valuable in efforts of estimating severe storm risk throughout the diurnal cycle based on automated detection of patterns consistently found atop severe storms. Furthermore, environmental conditions favorable for severe weather are well-known and are thought to be represented well by modern reanalysis products. Promoting resilience against such hazards on local and global scales is a chief goal the NASA Disasters program, which seeks to encourage use of satellite observations to mitigate risk. For instance, hail is the costliest severe weather hazard across the globe in terms of insured loss, but reporting inconsistencies for hail events globally make it difficult to develop models that can quantify the risk. Satellite observation and model reanalysis taken together have the potential to, with reasonable skill and specificity, characterize environmental conditions that are favorable for hazardous weather, and thereby enable creation of hazard climatologie. Such climatologies are particularly useful over regions without extensive radar networks or storm reporting. By mapping the multivariate combination of observed cloud features and reanalysis environmental parameters/indices to United States Next Generation Weather Radar (NEXRAD) radar-estimated Maximum Expected Size of Hail (MESH) by way of a deep neural network (DNN), estimates of likelihood for potentially severe hail can be produced. Such estimates are of greater complexity and efficiency than could be performed with previous multivariate or logistic regression analyses for observed points within convective systems. Statistical distributions of convective parameters from satellite and reanalysis are shown to highlight non-severe/severe class separation for well-known hailstorm predictors, e.g., overshooting cloud top characteristics, deep-layer wind shear, mid-level stability, helicity, and convective inhibition. These complex, multivariate predictor relationships are exploited within a DNN, which can efficiently produce a quantitative hail risk metric with better than 70% detection rate and under 30% false alarms. These hail classifications can then be aggregated across the satellite record to yield a hazard climatology for hail frequency and severity – knowledge of which is of particular interest to those who manage risk (e.g., insurers) and are seeking opportunities to identify hail-prone regions, particularly in developing nations. This NASA study uses satellite observations and model parameters in a DNN to perform climatological hailstorm analysis in support of catastrophe model development, with the hope of promoting risk resilience particularly in regions without adequate weather radar coverage.
High resistances prevent hotspots in parallel input and output operation. Nonvolatile computer memory combines ultrahigh storage density with extremely-low power dissipation. Accommodates about 1 billion bits in square centimeter of surface area. Bit written with expenditure of less than 1 nanojoule of energy and read with even lower energy. Developed for parallel input and output operation.
The feasibility of real-time processing schemes using artificial neural networks (ANNs) is investigated. A rationale for digital neural nets is presented and a general processor architecture for control applications is illustrated. Research results on ANN structures for real-time applications are given. Research results on ANN algorithms for real-time control are also shown.
Large amount of small Unmanned Aerial Vehicles (sUAVs) are projected to operate in the near future. Potential sUAV applications include, but not limited to, search and rescue, inspection and surveillance, aerial photography and video, precision agriculture, and parcel delivery. sUAVs are expected to operate in the uncontrolled Class G airspace, which is at or below 500 feet above ground level (AGL), where many static and dynamic constraints exist, such as ground properties and terrains, restricted areas, various winds, manned helicopters, and conflict avoidance among sUAVs. How to enable safe, efficient, and massive sUAV operations at the low altitude airspace remains a great challenge. NASA's Unmanned aircraft system Traffic Management (UTM) research initiative works on establishing infrastructure and developing policies, requirement, and rules to enable safe and efficient sUAVs' operations. To achieve this goal, it is important to gain insights of future UTM traffic operations through simulations, where the accurate trajectory model plays an extremely important role. On the other hand, like what happens in current aviation development, trajectory modeling should also serve as the foundation for any advanced concepts and tools in UTM. Accurate models of sUAV dynamics and control systems are very important considering the requirement of the meter level precision in UTM operations. The vehicle dynamics are relatively easy to derive and model, however, vehicle control systems remain unknown as they are usually kept by manufactures as a part of intellectual properties. That brings challenges to trajectory modeling for sUAVs. How to model the vehicle's trajectories with unknown control system? This work proposes to use a neural network to model a vehicle's trajectory. The neural network is first trained to learn the vehicle's responses at numerous conditions. Once being fully trained, given current vehicle states, winds, and desired future trajectory, the neural network should be able to predict the vehicle's future states at next time step. A complete 4-D trajectory are then generated step by step using the trained neural network. Experiments in this work show that the neural network can approximate the sUAV's model and predict the trajectory accurately.