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.
The design of a defect-tolerant control circuit for a set-associative cache memory is presented. The circuit maintains the stack ordering necessary for implementing the Least Recently Used (LRU) replacement algorithm. A discussion of programming techniques for bypassing defective blocks is included.
Hyper-redundant spatial manipulators possess fault-tolerant features because of their redundant structure. The kinematic control of these manipulators is investigated with special emphasis on fault-tolerant control. The manipulator tasks are viewed in the end-effector space while actuator commands are in joint-space, requiring an inverse kinematic algorithm to generate joint-angle commands from the end-effector ones. The rate-inverse kinematic control algorithm presented in this paper utilizes the pseudoinverse to accommodate for joint motor failures. An optimal scale factor for the robust inverse is derived.
The Optical Plume Anomaly Detection (OPAD) detects engine hardware degradation of flight vehicles through identification and quantification of elemental species found in the plume by analyzing the plume emission spectra in a real-time mode. Real-time performance of OPAD relies on extensive software which must report metal amounts in the plume faster than once every 0.5 sec. OPAD software previously written by NASA scientists performed most necessary functions at speeds which were far below what is needed for real-time operation. The research presented in this report improved the execution speed of the software by optimizing the code without changing the algorithms and converting it into a parallelized form which is executed in a shared-memory multiprocessor system. The resulting code was subjected to extensive timing analysis. The report also provides suggestions for further performance improvement by (1) identifying areas of algorithm optimization, (2) recommending commercially available multiprocessor architectures and operating systems to support real-time execution and (3) presenting an initial study of fault-tolerance requirements.
Quantum error correction provides a route to realizing large-scale quantum computation but incurs substantial resource overheads. Here, in this work, we highlight recent advances that reduce these overheads by co-designing different levels of the computational stack, including algorithms, quantum-error-correction strategies and hardware architecture. We then discuss opportunities for further optimization such as leveraging flexible qubit connectivity and quantum low-density parity check codes. These strategies can bring useful quantum computation closer to reality as experiments advance in the coming years.
An architecture has been developed for a high-performance VLSI digital signal processor that is highly reliable, fault-tolerant, and radiation-hard. The signal processor, part of a spacecraft receiver designed to support uplink radio science experiments at the outer planets, organizes the connections between redundant arithmetic resources, register files, and memory through a shuffle exchange communication network. The configuration of the network and the state of the processor resources are all under microprogram control, which both maps the resources according to algorithmic needs and reconfigures the processing should a failure occur. In addition, the microprogram is reloadable through the uplink to accommodate changes in the science objectives throughout the course of the mission. The processor will be implemented with silicon compiler tools, and its design will be verified through silicon compilation simulation at all levels from the resources to full functionality. By blending reconfiguration with redundancy the processor implementation is fault-tolerant and reliable, and possesses the long expected lifetime needed for a spacecraft mission to the outer planets.
Standard deep learning algorithms require differentiating large nonlinear networks, a process that is slow and power-hungry. Electronic contrastive local learning networks (CLLNs) offer potentially fast, efficient, and fault-tolerant hardware for analog machine learning, but existing implementations are linear, severely limiting their capabilities. These systems differ significantly from artificial neural networks as well as the brain, so the feasibility and utility of incorporating nonlinear elements have not been explored. Here, we introduce a nonlinear CLLN—an analog electronic network made of self-adjusting nonlinear resistive elements based on transistors. We demonstrate that the system learns tasks unachievable in linear systems, including XOR (exclusive or) and nonlinear regression, without a computer. We find our decentralized system reduces modes of training error in order (mean, slope, curvature), similar to spectral bias in artificial neural networks. The circuitry is robust to damage, retrainable in seconds, and performs learned tasks in microseconds while dissipating only picojoules of energy across each transistor. This suggests enormous potential for fast, low-power computing in edge systems like sensors, robotic controllers, and medical devices, as well as manufacturability at scale for performing and studying emergent learning.
Certification trails are a recently introduced and promising approach to fault-detection and fault-tolerance. Recent experimental work reveals many cases in which a certification-trail approach allows for significantly faster program execution time than a basic time-redundancy approach. Algorithms for answer-validation of abstract data types allow a certification trail approach to be used for a wide variety of problems. An attempt to assess the performance of algorithms utilizing certification trails on abstract data types is reported. Specifically, this method was applied to the following problems: heapsort, Hullman tree, shortest path, and skyline. Previous results used certification trails specific to a particular problem and implementation. The approach allows certification trails to be localized to 'data structure modules,' making the use of this technique transparent to the user of such modules.
The modeling and design of a fault-tolerant multiprocessor system is addressed. Of interest is the behavior of the system during recovery and restoration after a fault has occurred. The multiprocessor systems are based on the Algorithm to Architecture Mapping Model (ATAMM) and the fault considered is the death of a processor. The developed model is useful in the determination of performance bounds of the system during recovery and restoration. The performance bounds include time to recover from the fault, time to restore the system, and determination of any permanent delay in the input to output latency after the system has regained steady state. Implementation of an ATAMM based computer was developed for a four-processor generic VHSIC spaceborne computer (GVSC) as the target system. A simulation of the GVSC was also written on the code used in the ATAMM Multicomputer Operating System (AMOS). The simulation is used to verify the new model for tracking the propagation of the delay through the system and predicting the behavior of the transient state of recovery and restoration. The model is shown to accurately predict the transient behavior of an ATAMM based multicomputer during recovery and restoration.
An error propagation model has been developed for multimodule computing systems in which the main parameters are the distribution functions of error propagation times. A digraph model is used to represent a multimodule computing system, and error propagation in the system is modeled by general distributions of error propagation times between all pairs of modules. Two algorithms are developed to compute systematically and efficiently the distributions of error propagation times. Experiments are also conducted to measure the distributions of error propagation times with the fault-tolerant microprocessor (FTMP). Statistical analysis of experimental data shows that the error propagation times in FTMP do not follow a well-known distribution, thus justifying the use of general distributions in the present model.
A method and apparatus for achieving fault tolerance in a computer system having at least a first central processing unit and a second central processing unit. The method comprises the steps of first executing a first algorithm in the first central processing unit on input which produces a first output as well as a certification trail. Next, executing a second algorithm in the second central processing unit on the input and on at least a portion of the certification trail which produces a second output. The second algorithm has a faster execution time than the first algorithm for a given input. Then, comparing the first and second outputs such that an error result is produced if the first and second outputs are not the same. The step of executing a first algorithm and the step of executing a second algorithm preferably takes place over essentially the same time period.
Not Available
A new Architecture Design and Assessment System (ADAS) tool package is introduced, and a range of possible applications is illustrated. ADAS was used to evaluate the performance of an advanced fault-tolerant computer architecture in a modern flight control application. Bottlenecks were identified and possible solutions suggested. The tool was also used to inject faults into the architecture and evaluate the synchronization algorithm, and improvements are suggested. Finally, ADAS was used as a front end research tool to aid in the design of reconfiguration algorithms in a distributed array architecture.
A novel symmetrically skewed configuration for an attitude reference system (ARS) using three dynamically tuned gyros (DTGs) is developed. Simple schemes for autonomous detection and identification of a faulty DTG in real time and subsequent reconfiguration of the attitude estimation algorithm are proposed. The performance of the present configuration is shown to be better than that of configurations proposed earlier, and it is shown to have better features. It tolerates all types of failures of DTG failures, requires very simple computations, and gives less error in attitude estimate than the other configurations.
In a fault-tolerant parallel computer, a functional programming model can facilitate distributed checkpointing, error recovery, load balancing, and graceful degradation. Such a model has been implemented on the Draper Fault-Tolerant Parallel Processor (FTPP). When used in conjunction with the FTPP's fault detection and masking capabilities, this implementation results in a graceful degradation of system performance after faults. Three graceful degradation algorithms have been implemented and are presented. A user interface has been implemented which requires minimal cognitive overhead by the application programmer, masking such complexities as the system's redundancy, distributed nature, variable complement of processing resources, load balancing, fault occurrence and recovery. This user interface is described and its use demonstrated. The applicability of the functional programming style to the Activation Framework, a paradigm for intelligent systems, is then briefly described.
Fault detection and isolation for advanced rocket engine controllers are discussed focusing on advanced sensing systems and software which significantly improve component failure detection for engine safety and health management. Aerojet's Space Transportation Main Engine controller for the National Launch System is the state of the art in fault tolerant engine avionics. Health management systems provide high levels of automated fault coverage and significantly improve vehicle delivered reliability and lower preflight operations costs. Key technologies, including the sensor data validation algorithms and flight capable spectrometers, have been demonstrated in ground applications and are found to be suitable for bridging programs into flight applications.
We propose a novel floating-point encoding scheme that builds on prior work involving fixed-point encodings. We encode floating-point numbers using Two's Complement fixed-point mantissas and Two's Complement integral exponents. We used our proposed approach to develop quantum algorithms for fundamental arithmetic operations, such as bit-shifting, reciprocation, multiplication, and addition. We prototyped and investigated the performance of the floating-point encoding scheme on quantum computer simulations by performing reciprocation on randomly drawn inputs and by solving first-order ordinary differential equations, while varying the number of qubits in the encoding. We observed rapid convergence to the exact solutions as we increased the number of qubits and a significant reduction in the number of ancilla qubits required for reciprocation when compared with similar approaches.
In this paper, an enhanced on-line diagnostic system which utilizes dual-channel sensor measurements is developed for the aircraft engine application. The enhanced system is composed of a nonlinear on-board engine model (NOBEM), the hybrid Kalman filter (HKF) algorithm, and fault detection and isolation (FDI) logic. The NOBEM provides the analytical third channel against which the dual-channel measurements are compared. The NOBEM is further utilized as part of the HKF algorithm which estimates measured engine parameters. Engine parameters obtained from the dual-channel measurements, the NOBEM, and the HKF are compared against each other. When the discrepancy among the signals exceeds a tolerance level, the FDI logic determines the cause of discrepancy. Through this approach, the enhanced system achieves the following objectives: 1) anomaly detection, 2) component fault detection, and 3) sensor fault detection and isolation. The performance of the enhanced system is evaluated in a simulation environment using faults in sensors and components, and it is compared to an existing baseline system.
The goal of the Autonomous Propulsion System Technology (APST) project is to reduce pilot workload under both normal and anomalous conditions. Ongoing work under APST develops and leverages technologies that provide autonomous engine monitoring, diagnosing, and controller adaptation functions, resulting in an integrated suite of algorithms that maintain the propulsion system's performance and safety throughout its life. Engine-to-engine performance variation occurs among new engines because of manufacturing tolerances and assembly practices. As an engine wears, the performance changes as operability limits are reached. In addition to these normal phenomena, other unanticipated events such as sensor failures, bird ingestion, or component faults may occur, affecting pilot workload as well as compromising safety. APST will adapt the controller as necessary to achieve optimal performance for a normal aging engine, and the safety net of APST algorithms will examine and interpret data from a variety of onboard sources to detect, isolate, and if possible, accommodate faults. Situations that cannot be accommodated within the faulted engine itself will be referred to a higher level vehicle management system. This system will have the authority to redistribute the faulted engine's functionality among other engines, or to replan the mission based on this new engine health information. Work is currently underway in the areas of adaptive control to compensate for engine degradation due to aging, data fusion for diagnostics and prognostics of specific sensor and component faults, and foreign object ingestion detection. In addition, a framework is being defined for integrating all the components of APST into a unified system. A multivariable, adaptive, multimode control algorithm has been developed that accommodates degradation-induced thrust disturbances during throttle transients. The baseline controller of the engine model currently being investigated has multiple control modes that are selected according to some performance or operational criteria. As the engine degrades, parameters shift from their nominal values. Thus, when a new control mode is swapped in, a variable that is being brought under control might have an excessive initial error. The new adaptive algorithm adjusts the controller gains on the basis of the level of degradation to minimize the disruptive influence of the large error on other variables and to recover the desired thrust response.