Browse Topic: Real-time data

Items (254)
This SAE Aerospace Information Report (AIR) offers an overview of the aspects of intellectual property (IP) protection, legislative compliance, business model, and technologies which need to be considered and addressed to implement a data interoperability, secure business model and technology platform to enable prognostics and health management (PHM) in the digital age. While this information report is restricted to the aerospace domain and also to commercial aviation, the concepts are applicable to any other domain that employs data for supporting health management functionality.
G-31 Electronic Transactions for Aerospace Committee
Rain-Adaptive Intensity-Driven Object Detection for Autonomous Vehicles2020-01-00914/14/2020
Deep learning based approaches for object detection are heavily dependent on the nature of data used for training, especially for vehicles driving in cluttered urban environments. Consequently, the performance of Convolutional Neural Network (CNN) architectures designed and trained using data captured under clear weather and favorable conditions, could degrade rather significantly when tested under cloudy and rainy conditions. This naturally becomes a major safety issue for emerging autonomous vehicle platforms relying on CNN based object detection methods. Furthermore, despite a noticeable progress in the development of advanced visual deraining algorithms, they still have inherent limitations for improving the performance of state-of-the-art object detection. In this paper, we address this problem area by make the following contributions. We systematically study and quantify the influence of a wide range of rain intensities on the performance of popular deep learning based object detection that is trained with clear visual data. We show that even low rain intensities could significantly degrade the performance of object detection trained using clear visuals. Subsequently, we propose a Rain-Adaptive Intensity-Driven (RAID) deep learning framework for object detection under a variety of rain intensities. Controlled experiments based on rain simulations, which are seamlessly integrated with real visual data captured by moving vehicles in truly cluttered urban environments, show the superiority of the proposed RAID framework as compared with state-of-the-art deraining methods in conjunction with popular deep learning based object detection.
Hnewa, MazinRadha, Hayder
Digital Twins for Prognostic Profiling2019-28-245611/21/2019
Ability to have least failures in products on the field with minimum effort from the manufacturers is a major area of focus driven by Industry 4.0 initiatives. Amidst traditional methods of performing system/subsystem level tests often does not enable the complete coverage of a machine health performance predictions. This paper highlights a workable workflow that could be used as a template while considering system design especially employing Digital Twins that help in mimicking real-life scenarios early in the design cycle to increase product’s reliability as well as tend to near zero defects. With currently available disruptive technologies, systems integrated multi-domain 'mechatronics' systems operating in closed-loop/close-interaction. This poses great challenge to system health monitoring as failure of any component can trigger catastrophic system failures. It may be the reason that component failures, as per some aerospace reports, are found to be major contributing factors to aircraft loss-of-control. Essentially, it is either too expensive or impossible to monitor every component or subsystem of a complex machine and the current state of the Integrated Health Monitoring Systems seem to be quite inadequate. In this paper, we propose an approach that combines the best of the diagnostics and feature extraction techniques coupled with Artificial Intelligence as a solution to address the challenges of Prognostics Health Management (PHM) for complex systems. The paper also documents a standard procedure to apply the right technologies/tools at every stage so that a clear process can be applied for any similar complex system across the product development life cycle. In this paper we derive the health status of subsystems by looking at system level responses [1]. Distinguishing features are derived from the overall system level response through feature extraction methodologies and then fed into decision making frameworks that are implemented using both Convolutional Neural Networks [7, 18, 19], Machine Learning [4] and Deep Learning. Models are trained with distinguishable features through system simulations [20]. Employing rightly designed ML models provide the ability of classifying the failure modes as well as to analyze system faults/responses. Predictive modelling techniques are applied to the ML processed data to deliver useful prognostics on the criticality of the failure mode, RUL of the components/subsystems while system is in operation can be determined. The proposed concept can be easily adapted to various systems from varying domains [2]. The methodology evolved in this work can be easily extended for various use cases for instance in the Transportation domain the user can get alerts not only of failures ahead of time but also the remaining useful lifer as well as possible causes of such a failure. This would let prevent downtime of the overall vehicle/fleet and thereby ensures smooth operation of the entire service. As a case study, the present work demonstrates the a DPHM solution applied to electrical energy generator where failure mode effects of subsystems and their effect on the overall system performance are studied using Modeling and Simulation techniques. The overall work would finally lead in demonstrating a working recommendation/advisory system that understand the behavior as if it was a pure Digital Twin [24] and thereby giving a quick turn around for different use cases like study/analysis/what if/predict behavior under various operating conditions with a high level of confidence before the changes are tried on a real system.
Thukaram, PainuriMohan, Sreeram
Optimal Sizing and Energy Management of a Microgrid Using Single and Multi-Objective Particle Swarm Optimization under Autonomous and Grid Connected Mode2019-28-015810/11/2019
The conventional energy sources are getting depleted while at the same time the energy demand keeps growing. Hence, it is important to consider non-conventional energy sources to meet future energy demands. The renewable energy based microgrid system is one of the promising solutions to meet this increasing energy demand. The major parameters under consideration in a micro-grid system are cost-effectiveness, quality of service and energy management. This work concentrates on the energy management of the Photovoltaic/Wind based microgrid system connected to the fuel cell, microturbine and battery under Islanding (or) Autonomous mode and Grid-Connected Mode. The current model of PV, Wind and Battery systems are employed. The Wind, PV and Battery types are chosen from i-HOGA. The optimal combination of these sources with the aim of minimizing the operating cost, pollutant treatment cost and maximizing reliability using both single and multi-objective particle swarm optimization (PSO) has been considered. This microgrid has also been analyzed under three different strategies for both grids connected and islanded mode and the best energy management strategy is obtained after analysis. In addition to this, the type and number of PV, Wind, and Battery to meet the forecasted demand are determined under islanding mode using Multi-Objective Particle Swarm Optimization (MOPSO). A solitary best-accepted solution is attained from Fuzzy membership function. The algorithm proposed decides the optimal number of units and types of units selected to achieve the optimal cost. The simulation has been performed in MATLAB environment.
Dayalan, SuchitraRathinam, RajarajeswariValliappan, Subramaniyan
Intelligent Real Time Inspection of Rivet Quality Supported by Human-Robot-Collaboration2019-01-18869/16/2019
Aircraft production is facing various technical challenges, such as large product dimensions, complex joining processes and the organization of assembly tasks. Meeting the requirements that come with large dimensions, low tolerances and small batch sizes, in combination with complex joining processes, automation and labor-intensive inspection task, is often difficult to achieve in an economically viable way. ZeMA believes that a semi-automated approach is the most effective for optimizing aircraft section assembly. An effective optimization of aircraft production can be achieved with a semi-automated riveting process for solid rivets using Human-Robot-Collaboration in combination with an intuitive Human-Machine-Interaction operating concept. While using dynamic task sharing between human and robot based on their skills, and considering ergonomics, the determined ideal solution involves placing a robot inside the section barrel. The robot’s workspace is expanded by mounting it on top of a lifting unit so that it can properly position the anvil. In the meantime, the human performs the more complex tasks of inserting the solid rivets and operating the riveting hammer from the outside of the section barrel. By implementing a modular control system for configuration and operation of the assembly station with a variety of interaction possibilities, human and robot can perform the collaborative riveting process effectively. Additionally, due to high forces and vibrations, which are applied by the riveting hammer, a process specific tool has been developed to prevent damage to the robot system. By equipping the tool with sensors, such as a force torque and laser line sensor, it can not only monitor the riveting process in real time, but also perform inspection tasks using artificial intelligence algorithm. The sensor data will be passed through a trained machine learning model classifying whether the scanned part is of good or bad quality. With a limited amount of data, the “online” rivet classification using sensor data reaches a classification precision of 86%, whereas the rivet classification based on pictures reaches a precision of 97%. The result of the quality inspection will be sent to the employee using Mixed Reality glasses. This will allow the operator to react appropriately and carry out necessary maintenance. The results are part of the project "Modularity, Safety, Usability, Efficiency by Human-Robot-Collaboration - FourByThree" (no 637095), funded by European Union's Horizon 2020 research and innovation program at ZeMA and will present semi-automation shown in the HRC riveting process. Furthermore the research is funded by the Interreg V A Großregion within Robotix-Academy project (no 002-4-09-001). The rivet quality inspection using artificial intelligence has been supported by PIKON Deutschland AG.
Mueller, RainerVette, MatthiasMasiak, TobiasDuppe, BenjaminSchulz, Albert
An Improved Multi-Pedestrian Tracking System Based on Deep Neural Network2019-01-10554/2/2019
The intelligent vehicle driverless technology has become a very hot topic in the past two decades. To solve the road safety problem, which is one of the most important factors inhibiting the development of intelligent vehicle technology, the multi-target tracking system has attracted more and more attention in recent study since it can detect and track multiple objects in traffic scene so as to help the whole driving system plan the safe route. In this work, a novel multi-pedestrian tracking system based on deep neural network is proposed to improve the tracking efficiency while providing high recognition accuracy. The proposed tracking system consists of two parts: 1) pedestrian detector, and 2) pedestrian tracker. For the detector part, we first transform the image convolution operation in spatial-temporal domain to the coefficient product operation in complex frequency domain, and then replace the maximum pooling and mean pooling operation in the traditional SSD detection network by the proposed product operation, and the modified SSD model is used as the pedestrian detector. For the tracker part, we train a corresponding appearance model and calculate the appearance similarity of the detected targets between the continuous frames with respect to the cosine distance. To evaluate the performance of our proposed system, we implement the system on the MOT16 benchmark. The detector can improve tracking accuracy by up to 8.9% and the tracker updates the speed of whole system at a rate of 126Hz, our extensions reducing the number of identity switches by 15%. The experimental results demonstrate that the proposed multi-pedestrian tracking system provides better real-time performance and accuracy than the state-of-the-art methods, which can provide more accurate object information for the following auto-driving system with higher safety.
Gong, YuanChi, JianningYu, XiaoshengWu, ChengdongZhang, YifeiGao, Na
A Blockchain-Backed Database for Qualified Parts2019-01-13433/19/2019
Certain standard parts in the aerospace industry require qualification as a prerequisite to manufacturing, signifying that the manufacturer’s capacity to produce parts consistent with the performance specifications has been audited by a neutral third-party auditor, key customer, and/or group of customers. In at least some cases, a certifying authority provides manufacturers with certificates of qualification which they can then present to prospective customers, and/or lists qualified suppliers in a Qualified Parts List or Qualified Supplier List available from that qualification authority. If this list is in an infrequently updated and/or inconsistently styled format as might be found in a print or PDF document, potential customers wishing to integrate qualification information into their supplier tracking systems must use a potentially error-prone manual process that could lead to later reliance on out-of-date or even forged data. This paper proposes a blockchain-backed database for such applications, facilitating integration with integrators’ electronic systems including near real-time data updates and a reliable audit trail of changes, certificates that provide more reliable signals of data integrity, and a better user interface with enhanced search capabilities. Though piloted with some centralized control related to the centralized issuance of qualifications, the paper describes how blockchain technology in this application could allow a consortium of companies to manage such a database in a decentralized structure like a decentralized autonomous organization. The proposed database also introduces generalizable data structures which can lend powerful dynamism to other data stores in domains with similar data structures or challenges. This paper describes an example implementation converting the TS200 Qualified Manufacturers List to a blockchain-backed database with search and administrative interfaces. The paper further discusses practical challenges associated with implementing such a database and future directions for additional capabilities.
Towne, W. Ben
Data Interoperability for Aerospace IVHM Systems2019-01-13423/19/2019
Aerospace systems today are generating a lot of data and for the most part all this data is being generated by siloed entities (by various stakeholders like components/sub-system manufacturers, OEMs, operators) and ends up living within the four walls of these individual entities. For the industry to fully benefit from this data there needs to be a transparent way to share this data while strictly controlling the proprietary nature of the data and adhering to all contracts. The SAE HM-1 technical committee is writing an aerospace information report (AIR) 6904 to describe a digital data landscape and approach that can support health management [1]. Integrated vehicle health management (IVHM) systems cut across many disciplines and boundaries and can benefit from structured landscape and well defined approach. For example, data associated with a fault in an aircraft subsystem like the engine must travel through multiple systems and boundaries before it can be analyzed by the cognizant personnel. Today the landscape is pretty ad-hoc; with not many standards governing the data handling, storage, analysis, and disposal. This information report is a beginning in describing how a more systematic way of structuring the interactions might make the job of dealing with all this data a little easier. In this paper, the AIR 6904 is summarized to introduce it to larger aerospace community in order to improve it further in subsequent revisions incorporating the feedback.
Rajamani, RaviWhitfield, MartinKumar G. V. V., Ravi
Improving Manufacturing Efficiencies through Industry 4.0 Technologies in Aerospace2018-01-192910/30/2018
1 In the age of 4th industrial revolution, operational and information technologies are increasingly getting converged to help organizations improve their topline through new innovative products and services, and improve bottom line by improving efficiencies. This transformation is driven by convergence of many advanced technologies such as advanced sensor and communication technologies, big data, advanced analytics, Artificial Intelligence (AI), robotics, additive manufacturing, virtual and augmented reality (VR/AR). Enterprises digitization journey continues to adopt advanced technologies through multi-pronged approach to achieve their near-term and long-term goals. This paper summarizes Industry 4.0 journey, its relevance and applications to aerospace. It also summarizes how Industry 4.0 concepts can be applied to a composite manufacturing shop floor of aerospace components, how effective convergence of IoT, analytics, machine learning, AI and AR/VR help in improving the overall efficiency, reliability, availability and quality of the manufacturing shop floor by monitoring real time data to evaluate the overall performance of manufacturing plant “As Designed” Vs “As Operated” quantifying the business value.
Veluri, SastryKumar, RaviVasudevan, RamjiGorur, Ravi PrakashNampuraja, EnoseShankaraiah, MaheshTanjore, SimhaRao, Shama
A Comparative Study on Knock Occurrence for Different Fuel Octane Number2018-01-16749/10/2018
Combustion with knock is an abnormal phenomenon which constrains the engine performance, thermal efficiency and longevity. The advance timing of the ignition system requires it to be updated with respect to fuel octane number variation. The production series engines are calibrated by the manufacturer to run with a special fuel octane number. In the experiment, the engine was operated at different speeds, loads, spark advance timings and consumed commercial gasoline with research octane numbers (RON) 95, 97 and 100. A 1-dimensional validated engine combustion model was run in the GT-Power software to simulate the engine conditions required to define the knock envelope at the same engine operation conditions as experiment. The knock intensity investigation due to spark advance sweep shows that combustion with noise was started after a specific advance ignition timing and the audible knock occur by increasing the advance timing. Therefore, the engine operation was divided into three regions; knock-free, light knock and heavy knock. The results for heavy-knock were well suited to audible knock detected by knock sensor. The simulation results from knock model divide the engine operation into two regions; normal combustion and knock region. The knock region was well suited to light-knock and heavy-knock which has been defined using experimental results. Next, an artificial neural network (ANN) model has been designed to classify the different RONs using engine rotational speed signal. The model classified different RONs accurately after starting point of noisy combustion (light-knock). This point defined from experimental results and was well suited with starting point of knock index increment from simulation results. The simulation tool ability to predict the knock envelope will reduce the experimental cost and time to generate the spark timing look-up table.
Ghanaati, AliMuhamad Said, Mohd FaridMat Darus, Intan Zaurah
Autonomous vehicles at various stages will impact the future of transportation by improving reliability, comfort and safety of the passengers. In this paper, for an existing experimental vehicle, fitted with various sensors and actuators typically required by autonomous vehicles, a basic level-1 autonomous controller for braking and throttle actuations is proposed. This controller is primarily developed for stop-and-go scenarios along with the additional functionalities of automatic cruise control (ACC) and automatic emergency braking (AEB). Since the rigorous testing of autonomous vehicle in actual roads can be time consuming, costly and having safety issues, a simulation test-bench based approach is considered to develop and test the controller. The controller, based on practical data is developed in simulation environment to primarily maintain safe distance from surrounding traffic objects while fulfilling requirements such as jerk levels, conditional braking, speed limits, etc. In this work, only a longitudinal controller is developed for low speeds (<30 kmph) and low throttle scenarios for which a four-wheel based vehicle dynamics model is formulated excluding the nonlinear tire model. Experimental data while running the experimental vehicle in actual traffic is acquired from camera, triangulation of ultrasonic sensors, throttle, brake pedal position and velocity of the vehicle and is used for tuning and validation of the derived model, to ensure satisfactory accuracy. Accordingly, a relative distance and relative velocity dependent longitudinal controller comprising of several coordinated PID controllers is designed in stop-and-go scenario and in AEB mode. The captured pre-recorded traffic video along with acquired throttle, braking, speed and relative distance information is synced with the proposed controllers simulation execution for correlations, wherein the acquired relative distance data is used as reference to run the simulations. The proposed practical data based simulation test environment is successful in creating multiple test scenarios and the developed longitudinal controller is able to satisfactorily autonomously control the vehicle in the desired manner.
Goel, AyushSengupta, Somnath
Custom Data Logger for Real-Time Remote Field Data Collections17AERP10_1010/1/2017
Compact, energy efficient instruments have the same functionality as a personal computer. Army Engineer Research and Development Center, Vicksburg, Mississippi The U.S. Army Corps of Engineers (USACE), CHL, FRF, had a need for a remote real-time data collection system to control instruments and log and communicate data from five observing stations in the Currituck Sound Estuary, NC1. These stations, referred to as the Currituck Sound Array (CSA), collect a suite of meteorological and oceanographic data including wind, air temperature, humidity, incoming solar radiation (above and below water), waves, currents, water level, salinity, and water temperature, as well as turbidity and many other water quality parameters. This array of instruments has a variety of control commands, sample routines, and output data formats. Additionally, the CSA was designed to act as a natural laboratory for estuarine research and as an instrument and model test bed. These capabilities required a reliable and flexible system that would allow easy modification of sampling schemes, the ability to log as many as 15 instruments with a single logger, and allow the incorporation of additional and novel instrumentation with minimal effort and expense. The custom loggers were built upon single board computers (SBC) running the Linux operating system. They effectively have the same functionality as a personal computer, overcoming many of the limitations of off-the-shelf loggers. Additionally, off-the-shelf loggers typically operate on a very limited set of commands. These custom Linux-based loggers have a much more diverse and powerful selection of commands, overcoming many of the unique challenges of real-time data collection with robust code and programmatic “watchdogs” that can automatically make sure the logger, instruments, and communications are operating as intended.
Secure Deterministic L2/L3 Ethernet Networking for Integrated Architectures2017-01-21039/19/2017
Cybersecurity attacks exploit vulnerabilities related to the increased complexity and connectivity of critical infrastructure systems. This paper investigates the context and use of key security technologies, processes, challenges and use cases for the design of advanced integrated architectures with security, safety, and real-time performance considerations. In such architectures, deterministic Ethernet standards are used as a baseline for system integration in closed embedded systems or open mixed criticality systems. Security-informed safety development processes for integrated architectures are required to prevent catastrophic failures caused by environmental and cyber threats, due to expanding number of security vulnerabilities in complex and increasingly open systems. State-of-art safety/security processes for integrated systems in cross-industry environments are considered and similarities examined, for different types of integrated architectures. In integrated systems and IMA which share common resources, multi-level secure systems and composable modular architectures such as MILS based on separation kernels and ARINC653 API are gaining importance for design of safe and secure distributed applications with real-time performance requirements. Network security is a core component of the overall cyber-security and defense-in-depth capability for distributed architectures. Protection mechanism for information, interface and system integrity, communication availability, and data confidentiality are required for design of safe and secure integrated embedded infrastructure. In deterministic Ethernet networks with Time-Triggered Ethernet (SAE AS6802) and ARINC664 services can actively support security measures for mixed-criticality applications. The network partitioning, dataflow isolation, configuration protection, per-flow traffic policing, link and end-to-end encryptions or authentication, and internal network device partitioned architecture can be useful for design of open networked systems which can also accept previously unknown soft-time or bursty traffic, while hosting highly critical functions with temporal boundaries. After an overview of security issues in networks within integrated architectures, this paper continues with discussion of MACsec and IPsec mechanisms, packet firewalls, secure shells and Denial-Of-Service (DoS) protection mechanisms for secure and deterministic L2/L3 networking.
Hirschler, BerndJakovljevic, Mirko
Items per page:
1 – 50 of 254