Browse Topic: Big data

Items (62)
Find
Big Data: It's Not Just About the Raw DataF-0081-2025-01635/20/2025
Big Data technologies have become quite ubiquitous in the last years, allowing for the storage of substantial amounts of data, typically flight test data as recorded by the flight test installation. On recent helicopter prototypes, we generate in excess of 50 GB of raw data per flight hour, usually in a format not adequate for efficient large-scale processing. With some specific optimizations and the setup of a specialized infrastructure, there are now practicable means to store timeseries in ways that allow for requests spanning hundreds or thousands of flights to complete within minutes, opening the way to some substantial savings and new insights. However, to make the most of these data and make informed decisions it is often quite important to store contextual data that go beyond the pure timeseries data, typically on helicopters where optional installations can have a significant impact on aircraft performance or behavior. This paper explores the various kinds of data and metadata related with flight tests, how to collect them and relate them with one another in order to maximize raw data value and avoid some common pitfalls. We define more accurately the data of interest, where to collect them and some ideas to further improve data collection in the future.
Brisset, Nicolas
Robust Behavioral Cloning for Autonomous Vehicles Using End-to-End Imitation Learning12-04-03-00238/19/2021
In this work, we present a lightweight pipeline for robust behavioral cloning of a human driver using end-to-end imitation learning. The proposed pipeline was employed to train and deploy three distinct driving behavior models onto a simulated vehicle. The training phase comprised of data collection, balancing, augmentation, preprocessing, and training a neural network, following which the trained model was deployed onto the ego vehicle to predict steering commands based on the feed from an onboard camera. A novel coupled control law was formulated to generate longitudinal control commands on the go based on the predicted steering angle and other parameters such as the actual speed of the ego vehicle and the prescribed constraints for speed and steering. We analyzed the computational efficiency of the pipeline and evaluated the robustness of the trained models through exhaustive experimentation during the deployment phase. We also compared our approach against state-of-the-art implementation in order to comment on its validity.
Samak, Tanmay Vilas, Samak, Chinmay Vilas, Kandhasamy, Sivanathan
Defect Detection of Railway Fasteners Based on Improved PHOG Characteristics09-08-01-00023/23/2020
Aiming at the problem of low recognition rate and slow speed caused by the small proportion of key area information in feature vectors of original Pyramid Histogram of Gradients (PHOG) features, an improved feature extraction method of PHOG is proposed. The PHOG feature extraction method is combined with edge feature enhancement method based on Census transform to extract feature vectors of fasteners, and dimensionality reduction is processed by Kernel Principal Component Analysis (KPCA) method to reduce the interference of redundant information. The vector is inputted into the support vector machine for training in order to get the classifier model and realize the automatic identification of the fastener’s state. The simulation results show that compared with the traditional PHOG method, this feature extraction method improves the false detection rate by 2.7%, and the complexity of the algorithm is greatly reduced.
Wu, Chun-Ming, Hu, Jiaming, Zheng, Hongkuo
Collaboration: A Powerful, Effective Strategy for Medtech Companies Moving to Embrace Industry 4.0 and Drive Digital TransformationTBMG-362763/1/2020
The digital revolution under way in medtech manufacturing will continue to transform the industry into a more connected, efficient, and agile ecosystem. However, the challenge for many companies is to plot the best way to embrace various existing and new technologies and move toward smarter manufacturing — their version of the “factory of the future.” Every company is at a different stage of digital transformation. Some companies are wholeheartedly piloting, integrating, and scaling disruptive technologies within their operations in a bid to realize greater supply chain and internal operational benefits. Others are at a different place in their digital journey, identifying incremental opportunity areas where they can demonstrate initial impact and justify further investment.
2019 'Create the Future' Electronics/Sensors/IOT Category Winner: Early Detection of Battery FaultsTBMG-3544711/1/2019
Analysis of Accelerator Hardware for Autonomous Vehicles and Data Centers2019-01-261510/22/2019
The development of Autonomous Vehicles (AV) has become a popular subject in academia and industry. Companies and cities are quickly realizing the opportunities that AVs can generate from Mobility as a Service to traffic safety. The challenges for the infrastructure to incorporate AVs as a viable transportation source are immense, from an outdated infrastructure to radical Smart-City designs. Historically, the transportation infrastructure has faced challenges from underfunding, economics, and much needed improvements. With the current infrastructure unable to support many of the services required by a fully connected network, a transformation will be necessary to meet growing mobility needs. The role of accelerating technology in data centers are key for production operations among industry leaders such as Amazon and Microsoft for real-time processing. The same accelerating technology that has successfully impacted data centers will play the same role in much smaller micro data centers (mDC) for Smart-City design in the transportation infrastructure. These mDCs and Edge computing sites will be tasked with the latency, tasking caching and offloading (TCO), and processing of millions of connected vehicles simultaneously. With the recent upgrade of 5G from 4G wireless connectivity will invariably provide lower latency to Edge computing devices used in real-time applications. This paper provides an analysis of accelerator technology for real-time processing in the transportation infrastructure. Accelerator hardware such as FPGAs, GPUs, and ACISs will be highlighted from current research that support real-time capabilities. As the popularity of AVs and a connected network continues to grow, the role of accelerator technology will enable large scale real-time processing in AVs and the transportation infrastructure.
Brown, Kyle W.
Data Privacy in the Emerging Connected Mobility Services: Architecture, Use Cases, Privacy Risks, and Countermeasures11-02-01-000410/14/2019
The rapid development of connected and automated vehicle technologies together with cloud-based mobility services is transforming the transportation industry. As a result, huge amounts of consumer data are being collected and utilized to provide personalized mobility services. Using big data poses serious challenges to data privacy. To that end, the risks of privacy leakage are amplified by data aggregations from multiple sources and exchanging data with third-party service providers, in face of the recent advances in data analytics. This article provides a review of the connected vehicle landscape from case studies, system characteristics, and dataflows. It also identifies potential challenges and countermeasures.
Li, Huaxin, Ma, Di, Medjahed, Brahim, Kim, Yu Seung, Mitra, Pramita
Handling of Data from Heterogeneity of Vehicular Devices through Inter-Networking2019-28-015610/11/2019
Collection of various data from sensed data or raw availability of data from transcript or interdependency of data from various sources is a tedious task in a real time scenario like an Indian context is considered. Planning to find a solution to collect the data from various vehicular devices about the information related to the pollution becomes a cumbersome job. The need of the data, under what time duration data has to be transmitted, how they are interconnected and whether data needs to be stored or how they are processed is a major question that arise when dealing with collecting data and internetworking with various vehicular devices. A study of two different types of approaches for internetworking between the devices is discussed. One related to real time setup of mobile application and other with the dynamic cluster approach when the nodes are moving in a region was considered. Eliminating the speed of the vehicle with the movement of human formation of cluster with the mobile application to identify the various devices in the vicinity was identified and handoff between the clusters was also taken care in the first approach and in the second one performance metrics while establishing the multi-layer cluster was discussed.
Anuradha, Gnanaprakasam, Ramesh Babu, Kalivaradhan, Naiju, Chooriyaparambil Damodaran
Engineered Resilient SystemsTBMG-3535110/1/2019
Analyzing large data is the process of methodically and systematically making decisions to reduce incomprehensible datasets down to a manageable size that can be viewed and understood easily. These reductions are made by performing data analytics, producing metrics, identifying patterns, and/or producing any other criteria of comparison that can be mathematically modeled.
Executive Forum: Sensors & the IoTTBMG-351449/1/2019
No longer just a buzzword, the Internet of Things (IoT) is rapidly taking hold in many different industries, from aerospace and automotive, to medical and manufacturing. The IoT ecosystem incorporates Web-enabled smart devices that use processors, sensors, communication hardware, and software to collect, send, and act on data they acquire. Tech Briefs posed questions to executives in the sensors and IoT component industries for their views on issues concerning the role of the IoT in manufacturing.
SAE Truck & Off-Highway Engineering: August 201919TOFHP088/1/2019
Electric machines find their niche Smaller machines working in cities are particularly ripe for electrification, but most off-highway applications will retain the internal-combustion engine. Big data, big benefits Data mining helps users and equipment developers use data from on-vehicle sensors to work more efficiently. Data connectivity in harsh environments Ensuring high-speed data transmission requires OEM designers to think more about components, placement and the impact of environmental conditions early in design. Trucks with intuition Perceptive Automata partners with Volvo Trucks to demonstrate AI technology that can determine when pedestrians will cross the road. Change is afoot Cummins and its CTO help focus the 2019 SAE COMVEC on the fast-shifting technology landscape that faces both on- and off-highway sectors. Editorial AI, ADAS & AVs-oh my! Holistic hydraulic designs add more sensors, controls DSD helps 'balance the books' for buying electric truck technology JCB prepared to meet EU Stage V emissions Protecting a cyber-physical remote diagnostic communication system against cyberattacks SKF employs digital tools to enhance bearing design, performance New Mercedes-Benz Actros loaded with ADAS, active safety tech Q&A Peloton CEO Josh Switkes explains Level 4 'automated following' platooning
What M&E Can Teach the AV Industry About Data19AVEP07_057/1/2019
Media & entertainment offers important learnings on data retention, management, scalability and security. At first glance, autonomous vehicles would seem to have little in common with the Media and Entertainment (M&E) industry, beyond action-movie car chases and in-vehicle entertainment screens. But AVs and M&E do have a lot in common when it comes to their data. Both generate hundreds of petabytes of data per year. Both need to manage that data for retrieval and value and both need to store that data with security and protection.
Coari, Jason, Pastor, Mark
Sensors Expo Preview 2019TBMG-345846/1/2019
The 2019 Sensors Expo and Conference will be held at the McEnery Convention Center, San Jose, CA from June 25 – 27. Both the expo and conference are excellent opportunities to get a good sense of important sensor trends. Among the wide variety of themes at the conference, some strike me as being very significant for the current state of the industry.
Editorial19AVEP05_015/1/2019
AVs, data and ‘surveillance capitalism’ I finally hit my limit after seeing a Twitter post-from, almost paradoxically, a privacy engineer at Google-showing a taxi cab in Asia with a passenger-area screen announcing it was employing facial recognition “to deliver the most optimized content” to the rider. “The image data is discarded immediately,” the little tablet professed.
Visnic, Bill
Driver Workload in an Autonomous Vehicle2019-01-08724/2/2019
As intelligent automated vehicle technologies evolve, there is a greater need to understand and define the role of the human user, whether completely hands-off (L5) or partly hands-on. At all levels of automation, the human occupant may feel anxious or ill-at-ease. This may reflect as higher stress/workload. The study in this paper further refines how perceived workload may be determined based on occupant physiological measures. Because of great variation in individual personalities, age, driving experiences, gender, etc., a generic model applicable to all could not be developed. Rather, individual workload models that used physiological and vehicle measures were developed. Unlike some existing methods of workload estimation where one, or a few signals are used, such as electroencephalography (EEG), electrocardiography (ECG), we developed intelligent systems that use multiple physiological and vehicle signals based on an end-to-end deep neural learning architecture to make a robust estimation of workload. The deep neural learning system, MTS-CNN, is designed to learn workload patterns from synchronized, heterogeneous temporal signals. All data collected for training and testing are from real-world driving trips along the same route which comprised urban local roads and highways. Data from twenty participants whose driving experience ranged from a few months to several years were collected and analyzed. The experimental results indicate that the proposed driver workload estimation model is capable of learning well from the combined temporal physiological and vehicle signals and good performance was obtained on workload estimation.
Murphey, Yi, Kochhar, Dev S., Xie, Yongquan
On Collecting High Quality Labeled Data for Automatic Transportation Mode Detection2019-01-09214/2/2019
With the recent advancements in sensing and processing capabilities of consumer mobile devices (e.g., smartphone, tablet, etc.), they are becoming attractive choices for pervasive computing applications. Always-on monitoring of human movement patterns is one of those applications that has gained a lot of importance in the field of mobility and transportation research. Automatic detection of the current transportation mode (e.g., walking, biking, riding a shuttle, etc.) of a consumer using data from their smartphone sensors enables delivering of a number of customized services for multi-modal journey planning. Most accurate models for automatic mode detection are trained with supervised learning algorithms. In order to achieve high accuracy, the training datasets need to be sufficiently large, diverse, and correctly labeled. Specifically, the training data requires each type of mode data to be collected for a minimum duration that is necessary and sufficient for building high accuracy models. Collecting such data in an efficient manner is challenging because of the variability in the test subjects’ multi-modal journey patterns, e.g., using mostly private vehicles for commute, not sufficiently using rideshares, etc. In this paper, we describe a Design of Experiment (DoE) to efficiently collect supervised training dataset from user smartphones in a controlled environment. In this DoE, we asked the subjects to use a data logger app during a multi-modal trip designed around the Ford Dearborn campus with the right trip characteristics. The app persistently logged GPS and motion sensor data in the background and sent it to a remote Hadoop server. Location data was used to detect movement and enhance the quality of the data collection in real-time, e.g., the app paused data logging when the user is detected to be waiting between two transit modes.
Rao, Prashant, Melcher, David, Mitra, Pramita, Rao, Sriram
Connections dig deeper19TOFHP04_014/1/2019
More aspects of vehicle operation are being improved by leveraging cellular links. Commercial-vehicle suppliers and owners who have been at the forefront of utilizing connectivity are pushing to gain further improvements. Safety is being enhanced as drivers are monitored for speed and hours, while predictive maintenance and software updating are helping improve efficiency. Equipment suppliers aren't waiting until the advent of 5G cellular links to improve their offerings. Existing cellular systems are being used to improve operations by using data collected from vehicle networks.
Costlow, Terry
SAE WCX: Bringing You the Future of MobilityTBMG-339163/1/2019
WCX™ World Congress Experience 2019 — presented by SAE International — is for forward-thinking engineers, executives, OEMs, suppliers, decision-makers, disruptors, and the entire spectrum of the mobility engineering field. Held April 9-11 in Detroit, WCX will offer technical education, peer-to-peer networking, a technology-driven exhibit floor, and global mobility solutions. The event covers every corner of the mobility industry from IoT, Big Data, and connectivity, to unmanned vehicles, sustainability, and cybersecurity. More than 10,000 technical experts and business leaders attend WCX for a look into the latest trends and innovations.
LEAN Techniques for Effective, Efficient and Secure Information Processing in Automotive Homologation2019-26-03351/9/2019
It is an established fact that virtual knowledge based engineering has revolutionized R & D activities by streamlining processes, ensuring productivity and accuracy. This has resulted in freeing up time for quality interpretational work and decision making for engineering the best of products. Subsequently, homologation is a mandatory requisite activity for product signoff. It certifies the quality of the product and is an important factor in giving the product an authenticity for sale in the market. Homologation entails compliance to regulations existing in form of well-established standards which elaborate systematic and detailed guidelines on conducting physical testing for automotive systems, sub-systems or components for specific vehicle types. The contemporary homologation scenario encompasses heavy usage of virtual platform tasks like data acquisition, application of heuristics from the standards, post processing, classification, test output report and type approval certificate generation. It is highly desirable that the homologation procedure needs to be streamlined and of high fidelity through seamless integration of all steps involved in the information processing as well as effective capture of homologation knowledge in a virtual form. This can be achieved by LEAN knowledge based techniques. The author hereby elucidates such a LEAN knowledge based framework that captures the heuristics outlined in the homologation standards in a form comprising of a structured taxonomy to address each of the compliance requirements, and seamlessly integrates them with all the upstream and downstream information processing tasks involved in the certification. Two sample tools are showcased for expounding the efficacy of this framework. These tools are integrated into the daily test procedures followed by the testing personnel, ensuring phenomenal productivity and accuracy. In addition, these tools leverage frugal automation platforms available right on the desktop, rendering them highly cost effective. Thus both tacit and explicit regulatory and process knowledge is captured and secured effectively which also conforms to requirements of system standards such as IS0 9001 and IS0 27001.
Thipse, Yogesh
Bring innovation, disruption in-house18TOFHP10_0110/1/2018
In-house or outsource? That is the question facing commercial-vehicle and off-highway manufacturers playing in this brave new world of big data, artificial intelligence and vehicles on the verge of becoming highly automated. A profusion of tech and service providers are lining up to assist manufacturers when and where they need it, but several executives speaking at the recent SAE COMVEC Technology Connection conference indicated they'd prefer to keep as much knowledge and talent as possible, in-house.
Gehm, Ryan
Quotes from COMVEC 201818TOFHP10_0910/1/2018
Industry leaders spoke extensively about all things autonomous-ADAS, big data, connectivity, cybersecurity, machine learning-at the annual SAE event. Here's some of what they had to say. Troy Clarke, Chairman, President & CEO, Navistar “We could [fully] automate 10, 20 percent of the truck fleet that exists and really not disrupt one driver because there's that much need for drivers today.”
SAE Truck & Off-Highway Engineering: October 201818TOFHP1010/1/2018
Are higher voltage architectures imminent? As the limitations of current 12V architectures become more apparent, the commercial vehicle industry could be on the verge of adopting 48V. The only question is, will improvements in other technologies offer something better? Sensing changes in autonomous trucks Requirements for sensors and controls for commercial vehicles differ significantly from those used for cars. Many paths lead to reduced emissions A wide range of ICE technologies are needed to meet increasingly stringent emissions regulations. Quotes from COMVEC 2018 Industry leaders spoke extensively about all things autonomous-ADAS, big data, connectivity, cybersecurity, machine learning-at the annual SAE event. Here's some of what they had to say. Fuel-cell Class 8-take 2.0 With a longer-range and more-refined fuel cell-powered heavy-duty truck, Toyota aims to eventually eliminate emissions from trucks serving increasingly congested California ports. Editorial Bring innovation, disruption in-house Adding 3D printing to design, manufacturing processes Upstream devoted to truck cybersecurity threats Jacobs employs cylinder deactivation in HD engines to lower CO2, NOx Emissions reductions continue to disrupt CV industry Mercedes doubles down on electric vans and buses, considers fuel cells Off-road bus from Torsus transports to hard-to-reach places Q&A Perkins pursues plug-and-play connectivity
A Nanotech Sensor Turns Molecular Fingerprints into Bar CodesTBMG-328719/1/2018
Infrared spectroscopy is the benchmark method for detecting and analyzing organic compounds. However, that requires complicated procedures and large, expensive instruments, making device miniaturization challenging and hindering its use for some industrial and medical applications. It is also difficult to use it for data collection in the field, such as measuring pollutant concentrations. Furthermore, since it is fundamentally limited by low sensitivity, it requires large samples.
A Real-Time Traffic Light Detection Algorithm Based on Adaptive Edge Information2018-01-16208/7/2018
Traffic light detection has great significant for unmanned vehicle and driver assistance system. Meanwhile many detection algorithms have been proposed in recent years. However, traffic light detection still cannot achieve a desirable result under complicated illumination, bad weather condition and complex road environment. Besides, it is difficult to detect multi-scale traffic lights by embedded devices simultaneously, especially the tiny ones. To solve these problems, this paper presents a robust vision-based method to detect traffic light, the method contains main two stages: the region proposal stage and the traffic light recognition stage. On region proposal stage, we utilize lane detection to remove partial background from the original image. Then, we apply adaptive canny edge detection to highlight region proposal in Cr color channel, where red or green color proposals can be separated easily. Finally, extract the enlarged traffic light RoI (Region of Interest) to classify. On traffic light recognition stage, a tinny but effective convolution neural network (CNN), named TLRNet, classifies each traffic light RoI into its own class. In fact, deep learning (DL) is bad for detecting small object in many fields, so we use region proposal stage to get RoI and classification by CNN to achieve a good result. We validate our method both on Laboratory for Intelligent and Safe Automobiles (LISA) Traffic Lights Dataset and video sequences captured from Beijing’s streets. The experimental results prove that the proposed method can achieve a good result for the multi-scales traffic lights in the TX1 embedded platform, and reach a real-time performance at 28fps.
Yu, Guizhen, Lei, Ao, Li, Honggang, Wang, Yunpeng, Wang, Zhangyu, Hu, Chaowei
The Use of “Big Data” for the Analysis and Design of Vehicle Sound Packages2018-01-15706/13/2018
With the ever-decreasing timescales and increased performance requirements afforded to OEM’s, it has become essential that NVH suppliers provide optimum palliative solutions that comply with a vehicles acoustic targets. The acoustic effect of any palliative treatment attached to a vehicle body system depends on its ability to attenuate noise energy passing through or radiating from the system or its interaction with reflected sound from other areas. Acoustic performance uses targets relating to sound insertion loss (SIL) and / or sound absorption and these are identified to the component supplier by the OEM at the “request for quotation” (RFQ) stage. For many potential suppliers, especially those with a limited portfolio of material options, success or failure is quite straightforward. However, the problem occurs when the material and processing opportunities cover wide parameters and the available combinations and permutations are extensive. It is no longer a simple choice to get the best solution. Ultimately, competitiveness relies on the optimum choice of material types, combinations and processing along with associated cost and this requires a detailed understanding of the ‘physics’ involved. Whilst material prediction software is frequently used to spot check performance prior to actual material testing this technique cannot guarantee success. It is also very time consuming and requires considerable training. The aim of this project was to use “Big Data” to automate the selection process. This paper describes the Authors work with “Big Data” combined with associated algorithms, so that once a system target is received a range of suitable solutions can be offered without pre-determination of parameters. It covers the creation of the “Big Data” landscapes and the integration of the procedure into a web based easily accessible application.
Morris-Kirby, Rod
Human Emotion Based Interior Lighting Control2018-01-10424/3/2018
In recent years, research on Human Computer Interaction (HCI) based on emotion recognition using behavioral and physiological signals have attracted immense interest in research circles. Lighting inside the automotive make us feel differently about our driving and how we feel or behave. From the literature, it is observed that ambient lighting makes an impact on the driving experience and it delivers an emotional atmosphere inside the automotive. Driving fatigue can be reduced if the lighting is controlled properly. These days, ambient interior lighting can be considered to be the point of fashion for high end automotive and also impact driver’s mood and comfort. There are different types of automotive based lighting automation systems available but emotion based control is in early or nascent stages of research. Speech controlled light control systems, control the light by the recognition of speech of the user and by using facial expressions lighting can be controlled. Facial/speech signals consist of both outward physical expression and the inborn emotions. These emotional signals thus exhibited vary from situation to situation and are mostly dependent on the conditions. In this work, we attempted an emotion based interior lighting control. Based on the emotions observed through the Emotion Recognition System (ERS), the lighting can be modified in a predefined fashion. In the proposed ERS, five types of emotions are considered, like happy, sad, angry, neutral and disgust. Live image expression and voice data of the driver/passengers are considered as inputs to the system and based on the output from the ERS, interior lighting is controlled. Standard databases are used for training ERS system. The proposed algorithm is tested, and the results demonstrate the approach and the reliability of the method to obtain the solution for lighting control. Machine Learning methods like Convolution Neural Networks (CNN) are used for classification of features in ERS system.
Nandyala, Sivaprasad, K, Gayathri, D H, Sharath, Manalikandy, Mithun
Cold Chain Management Using Model Based Design, Machine Learning Algorithms and Data Analytics2018-01-12014/3/2018
In the food industry, there is an increased demand for generic pharmaceutical products and perishable food without compromising with the changes in texture and taste that occur in the transit. With this demand, there is a need for better visibility of products in the logistics network, to minimize wastage, to ensure product integrity, influence productivity, transparently track the fleet and to identify pathogens before a potential outbreak. In Cold Chain Management, information is power: with potentially billions of dollars’ worth of cargo (such as food items, vaccines, serums, tests or chemicals) at stake worldwide. Hence, careful live monitoring, inspection, supervision, validation and documentation of business-critical information is essential. In this paper, we have proposed a framework for Cold Chain Management using Internet of Things (IoT) combined with other technological innovations such as: Cloud Computing, Machine Learning and Big Data Analytics to revolutionize the cold transport industry. By establishing such an architecture, we have tried to monitor, visualize, track and control various platform dependent parameters thereby providing a complete solution across the fleet cycle with assured freshness and palpability.
Khanuja, Gagandeep Singh, D H, Sharath, Nandyala, Sivaprasad, Palaniyandi, Balamurali
Validation of Automated Prediction of Blood Product Needs Algorithm Processing Continuous Non-Invasive Vital Signs Streams (ONPOINT4)TBMG-287884/1/2018
Hemorrhagic shock occurs frequently in natural and man-made disaster scenarios. To control bleeding and to provide necessary resuscitation, swift and accurate diagnosis and decision-making are required. Early recognition of bleeding and the need for targeted interventions could improve both survival and resource management, allowing the receiving hospital to prepare required blood, surgeons, or other resources in advance of patient arrival and to conserve valuable resources in those patients who are not bleeding. Resources can be saved through avoidance of over-triage, thereby reducing unnecessary air transport, unnecessary blood transfusions, and unnecessary evaluation with labs, X-rays, and computed tomography scans, which is important in all resource-constrained and austere environments.
Big Data a focus for 2018 SAE President Mircea Gradu18TOFHP02_042/1/2018
Velodyne CQO sees use of Big Data as a way to capitalize on multiple industry trends including vehicle electrification and automated vehicles. A highly respected figure in the automotive industry, Mircea Gradu, Ph.D., began his one-year term as SAE President in January amid monumental technology shifts that are rocking the automotive industry. SAE has been a source of progress and stability almost since the industry's inception, and looks assured to continue in that way under Gradu's leadership. The Senior Vice President of Validation/Chief Quality Officer at Velodyne LiDAR Inc. is eager to get going.
Ponticel, Patrick
Automotive Engineering: February 201818AUTP022/1/2018
Leaf turns to the 2020s Nissan's pioneering battery-EV finally gets styling, propulsion and feature upgrades that put it back in the thick of the expanding electric-car race. A rebel's guide to chassis engineering Before the advent of FEA, engineers used their hands to study chassis strength and stiffness. The approach still offers benefits today, notes the man who wrote chassis engineering's "bible." Auto tech comes on strong at CES 2018 The automotive sector showed up at CES with autonomy announcements by the score and a spectrum of new-technology unveilings. New SAE President focused on Big Data Mircea Gradu, senior VP and quality chief at lidar specialist Velodyne, targets data analytics, electrification and connectivity, and STEM initiatives during his 2018 term. Editorial: Tapping the brakes on autonomy Supplier Eye Meeting the less-is-more challenge of global scale The Navigator Autonomous business models vs. current cash generators Plug-in hybrid brings Honda into sharper Clarity 2018 Detroit auto show: Achates Power OP engine begins real-world test phase Workhorse multiphysics tool gets upgrades Toyota's new Structural Blue paint shimmers on Lexus LC500 Chevy unveils larger, lighter 2019 Silverado at Detroit auto show 2018 Detroit auto show: Ford Ranger returns 2019 XC40 showcases Volvo's new CMA platform What We're Driving Q&A Dassault's Mike LaLande
Tech Briefs April 201818AERP022/1/2018
Multi-Domain Command and Control (MDC2) Changing the Face of Modern Warfare Loop Thermosyphons Gravity-Driven Two-Phase Cooling for the 21st Century CompactPCI Serial Space A New Embedded System Specification Takes on Extreme Environments Plain Bearings for Aerospace Applications Additive Manufacturing Materials for RF Components NASA CubeSats: Pushing the Boundaries of Technology Content Addressable Memory (CAM) Technologies for Big Data and Intelligent Electronics Enabled By Magneto-Electric Ternary CAM New associative memory approach based on the propagation of spikes provides ultra-low energy search operations. Natural DNA-Based Nonvolatile Resistive Switching Memory Reliable memory devices can be realized within a single macromolecule based on natural DNA. A Mechanistic Analysis of Oxygen Vacancy Driven Conductive Filament Formation in Resistive Random Access Memory Metal/NiO/Metal Structures Study could lead to more efficient electrically switchable resistive random access memory devices. pH-Dependent Spin State Population and 19F NMR Chemical Shift Via Remote Ligand Protonation in An Iron(II) Complex An FeII complex that features a pH-dependent spin state population demonstrates potential as a 19F chemical shift-based pH sensor.
SAE Truck & Off-Highway Engineering: February 201818TOFHP022/1/2018
Electrification not a one-size-fits-all solution Efforts in the off-highway industry have been under way for decades, but the technology still faces implementation challenges. SuperTruck redux A year plus into phase two of the promising DOE program to dramatically improve freight efficiency, project leads from three of the participating truck OEMs share their progress thus far and technology paths moving forward. From research to reality Danfoss works closely with off-highway OEMs throughout the development process, testing and demonstrating machines packed with new technology features. Big Data a focus for 2018 SAE President Mircea Gradu Velodyne CQO sees use of Big Data as a way to capitalize on multiple industry trends including vehicle electrification and automated vehicles. Editorial Commercial vehicles invade CES Komatsu building AI into construction sites Gaining the Manufacturing 4.0 advantage with data-driven in-process testing Paccar reveals new and future tech at CES 2018 ChemSEI-Linker extends lithium battery life, increases EV range Powertrain mix for CVs in Europe set to diversify, but cost is critical Toyota unveils autonomous e-commerce concept vehicle, development alliance Q&A Director of the U.S. Army Tank Automotive Research, Development and Engineering Center (TARDEC) Dr. Paul D. Rogers talks strategic direction for the multitude of investments related to 270-plus Army systems
Content Addressable Memory (CAM) Technologies for Big Data and Intelligent Electronics Enabled By Magneto-Electric Ternary CAMTBMG-283852/1/2018
Content addressable memory (CAM) is one of the most promising hardware solutions for high-speed data searching and has many practical applications such as anti-virus scanners, internet protocol (IP) filters, and network switches. Since CAM stores the data in its internal memory elements and compares them with the search data in parallel, it can achieve much faster speed compared to the software lookup.
Fully Automated Quality Control of Cylinder Bores from Internal Combustion Engines and Its Implications for Industry 4.02017-36-008211/7/2017
Internet has transformed all industries and the automotive sector is on its list. It is true that manufacturing has experienced great advances in recent years, but the massive use of internet in industry is about to revolutionize it once again. The Internet of Things (IoT), the Big Data Analytics and the use of RFID technology will revolutionize manufacturing, giving to the so called “smart factories” the ability for self-diagnosis, self-configuration and self-optimization. That is what we call Industry 4.0, or the fourth industrial revolution. However, how will Industry 4.0 affect automakers and end users of vehicles? What are the challenges to bring Industry 4.0 innovations to the manufacturing industry? The present work discusses future trends in engine manufacturing, focusing on the quality control of its main components. Also, a fully automated inspection technique for quality control of cylinder bores from internal combustion engines is presented. The proposed method, which is based upon topography decomposition and multiscale analysis, was used to compare two honing variants. The results show that standardized roughness parameters are not enough to properly evaluate honed surfaces. Finally, this work shows that the proposed methodology for characterization of honed surfaces can not only increase quality and reliability of engine components, but can also allow more significant advances in engine optimization. Thereby, the described method can support automakers to attend the demands of increasingly stringent markets.
Obara, R.B., Guedes, L.C.
CornerStone Knowledge Acquisition and Synthesis FrameworkTBMG-275429/1/2017
In general, and especially in the “big data” era, there is often a failure to collect sufficient data about the data (metadata). This lack of metadata drastically reduces the potential use of the data, and the attempt to rectify this situation after the fact is often difficult, if not impossible. The necessary metadata must first be located, and then organized and associated with the data. Often those who know about the metadata are no longer available. The proposed alternative is to prepare the data for long-term preservation based on currently accepted best-practices for long-term digital archives. This means that both the data and the metadata are simultaneously collected and prepared for the archive. This methodology provides the best chance that the necessary metadata will be available for future analysis, and ensures use of the data being collected and preserved.
Data Optimization Technology Helps Wind Farm Operator Examine CostsTBMG-271907/1/2017
Seeq Seattle, WA For more info click
IoT connections transform manufacturing, vehicle validation17TOFHP06_046/1/2017
Commercial vehicles are at the forefront of the Internet of Things (IoT) movement, leveraging the connectivity that many have deployed for years. Equipment manufacturers are using IoT connections to improve their manufacturing operations, upgrade diagnostics to prognostics and improve maintenance operations. Fleet managers have been benefiting from telematic links for years, monitoring trucks and off-highway vehicles to improve efficiency. Before long, most new equipment buyers will be able to benefit from connectivity.
Costlow, Terry
Intelligent Software Enables Design of Greener, More Efficient Construction VehiclesTBMG-266504/1/2017
University of Warwick (UK) engineers are optimizing the fuel economy of next-generation offhighway vehicles by introducing new intelligent power systems for improved engine operation. This could lead to significant fuel savings and fewer carbon emissions for the industry. The team is analyzing a current fleet of construction vehicles to better understand the opportunities for emissions reduction and intelligent control.
Cloud-Driven Traffic Monitoring and Control Based on Smart Virtual Infrastructure2017-01-00923/28/2017
The new cyber-technological culture of the transport control based on virtual road signs and streetlight signals on the screen of car is the future of Humanity. A cyber-physical system (CPS) Smart Cloud Traffic Control, which realizes the mentioned culture, is proposed; it is characterized by the presence of the digitized regulatory rules, vehicles, infrastructure components, and also accurate monitoring, active cloud streetlight-free cyber control of road users, traffic lights, automatic output of operational regulatory actions (virtual traffic signs and traffic signals) to monitor of each vehicle. The main components of the cyber-physical system are the following: infrastructure, road users and rules, which have digital representation in cyberspace to realize a route, based on digital monitoring and cloud mobile control. We offer innovative services, which implement digital monitoring and cloud control as a scalable prototype of a global system that uses the following technology: precise positioning of moving and stationary objects, digital cartography, cyber-physical systems and Internet of Things, Advanced Wireless Communication and Big Data Analytics. The basic idea is to improve the quality and safety of traffic through the implementation of metric regulation of traffic, based on digital monitoring and active cloud cyber control, and also the use of intelligent virtual traffic lights and signs, which gives an opportunity to significantly improve the comfort of a car trip, reduce the overhead in time and cost of route execution. Components for the implementation of global cloud traffic control services are the following: 1) Smart is the definition of the process or phenomenon associated with the network interaction of the addressable system components in time and space between themselves and the environment, based on self-learning technologies to achieve their goals. 2) The Smart Cyber-Physical System is a set of communicatively connected to the network addressable virtual and real components in the digitized metric space with features of adequate physical monitoring, optimal cloud control and self-learning in real time to achieve their goals. 3) Internet of Things is a structure of cyber-physical systems, combining the centers of large data, knowledge, services and applications aimed to monitor and control of smart processes and phenomena in the digitized physical space by using the sensors actuators to provide high standards of living and saving the planet environment. 4) Computing is a branch of knowledge, focused on research, design and application of systems, networks and cloud-mobile services for monitoring and control of cyber-physical processes and phenomena. The development of computing, the main function of which is cyber control, should only be considered in conjunction with the real or the physical world, a part of which is humanity. There is interaction between the two worlds, the real and the virtual ones: humanity always poorly manages the real world and creates computing as his assistant. As a perfect mechanism, computing takes control of technological processes in humanity. 5) The market feasibility of the global cloud services for traffic control without physical infrastructure, traffic lights and road signs is at least 100 billion dollars. The economic effect of the transfer of road infrastructure in cyberspace, including the license plates is 500 billion dollars a year.
Hahanov, Vladimir, Gharibi, Wajeb, Litvinova, Eugenia, Chumachenko, Svitlana, Ziarmand, Arthur, Englesi, Irina, Gritsuk, Igor, Volkov, Vladimir, Khakhanova, Anastasiia
Secure and Privacy-Preserving Data Collection Mechanisms for Connected Vehicles2017-01-16603/28/2017
Nowadays, the automotive industry is experiencing the advent of unprecedented applications with connected devices, such as identifying safe users for insurance companies or assessing vehicle health. To enable such applications, driving behavior data are collected from vehicles and provided to third parties (e.g., insurance firms, car sharing businesses, healthcare providers). In the new wave of IoT (Internet of Things), driving statistics and users’ data generated from wearable devices can be exploited to better assess driving behaviors and construct driver models. We propose a framework for securely collecting data from multiple sources (e.g., vehicles and brought-in devices) and integrating them in the cloud to enable next-generation services with guaranteed user privacy protection. To achieve this goal, we design fine-grained privacy-aware data collection and upload policies that balance between enforcing privacy requirements and optimizing resource consumption (e.g., processing, network bandwidth). The optimal policy will be determined by the privacy index of the integrated multi-source data to be used by the specific service and the desired resource usage. Real-world experiments and privacy leakage analysis are conducted to address privacy issues in vehicle data collection and integration, raise public awareness around privacy leakage, and validate the proposed system.
Li, Huaxin, Ma, Di, Medjahed, Brahim, Wang, Qianyi, Kim, Yu Seung, Mitra, Pramita
Statistical Characterization, Pattern Identification, and Analysis of Big Data2017-01-02363/28/2017
In the Big Data era, the capability in statistical and probabilistic data characterization, data pattern identification, data modeling and analysis is critical to understand the data, to find the trends in the data, and to make better use of the data. In this paper the fundamental probability concepts and several commonly used probabilistic distribution functions, such as the Weibull for spectrum events and the Pareto for extreme/rare events, are described first. An event quadrant is subsequently established based on the commonality/rarity and impact/effect of the probabilistic events. Level of measurement, which is the key for quantitative measurement of the data, is also discussed based on the framework of probability. The damage density function, which is a measure of the relative damage contribution of each constituent is proposed. The new measure demonstrates its capability in distinguishing between the extreme/rare events and the spectrum events. Several case studies including vehicle reliability, vehicle road test score, warranty, salary distribution of an institution, the city population distribution in 3 countries, and the earthquake distribution worldwide and in the USA, are provided to demonstrate the role of the statistical and probabilistic approaches in the characterization and analysis of the big data.
Wei, Zhigang, Nikbin, Kamran
Automotive Engineering: November 3, 201616AUTP1111/3/2016
SAE Convergence 2016 Talk of the healthy aspects of disruption mingles with SAE's renowned technical emphasis to foster the auto industry's continuing evolution toward electrification and autonomy. The Battery Man Speaks The speed of progress in automotive lithium batteries has impressed AABC's Dr. Menahem Anderman. So has silicon-graphite anode technology development from Tesla and Panasonic. Industry 4.0: The smart factory arrives The plants that produce automotive systems and vehicles are increasingly employing intelligent systems, Big Data and advanced analytics to improve quality, safety and efficiency. Editorial: Promise of 48 volts is no shock Nissan unveils variable-compression-ratio ICE for 2018 Infiniti production model Optimizing engine oil warm-up strategies for 'real-world' driving In search of higher-energy-content batteries Making Multiphysics fast and convenient I.D. Concept heralds VW's electric Bolt fighter for 2020 Aero-slick Land Rover Discovery sheds 1058 lb, gains features Honda's all-new 2017 CR-V: Larger, lighter, turbocharged Ram Rebel TRX concept has 2017 Ford Raptor in its sights Bob Lutz on what makes a great engineer
Fiber Optic Multi-Sensing Platforms16AERP10_0410/1/2016
Enabling Innovation Across Aerospace Organizations Eventually, technology advances to the point where solutions that have been “good enough” for decades are no longer “good enough” for the innovations of today. The philosophy of “good enough” is widely applied when developing a new product or solution. Businesses have to make decisions about what new technology features will receive the most attention to keep projects within scope and completed on time. This philosophy has a downside when applied to testing and monitoring. For decades, point sensing solutions like strain gauges and thermocouples have been “good enough” for testing aircraft durability, however, the mindset of “good enough” often blocks innovation. It is not uncommon for researchers to find that they have innovated beyond their ability to test with legacy technologies.
Fiber Optic Multi-Sensing PlatformsTBMG-2555810/1/2016
Eventually, technology advances to the point where solutions that have been “good enough” for decades are no longer “good enough” for the innovations of today. The philosophy of “good enough” is widely applied when developing a new product or solution. Businesses have to make decisions about what new technology features will receive the most attention to keep projects within scope and completed on time.
Data Encoding and Parallelization Porting Techniques to Transform Binary Data Formats to Hadoop/MapReduceTBMG-245755/1/2016
The objective of this invention is to transform applications that process big data in arbitrary formats using the MapReduce parallel programming model and executed on the open-source Hadoop platform. Many legacy data-intensive applications do not scale out of a single server. Some applications can load-balance to multicore, but scalability is limited by intensive IO. Hadoop/MapReduce is a massively scalable programming paradigm modeled after Google’s implementation. There are many use cases to show Hadoop/MapReduce can scale linearly to hundreds or even thousands of servers for big data analysis. The problem is how to transparently transform a legacy serial data processing code to take full advantage of Hadoop/MapReduce parallel processing framework.
Big-Data Based Online State of Charge Estimation and Energy Consumption Prediction for Electric Vehicles2016-01-12004/5/2016
Whether the available energy of the on-board battery pack is enough for the driver’s next trip is a major contributor in slowing the growth rate of Electric Vehicles (EVs). What’s more, the actual capacity of the battery pack depend on so many factors that a real-time estimation of the state of charge of the battery pack is often difficult. We proposed a big-data based algorithm to build a battery pack dynamic model for the online state of charge estimation and a stochastic model for the energy consumption prediction. And the good performance of sensors, high-bandwidth communication systems and cloud servers make it convenient to measure and collect the related data, which are grouped into three categories: standard, historical and real-time data. First a resistance-capacitance ( RC )-equivalent circuit is taken consideration to simplify the battery dynamics. And the nonlinear relationship between the open-circuit voltage (Voc ) and the state of charge ( SOC ) is described by five linear piecewise functions, achieving good fitting effect and accuracy. The moving window recursive least square algorithm is utilized to identify the parameters of the battery model online with the historical and real-time data. A Luenberger state observer is used for online estimation of SOC, the estimation accuracy of which is not dependent on the actual capacity’s value. Afterwards, giving more insight into the relationship between the received data from different sources and the estimated SOC, we take the objective law of energy consumption as a Gaussian distribution, which can provide the confidence interval analysis to confine decision-making about whether to have the trip to the driver.
Zhang, Zhiyun, Huang, Miaohua, Chen, Yupu, Gao, Dong
Generation and Usage of Virtual Data for the Development of Perception Algorithms Using Vision2016-01-01704/5/2016
Camera data generated in a 3D virtual environment has been used to train object detection and identification algorithms. 40 common US road traffic signs were used as the objects of interest during the investigation of these methods. Traffic signs were placed randomly alongside the road in front of a camera in a virtual driving environment, after the camera itself was randomly placed along the road at an appropriate height for a camera located on a vehicle’s rear view mirror. In order to best represent the real world, effects such as shadows, occlusions, washout/fade, skew, rotations, reflections, fog, rain, snow and varied illumination were randomly included in the generated data. Images were generated at a rate of approximately one thousand per minute, and the image data was automatically annotated with the true location of each sign within each image, to facilitate supervised learning as well as testing of the trained algorithms. A deep convolutional neural network was built using 8 hidden layers, 1.5 million free parameters, and 250,000 neurons, with unique configurations optimal for traffic sign classification. This network was then trained using the above mentioned dataset. A high cross-validation accuracy of 98% with stable k-fold validation energy was achieved. This network, trained using virtual images, was then tested on real-world images with promising results, and the network was able to consistently classify signs that appear much smaller and farther away than those in the images it was trained on. The algorithm also attempted to classify signs for which it had not been trained, and predictably classified such signs using the most similar label.
Nariyambut Murali, Vidya, Micks, Ashley, Goh, Madeline J., Liu, Dongran
A Robust, Real-Time Collaboration Technology for Decision Support using Common Operating Picture (COP) EnvironmentsTBMG-242724/1/2016
StormCenter Communications, Inc. software takes advantage of cloud computing solutions to solve the growing need for real-time collaboration when accessing a Common Operating Picture (COP) from any device without having to give up all the vital tools and display functionality; all this is performed without disrupting current workflows. A COP is a platform shared by multiple stakeholders to provide a single, identical display of relevant operational information (i.e., mapping information, network operations and cyber security, sensors, weather, critical infrastructure) and tools to mark and manipulate the data. In the era of Big Data and the need for improved decision-making with the presence of multiple stakeholders across different disciplines, collaboration when using a COP is essential to success. This is especially important when dealing with decision-makers who are frequently non-data experts who need to take action immediately based on their situational awareness.
DAWN: a Simulation Model for Evaluating Costs and Tradeoffs of Big Data Science ArchitecturesTBMG-239022/1/2016
Data science is emerging as a critical area of research and technology to advance scientific discovery, knowledge, and decision-making through systematic computational approaches to analyzing massive data sets. The sheer volume of data increase, coupled with the highly distributed and heterogeneous nature of scientific data sets, is requiring new approaches to how the data will be ultimately managed and analyzed. This requires evaluating the scalability and distribution of complex software architectures. DAWN (Distributed Analytics, Workflows and Numeric) is a model for simulating the execution of data processing workflows on arbitrary data system architectures. DAWN was developed to provide NASA and the scientific community at large with an additional tool to prepare for the upcoming Big Data deluge in science.
The Advanced Data Analytics Platform (ADAPT): Concept, Design, Architecture, and OperationTBMG-238822/1/2016
NASA scientists are uniquely positioned to research and understand the processes affecting the Earth’s climate. To study these important processes, scientists must address the Big Data challenges posed by working with massive amounts of observational and climate model output data. The Advanced Data Analytics Platform (ADAPT) is a cyber infrastructure resource specifically designed to reduce the friction between scientists and data. The system includes a high-performance storage cloud surrounded by large-scale compute resources. A very high-performing network enables fast access to the data stored within ADAPT. Furthermore, the system allows users to bring their applications to the data and define the environment in which those applications run. The science results can then be stored for future analysis or shared through static and dynamic data services within ADAPT without having to move the data or make additional copies. The agility, flexibility, and extensibility of the system make it ideal for NASA scientists to produce science results quickly by analyzing large data sets.
Items per page:
1 – 50 of 62