Search

In superconducting magnets, the irreversible transition of a portion of the conductor to resistive state is called a “quench.” Having large stored energy, magnets can be damaged by quenches due to localized heating, high voltage, or large force transients. Unfortunately, current quench protection systems can only detect a quench after it happens, and mitigating risks in Low Temperature Superconducting (LTS) accelerator magnets often requires fast response (down to ms). Additionally, protection of High Temperature Superconducting (HTS) magnets is still suffering from prohibitively slow quench detection. In this study, we lay the groundwork for a quench prediction system using an auto-encoder fully-connected deep neural network. After dynamically trained with data features extracted from acoustic sensors around the magnet, the system detects anomalous events seconds before the quench in most of our data. While the exact nature of the events is under investigation, we show that the system can “forecast” a quench before it happens under magnet training conditions through a randomized experiment. This opens up the way of integrated data processing, potentially leading to faster and better diagnostics and detection of magnet quenches. [ABSTRACT FROM AUTHOR]

Energy footprinting has the potential to raise awareness of energy consumption and lead to energy-saving behavior. However, current methods are largely restricted to single buildings; these methods require energy and occupancy monitoring sensor deployments, which can be expensive and difficult to deploy at scale. Further, current methods for estimating energy consumption and population at scale cannot provide fine enough temporal or spatial granularity for a reasonable personal energy footprint estimate. In this work, we present a data-driven system for city-wide estimation of personal energy footprints. This system takes advantage of existing sensing infrastructure and data sources in urban cities to provide energy and population estimates at the building level, even in built environments that do not have existing or accessible energy or population data. [ABSTRACT FROM AUTHOR]

The identification of plant species is fundamental for the effective study and management of biodiversity. In a manual identification process, different characteristics of plants are measured as identification keys which are examined sequentially and adaptively to identify plant species. However, the manual process is laborious and time-consuming. Recently, technological development has called for more efficient methods to meet species’ identification requirements, such as developing digital-image-processing and pattern-recognition techniques. Despite several existing studies, there are still challenges in automating the identification of plant species accurately. This study proposed designing and developing an automated real-time plant species identification system of medicinal plants found across the Borneo region. The system is composed of a computer vision system that is used for training and testing a deep learning model, a knowledge base that acts as a dynamic database for storing plant images, together with auxiliary data, and a front-end mobile application as a user interface to the identification and feedback system. For the plant species identification task, an EfficientNet-B1-based deep learning model was adapted and trained/tested on a combined public and private plant species dataset. The proposed model achieved 87% and 84% Top-1 accuracies on a test set for the private and public datasets, respectively, which is more than a 10% accuracy improvement compared to the baseline model. During real-time system testing on the actual samples, using our mobile application, the accuracy slightly dropped to 78.5% (Top-1) and 82.6% (Top-5), which may be related to training data and testing conditions variability. A unique feature of the study is the provision of crowdsourcing feedback and geo-mapping of the species in the Borneo region, with the help of the mobile application. Nevertheless, the proposed system showed a promising direction toward real-time plant species identification system.

This article proposes a low-computation framework for on-board calculation of energy-optimal cruising speed, and uses the framework to investigate the critical parameters for energy-optimal cruising speed determination. The main features of the proposed framework are as follows: first, inclusion of all internal and external vehicle losses, particularly the accessory loads, which have a large impact on optimal cruising speed and have not been investigated in other works, second, use of accurate motor/inverter loss look-up-tables rather than approximated polynomial models, and, third, no requirement for connected data or knowledge of the future route, as the optimal cruising speed is calculated precisely for the current state of the vehicle and surroundings. To validate the framework, three electric vehicle models are developed in MATLAB/Simulink and tuned to match real-world data for a Chevrolet Bolt, Chevrolet Spark, and Tesla Model S. Furthermore, an algorithm is developed to determine near-optimal speed transitions when the optimal cruising speed changes. [ABSTRACT FROM AUTHOR]

2-D sparse arrays may push the development of low-cost 3-D systems, not needing to control thousands of elements by expensive application-specific integrated circuits (ASICs). However, there is still some concern about their suitability in applications, such as Doppler investigation, which inherently involve poor signal-to-noise ratios (SNRs). In this article, a novel real-time 3-D pulsed-wave (PW) Doppler system, based on a 256-element 2-D spiral array, is presented. Coded transmission (TX) and matched filtering were implemented to improve the system SNR. Standard sonograms as well as multigate spectral Doppler (MSD) profiles, along lines that can be arbitrarily located in different planes, are presented. The performance of the system was assessed quantitatively on experimental data obtained from a straight tube flow phantom. An SNR increase of 11.4 dB was measured by transmitting linear chirps instead of standard sinusoidal bursts. For a qualitative assessment of the system performance in more realistic conditions, an anthropomorphic phantom of the carotid arteries was used. Finally, real-time B-mode and MSD images were obtained from healthy volunteers. [ABSTRACT FROM AUTHOR]

Harmful Algal Blooms (HAB) pose significant challenges to fisheries management and food and water security. The onset of a HAB is notoriously difficult to predict. Traditional methods of algal species identification under a microscope are also laborious and time consuming. A real-time system for identification and concentration measurement of algal bloom species has been developed at a marine fish culture zone (FCZ) in the subtropical coastal waters of Hong Kong. The system is based on analysis of high frequency algal cell images obtained from an underwater Imaging FlowCytobot (IFCB) deployed on the fish farm. An explainable supervised machine learning technique has been successfuly developed. The algal species classifier is trained by presenting a wide range of extracted image features to a random forest algorithm. An optimized set of 25 features is identified by a recursive feature elimination technique. The random forest (RF) classifier can identify 15 target HAB classes with an overall out-of-bag accuracy of 94.2%, with individual F 1 score ranging from 0.8 to 1.0. The classifier performs equally well as a Convolution Neural Network (CNN) developed using transfer learning techniques. Based on the classifier, an automated real-time species identification and cell counting protocol has been developed, with a response time of 10 min after data collection. This work represents the first successful attempt of continuous algal species monitoring by IFCB and artificial intelligence (AI) – based detection of HAB in subtropical coastal waters. [ABSTRACT FROM AUTHOR]

Shared resources on the multicore chip, such as main memory, are increasingly becoming a point of contention. Traditional real-time task scheduling policies focus on solely on the CPU, and do not take in account memory access and cache effects. In this paper, we propose parallel real-time tasks scheduling (PRTTS) policy on multicore platforms. Each set of tasks is represented as a directed acyclic graph (DAG). The priorities of tasks are assigned according to task periods Rate Monotonic (RM). Each task is composed of three phases. The first phase is read memory stage, the second phase is execution phase and the third phase is write memory phase. The tasks use locks and critical sections to protect data access. The global scheduler maintains the task pool in which tasks are ready to be executed which can run on any core. PRTTS scheduling policy consists of two levels: the first level scheduling schedules ready real-time tasks in the task pool to cores, and the second level scheduling schedules real-time tasks on cores. Tasks can preempt the core on running tasks of low priority. The priorities of tasks which want to access memory are dynamically increased above all tasks that do not access memory. When the data accessed by a task is in the cache, the priority of the task is raised to the highest priority, and the task is scheduled immediately to preempt the core on running the task not accessing memory. After accessing memory, the priority of these tasks is restored to the original priority and these tasks are pended, the preempted task continues to run on the core. This paper analyzes the schedulability of PRTTS scheduling policy. We derive an upper-bound on the worst-case response-time for parallel real-time tasks. A series of extensive simulation experiments have been performed to evaluate the performance of proposed PRTTS scheduling policy. The results of simulation experiment show that PRTTS scheduling policy offers better performance in terms of core utilization and schedulability rate of tasks. [ABSTRACT FROM AUTHOR]

The rapid development of various types of real-time control systems raise new challenges on their heterogeneity and knowledge explicitly sharing issues. In this study, we propose an ontology-based model, named OntoEvent, to define and detect complex event in high-speed train control system. OntoEvent defines control logics using ontology structure and describes functionalities using logical, temporal operators and attribute relations. This ontology-based event processing model supports dynamic reconfiguration of functions and sharing between different components of the railway system. A pipelined construction framework is designed to transform OntoEvent model into semantic-consistent detection model. We implement a prototype control system, to evaluate the efficiency and performance of OntoEvent. Experimental results on this prototype system prove that OntoEvent-based event detection model outperforms other two selected models in results correctness, processing throughput and real-time performance, especially when processing a large amount of complex events. [ABSTRACT FROM AUTHOR]

Real-time simulation-based validation plays an essential role in the early stage development of electric traction systems. The high-fidelity real-time simulation relies on the accurate modeling of the power electronics and motors and usually represents the nonlinear characteristics as many as possible while obeying the computing time constraints. In this paper, an artificial neural network (ANN) aided modeling approach is proposed for the field-programmable gate array (FPGA)-based real-time simulation to deal with the nonlinearities in permanent magnet synchronous motor (PMSM) and its drive. With the help of the ANN, the switches power losses, and the PMSM nonlinear flux linkage and electromagnetic torque can be modeled. An electro-thermal model of inverters and the finite element analysis (FEA)-based PMSM model are thus enabled in the FPGA-based real-time simulation, which can significantly improve the performances of the real-time simulation-based test bench. A two-level inverter fed PMSM drive system is modeled using the ANN-aided modeling approach and simulated in the National Instrument PXIe FlexRIO FPGA real-time system. The accuracy and effectiveness of the proposed approaches are tested and validated by comparing the results with the offline simulation tools.

Over the last few decades, the accuracy requirements for wind tunnel testing have become much more stringent as aircraft industry strives to achieve the best performance of their products. Investments in advanced capabilities are necessary to obtain more accurate test results more efficiently, ensuring that the wind tunnels stay productive well into the future. The VTI Belgrade T-38 wind tunnel responded to these challenges by applying a hierarchical approach to design a distributed multilevel control system, a part of which is the variable-geometry nozzle positioning system, presented in this paper. It follows the hierarchy of the entire wind tunnel control system, with critical positioning loops at the field-programmable gate array (FPGA), and the user interface with real-time data analysis capability implemented on a central wind tunnel computer. In addition to more accurate positioning, the FPGA-based control system significantly improved overall operation efficiency, speed and reliability. Since the nozzle geometry uniquely determines supersonic Mach number, the improved positioning accuracy is verified in wind tunnel tests.

Since of the advent of Industry 4.0, embedded systems have become an indispensable component of our life. However, one of the most significant disadvantages of these gadgets is their high power consumption. It was demonstrated that making efficient use of the device’s central processing unit (CPU) enhances its energy efficiency. The use of the particle swarm optimization (PSO) over an embedded environment achieves many resource problems. Difficulties of online implementation arise primarily from the unavoidable lengthy simulation time to evaluate a candidate solution. In this paper, an embedded two-level PSO (E2L-PSO) for intelligent real-time simulation is introduced. This algorithm is proposed to be executed online and adapted to embedded applications. An automatic adaptation of the asynchronous embedded two-level PSO algorithm to CPU is completed. The Flexible Job Shop Scheduling Problem (FJSP) is selected to solve, due to its importance in the Industry 4.0 era. An analysis of the run-time performance on handling E2L-PSO over an STM32F407VG-Discovery card and a Raspberry Pi B+ card is conducted. By the experimental study, such optimization decreases the CPU time consumption by 10% to 70%, according to the CPU reduction needed (soft, medium, or hard reduction).

Abstract Sensing the number of people occupying a building in real-time facilitates a number of pervasive applications within the area of building energy optimization and adaptive control. To ascertain occupant counts, the adoption of camera-based sensors i.e. 3D stereo-vision and thermal cameras have grown significantly. However, camera-based sensors can only produce occupant counts with accumulating errors. Existing methods for correcting such errors can only correct erroneous count data at the end of the day and not in real-time. However, many applications depend on real-time corrected counts. In this paper, we present an algorithm named PreCount for accurately correcting raw counts in real-time. The core idea of PreCount is to learn error estimates from the past. We evaluated the accuracy of the PreCount algorithm using datasets from four buildings. Also, the Normalized Root Mean Squared Error was used to evaluate the performance of PreCount. Our evaluation results show that in real-time PreCount achieved a significantly lower Normalized Root Mean Squared Error compared to raw counts and other correction approach with a maximum error reduction of 68% when benchmarked with ground truth data. By presenting a more accurate algorithm for estimating occupant counts in real-time, we hope to enable buildings to better serve the actual number of people to improve both occupant comfort and energy efficiency.

Aiming at addressing the contradiction between the high-speed real-time positioning and multi-channel signal processing in multi-beam sonar systems, in this work we present a real-time multi-beam sonar system based on a Field Programmable Gate Array (FPGA) and Digital Signal Processing (DSP) from two perspectives, i.e., hardware implementation and software optimization. In terms of hardware, an efficient high-voltage pulse transmitting module and a multi-channel data acquisition module with time versus gain (TVG) compensation with characteristics such as low noise and high phase amplitude consistency, are proposed. In terms of algorithms, we study three beamforming methods, namely delay-and-sum (D&S), direct-method (DM) and Chirp Zeta Transform (CZT). We compare the computational efficiency of DM and CZT in the digital domain. In terms of software, according to the transmission bandwidth of the Gigabit Ethernet and a serial rapid IO (SRIO) interface, the data transmission paths of the acquired data and the beam pattern between the FPGA, the DSP, and a personal computer (PC) are planned. A master-slave multi-core pipelined signal processing architecture is designed based on DSP, which enhances the data throughput of the signal processor by seven times as compared with that of the single-core operation. The experimental results reveal that the sound source level of the transmitting module is around 190.25 dB, the transmitting beam width is 64° × 64°, the background noise of the acquisition module is less than 4 μVrms, the amplitude consistency error of each channel is less than −6.55 dB, and the phase consistency error is less than 0.2°. It is noteworthy that the beam number of the sonar system is 90 × 90, the scanning angle interval is 0.33°, the working distance ranges from 5 m to 40 m, and the maximum distance resolution is 0.384 m. In the positioning experiment performed in this work; the 3-D real-time position of the baffle placed in the detection sector is realized. Please note that the maximum deviation of azimuth is 2°, the maximum deviation of elevation is 2.3°, and the maximum distance deviation is 0.379 m. [ABSTRACT FROM AUTHOR]

In this work, six voiced/unvoiced speech classifiers based on the autocorrelation function (ACF), average magnitude difference function (AMDF), cepstrum, weighted ACF (WACF), zero crossing rate and energy of the signal (ZCR‐E), and neural networks (NNs) have been simulated and implemented in real time using the TMS320C6713 DSP starter kit. These speech classifiers have been integrated into a linear‐predictive‐coding‐based speech analysis‐synthesis system and their performance has been compared in terms of the percentage of the voiced/unvoiced classification accuracy, speech quality, and computation time. The results of the percentage of the voiced/unvoiced classification accuracy and speech quality show that the NN‐based speech classifier performs better than the ACF‐, AMDF‐, cepstrum‐, WACF‐ and ZCR‐E‐based speech classifiers for both clean and noisy environments. The computation time results show that the AMDF‐based speech classifier is computationally simple, and thus its computation time is less than that of other speech classifiers, while that of the NN‐based speech classifier is greater compared with other classifiers. [ABSTRACT FROM AUTHOR]

This paper deals with real-time scheduling of homogeneous multicore platforms powered by a battery to be periodically recharged. A system is composed of reconfigurable real-time dependent and periodic tasks to be assigned to different cores interconnected by a network-on-chip (NoC). The system is subject to reconfigurations, which are automatic operations allowing the addition and/or removal of tasks as well as their exchanged messages on the NoC. Consequently, any reconfiguration can violate real-time and energy constraints on cores as well as the NoC when the energy is unavailable until the next recharge. A novel periodic task model based on elastic coefficients and a new scheduling strategy are proposed to compute useful temporal parameters allowing for tasks and messages to meet the related constraints while controlling the communication cost on the NoC. This strategy is compared with an integer linear programming-based optimal solution, and a tool named OptimalMappingTasks is developed to run different simulations that prove the originality of this paper’s contribution. [ABSTRACT FROM AUTHOR]

We developed a vibration measurement system by fiber Bragg gratings (FBGs) with a pulse light. The pulse light is produced by modulating a broadband light with an optical fiber switch at a modulation frequency of 40 kHz. This system enables to separate and detect reflection signals from multiple FBGs like a pulse train. This system achieves multipoint and real‐time vibration measurement by FBGs with a time resolution of 25 µs. [ABSTRACT FROM AUTHOR]

Real-time embedded platforms with resource constraints can take the benefits of mixed-criticality system where applications with different criticality-level share computational resources, with isolation in the temporal and spatial domain. A conventional software-based isolation mechanism adds additional overhead and requires certification with the highest level of criticality present in the system, which is often an expensive process. In this article, we present a different approach where the required isolation is established at the hardware-level by featuring partitions within the processor. A four-stage pipelined soft-processor with replicated resources in the data-path is introduced to establish isolation and avert interference between the partitions. A cycle-accurate scheduling mechanism is implemented in the hardware for hard-real-time partition scheduling that can accommodate different periodicity and execution time for each partition as per user needs, while preserving time-predictability at the individual application level. Applications running within a partition has no sense of the virtualization and can execute either on a host-software or directly on the hardware. The proposed architecture is implemented on FPGA thread and demonstrated with an avionics use case. [ABSTRACT FROM AUTHOR]

The integrated navigation system highly relies on the accuracy of measurements of sensors that are susceptible to unknown disturbances. In order to improve the reliability and safety of the navigation system, there is an increasing need for the fault detection of the sensors. In the present study, a hybrid data-driven fault detection strategy is proposed, which is based on residual sequence analysis. Currently, the one-class support vector machine is one of the most popular fault detection methods for navigation systems with many successful cases. Therefore, the one-class support vector machine is combined with time-series similarity measure and modified principal components analysis approaches. The similarity measurement of multi-sequence residuals between a real-time sample and normal condition samples is computed to construct learning features for one-class support vector machine. Similarly, the modified principal components analysis scheme is applied to project residuals onto subspaces and obtain learning features. Moreover, the one-class support vector machine model is applied for abnormal detection if unexpected sensor faults exhibit in measurements and residuals. Finally, experiments are carried out to evaluate the performance of the proposed strategy for abrupt faults and soft faults on navigation sensors. Experimental results show that the hybrid data-driven fault detection strategy can effectively detect these faults with short time delay and high accuracy. [ABSTRACT FROM AUTHOR]

This paper presents a framework for complete simulation and verification of Serial Digital Interface (SDI) video using a verilog test-bench and geared toward FPGAs. This framework permits simulating the entire process: from test video signal generation to protocol verification in the FPGA which implements the Device Under Test (DUT). The novelty in the design is the combination of a customized test video signal generator with an implementation clone of DUT transceiver for in-depth protocol debugging. Identical input test patterns of the video protocol under test are generated and fed to DUT for verification. Thus, the model not only permits to evaluate the SDI transport layer but also validates the implementation at ultra low pixel level of the video format. This approach provides two advantages: cost saving in terms of additional lab test equipment and delivering all-in-one test solution to verify design and implementation. A practical implementation using a test example of a macroblock processing chain using SDI video interface shows the viability of the proposed framework for video protocol testing. [ABSTRACT FROM AUTHOR]

Summary: We investigate the challenges of building an end‐to‐end cloud pipeline for real‐time intelligent visual inspection system for use in automotive manufacturing. Current methods of visual detection in automotive assembly are highly labor intensive, and thus prone to errors. An automated process is sought that can operate within the real‐time constraints of the assembly line and can reduce errors. Components of the cloud pipeline include capture of a large set of high‐definition images from a camera setup at the assembly location, transfer and storage of the images as needed, execution of object detection, and notification to a human operator when a fault is detected. The end‐to‐end execution must complete within a fixed time frame before the next car arrives in the assembly line. In this article, we report the design, development, and experimental evaluation of the tradeoffs of performance, accuracy, and scalability for a cloud system. [ABSTRACT FROM AUTHOR]

Applications that require short-range ultra-high bitrate communication, such as cable removal in virtual reality games and communication between autonomous vehicles, are examining solutions such as millimetre wave wireless (mmWave). When using mmWave, steerable directional antennas are used to mitigate the severe signal power attenuation common with high frequencies. Nonetheless, even small movements in the user device can cause a sudden drop in data-rate down (even to 0 bits/s) making mmWave channels unstable and unusable. To make the channel more stable for the aforementioned applications, which are vulnerable due to frequent blockages and fast movement, we designed and developed a robust solution based on a double link mmWave system. We duplicate the radio transceivers (RT) of a user device (UD) to increase the probability of finding line of sight to an access point (AP) representing the other side of the communication channel. The AP selects one RT of the UD for communication, based on continuous measurement of quality compared to the channel of the other RT. This concept was implemented in a laboratory environment and evaluated using a series of controlled experiments. The experiments serve to validate that using double links is feasible, and is considerably more robust and it can double the link utilization, compared to only using one mmWave link. These results show great promise for the concept, by demonstrating that using multiple mmWave links yields ultra-high bit-rate wireless communication with no disruption, even in the presence of blockages and mobility. [ABSTRACT FROM AUTHOR]

• A novel real-time counting system is developed for monitoring high pedestrian flows. • The proposed system uses distance measurements from a two-dimensional LiDAR sensor. • Clustering algorithm is developed to count pedestrians and identify their direction. • The results show that the system accurately counts more than 97% of the pedestrians. • The accuracy of the proposed system is higher than the traditional technologies. Automated monitoring of pedestrians on non-motorized facilities with high pedestrian flows is challenging. Several automated sensor solutions are commercially available that have been evaluated in the literature including traditional point-based sensors, such as inductive loop detectors for bicycles and infrared sensors for pedestrians. More recently, image-based systems, based on video cameras or thermal video cameras, have been developed. Despite the various options, some key limitations of existing solutions exist, in particular, the lack of low-cost solutions using embedded systems capable of performing in real-time under high volume (flow) conditions. This work aims at developing and evaluating the performance of a novel, real-time counting system, developed for environments with high pedestrian flows. The proposed system is based on emerging LiDAR (Light Detection and Ranging) technology. As an input, the system uses the distance measurements from a two-dimensional LiDAR sensor with a set of distinct laser channels and a given angular resolution between each channel. The developed system processes those measurements using a clustering algorithm to detect, count, and identify the direction of travel of each pedestrian. The system's performance is evaluated by comparing its directional counting outputs with manual counts (ground truth) using disaggregate and aggregate (15-minutes interval) counts at two different monitoring locations. The results demonstrate that the system accurately counts more than 97% of the pedestrians at the disaggregate level, with a false direction detection rate of 1.1%. The over-counting error is 0.7% and the under-counting errors are 1.3% and 2.7% for the two selected sites. At the aggregate level (15-minutes interval), the average absolute percentage deviations (AAPDs) are 1.6% and 4.3% while the weighted AAPDs are 1.5% and 3.5% for the first and second sites, respectively. The accuracy of the proposed system is higher than the traditional technologies used for the same purpose. [ABSTRACT FROM AUTHOR]