Hasil untuk "Machine design and drawing"

Chyan Zheng Siow, Qingwei Song, Azhar Aulia Saputra et al.

Abstract Mobility support robots can be made available for use by patients in hospital settings. To prevent the robot from being used by unauthorized patients, an identification system must be installed on the robot. The FaceNet system is unable to recognize users when the camera is positioned to view from below, and the true recognition rate is lower when the threshold is set to a higher level. To address these issues, this paper proposes an adaptive face recognition system with a higher threshold setting and the capacity to adapt to bottom camera viewing angles. To achieve high threshold settings, this paper proposes a novel training strategy that enlarges the output distribution range through Negative Feature Augmentation (NFA) when training a Siamese network. The objective of NFA is to introduce additional negative sample pairs during the training phase, thereby enabling the network to expand the recognition distribution range. This hypothesis is demonstrated on the MNIST dataset. Furthermore, the proposed method was tested on the LFW benchmark dataset, resulting in an improved recognition rate of 0.86 when the threshold was set to 0.8. Subsequently, in order to achieve the ability to adaptively recognize users from bottom perspectives, we introduce the use of Growing Neural Gas (GNG) to track feature changes when the user faces different directions while using the robot. Furthermore, we propose methodologies for aligning eyes and standardizing faces, thereby ensuring that the input face is aligned with the learned features. In our experiments, we collected RGB videos of 16 college students using RT-1, and the results demonstrated a high accuracy rate of 0.80, outperforming other methods.

Zdenka Bulková, Juraj Čamaj, Jozef Gašparík

The current technical condition of facilities designated for the technical–hygienic maintenance of railway rolling stock is unsatisfactory, as they are neither technologically nor technically equipped to meet the required quality standards. Maintenance is often carried out in open spaces or directly on the tracks of major railway junctions, which prevents year-round execution of these services and causes operational limitations. This article analyses and proposes solutions for the technical–hygienic maintenance center (THU) of railway rolling stock at the Nové Zámky railway station in Slovakia, focusing on improving the efficiency and quality of the provided services. The analysis includes an assessment of technological procedures, identification of operational deficiencies, and a comparison of current maintenance standards with the requirements for contemporary railway systems, such as automated diagnostic platforms, predictive maintenance modules, and modular cleaning infrastructure. The optimization of THU services considers the average time norms for selected technological procedures and the characteristics of train sets passing through the center. The proposed solution involves a more efficient scheduling of operations in line with the valid railway traffic timetable and train set circulation, utilizing a graphical planning method for modelling and optimizing the facility’s service processes. The implementation of optimization measures can lead to increased capacity and efficiency of maintenance, reduced time required for individual procedures, and lower operational costs. The study’s results provide practical recommendations for improving the quality of technical–hygienic maintenance at railway junction stations, contributing to greater railway transport reliability and an overall improvement in passenger comfort. Additionally, the findings offer a transferable framework that may inform the planning and modernization of maintenance facilities at other regional railway stations facing similar infrastructural and operational challenges.

Yuka Sone, Chifuyu Matsumoto, Jinseok Woo et al.

Abstract In recent years, IoT technologies have made our daily lives more convenient and comfortable. In particular, these technologies are being actively utilized in our living environments through the development of various smart home appliances. As the pace of Digital Transformation (DX) quickens and digital platforms become integrated into diverse residential settings, the notion of the home we inhabit is taking on greater significance beyond merely a place for rest. Therefore, this paper explains the development of user-friendly interfaces for smart home systems used in our living rooms. An appropriate service can be provided accordingly if the user’s intent is specifically understood. Therefore, this paper explores the relationship between the user’s gaze and the control targets of the smart home using Mixed Reality (MR) devices. In this study, we introduce a novel smart home control system from a range of options. Our proposed system enables user analysis and remote control through the utilization of an MR device capable of tracking a user’s gaze. In addition, we investigate the perception of user interfaces and analyze the survey results conducted after using the interface system, discussing the validity of the proposed system.

Yuta Murakami, Tomohito Takubo, Tetsuo Tsujioka

Abstract Global Navigation Satellite Systems is a positioning system that receives signals from satellites. RTK-GNSS positioning uses multiple satellites and base stations. In RTK-GNSS positioning, the accuracy of the positioning solution is determined by a Ratio-test, classifying it as either a accurate solution (FIX) or an inaccurate one. When classified as FIX, the positional error is within a few centimeters. However, the positioning solutions judged to be FIX may contain inaccurate data called Wrong FIX. Removing Wrong FIX is the current challenge. In this paper, we propose Wrong FIX detection method for using the geometry of the surrounding environment. The proposed method identifies Wrong FIX by leveraging the fact that point clouds obtained from two positions will precisely overlap when their relative positions are accurately aligned. Experiments were conducted to demonstrate the effectiveness of the proposed method, and it was confirmed that the accuracy was improved by comparing it with the Root Mean Squared Error and Binary Accuracy of the Ratio-test.

Jost Arndt, Utku Isil, Michael Detzel et al.

Many physical processes can be expressed through partial differential equations (PDEs). Real-world measurements of such processes are often collected at irregularly distributed points in space, which can be effectively represented as graphs; however, there are currently only a few existing datasets. Our work aims to make advancements in the field of PDE-modeling accessible to the temporal graph machine learning community, while addressing the data scarcity problem, by creating and utilizing datasets based on PDEs. In this work, we create and use synthetic datasets based on PDEs to support spatio-temporal graph modeling in machine learning for different applications. More precisely, we showcase three equations to model different types of disasters and hazards in the fields of epidemiology, atmospheric particles, and tsunami waves. Further, we show how such created datasets can be used by benchmarking several machine learning models on the epidemiological dataset. Additionally, we show how pre-training on this dataset can improve model performance on real-world epidemiological data. The presented methods enable others to create datasets and benchmarks customized to individual requirements. The source code for our methodology and the three created datasets can be found on https://github.com/github-usr-ano/Temporal_Graph_Data_PDEs.

Rahul D Ray

Physics-constrained data generation is essential for machine learning in scientific domains where real data are scarce; however, existing approaches often over-constrain models without identifying which physical components are necessary. We present a systematic ablation study of a physics-informed grating coupler spectrum generator that maps five geometric parameters to 100-point spectral responses. By selectively removing explicit energy conservation enforcement, Fabry-Perot oscillations, bandwidth variation, and noise, we uncover a physics constraint paradox: explicit energy conservation enforcement is mathematically redundant when the underlying equations are physically consistent, with constrained and unconstrained variants achieving identical conservation accuracy (mean error approximately 7 x 10^-9). In contrast, Fabry-Perot oscillations dominate threshold-based bandwidth variability, accounting for a 72 percent reduction in half-maximum bandwidth spread when removed (with bandwidth spread reduced from 132.3 nm to 37.4 nm). We further identify a subtle pitfall: standard noise-addition-plus-renormalization pipelines introduce 0.5 percent unphysical negative absorption values. The generator operates at 200 samples per second, enabling high-throughput data generation and remaining orders of magnitude faster than typical full-wave solvers reported in the literature. Finally, downstream machine learning evaluation reveals a clear physics-learnability trade-off: while central wavelength prediction remains unaffected, removing Fabry-Perot oscillations improves bandwidth prediction accuracy by 31.3 percent in R-squared and reduces RMSE by 73.8 percent. These findings provide actionable guidance for physics-informed dataset design and highlight machine learning performance as a diagnostic tool for assessing constraint relevance.

Bruno Rodrigues, Inês Ochoa, Agostinho Gomes

Calorimeters are a crucial component in modern particle detectors. They are responsible for providing accurate energy measurements of particles produced in high-energy collisions. The demanding requirements set for next-generation collider experiments impose new challenges on the design of new detectors, and a systematic approach to their optimization is increasingly necessary. The performance of calorimeters is primarily characterized by their energy resolution, parameterized by a stochastic and a constant term, related to sampling fluctuations and non-uniformities respectively. To improve the reconstruction quality of physics objects in the calorimeter, both terms need to be taken into account. Changes in a longitudinally constrained design usually result in a trade-off between these terms, making optimization a non-trivial task. This work focuses on the optimization of a hadronic sampling calorimeter, based on the FCC-ee ALLEGRO detector concept. By controlling the absorber layer thickness in a Geant4 simulation, the impact of the passive to active material proportion on the deposited energy distribution and resolution can be analyzed. Our methodology aims at exploring the design space with practical considerations, paving the way for the development of a closed optimization framework that can evaluate multiple designs against physics performance targets.

Valerian Pinzaru, Carmen Bujoreanu, Ovidiu Rapeanu

The front hood striker assembly, integral to the hood structure, experiences bending due to various factors such as repeated opening and closing, road impacts, and aerodynamic forces. This paper presents a numerical and experimental study of various striker assembly geometries, focusing on their effects on static stiffness performance. A static load is applied to generate displacement and calculate stiffness as the load-to-displacement ratio. It was discovered that by increasing the bending angle of the striker wire, an increase in section moment of inertia is achieved for the Z direction and an increase in the static stiffness on this particular direction, while for the Y direction a stiffness loss is observed for angles bigger than 95 degrees. Therefore, by improving the shape and the bending angle of the striker wire a good level of stiffness can be achieved while reducing the thickness and overall weight of the striker reinforcer.

Mo’ath Shatnawi, Maram Bani Younes

The rise in autonomous vehicles further impacts road networks and driving conditions over the road networks. Cameras and sensors allow these vehicles to gather the characteristics of their surrounding traffic. One crucial factor in this environment is the appearance of emergency vehicles, which require special rules and priorities. Machine learning and deep learning techniques are used to develop intelligent models for detecting emergency vehicles from images. Vehicles use this model to analyze regularly captured road environment photos, requiring swift actions for safety on road networks. In this work, we mainly developed a Generative Adversarial Network (GAN) model that generates new emergency vehicles. This is to introduce a comprehensive expanded dataset that assists emergency vehicles detection and classification processes. Then, using Convolutional Neural Networks (CNNs), we constructed a vehicle detection model demonstrating satisfactory performance in identifying emergency vehicles. The detection model yielded an accuracy of 90.9% using the newly generated dataset. To ensure the reliability of the dataset, we employed 10-fold cross-validation, achieving accuracy exceeding 87%. Our work highlights the significance of accurate datasets in developing intelligent models for emergency vehicle detection. Finally, we validated the accuracy of our model using an external dataset. We compared our proposed model’s performance against four other online models, all evaluated using the same external dataset. Our proposed model achieved an accuracy of 85% on the external dataset.

Christian Wacker, Niklas Dierks, Arno Kwade et al.

Abstract Innovative soft robotic grippers, such as granular grippers, enable the automated handling of a wide spectrum of different geometries, increasing the flexibility and robustness of industrial production systems. Granular grippers vary in their design as well as in their configuration, which affects the specific characteristics and capabilities regarding grippable objects. Relevant aspects are the selection of granulates and membranes, as they affect the deformability. This influences the achievable gripping forces, which vary with the gripped objects geometry. On the basis of experimental studies, the modeling of interpolations as well as through experimental validations, the present research investigates the influences of different configurations on the achievable gripping forces for a specific concept of an innovative vacuum-based granular gripper. Specifically, the focus lies on design as well as configuration parameters, which could influence the achievable gripping force. Influencing parameters are determined based on a literature review of similar gripping concepts. Various adjustment possibilities are identified, such as materials of granulates or membranes. The possible configuration options are experimentally analyzed with a one-factor-at-a-time approach. The possibility of modelling the effects of their interrelations on the achievable gripping force is examined with approaches for linear models and compared to interpolations based on Machine Learning. Especially the granulate filling level and the membrane configuration exhibit the largest influences, which were best predicted with the approach based on artificial neural networks. A selection of an optimized gripper configuration for a specified object set as well as possible further developments such as a continuous expandability of the approaches and integrations with simulations are discussed. As a result of these analyses, this research provides methodologies for an optimized selection of a gripper configuration for an improved object-specific achievable gripping force and allows for more efficient handling processes with the examined type of vacuum-based granular gripper.

Paul Arévalo, Danny Ochoa-Correa, Edisson Villa-Ávila

The increasing demand for more efficient and sustainable power systems, driven by the integration of renewable energy, underscores the critical role of energy storage systems (ESS) and electric vehicles (EVs) in optimizing microgrid operations. This paper provides a systematic literature review, conducted in accordance with the PRISMA 2020 Statement, focusing on studies published between 2014 and 2024 and sourced from Web of Science and Scopus, resulting in 97 selected works. The review highlights the potential of EVs, not only as sustainable transport solutions but also as mobile storage resources, enhancing microgrid flexibility and stability through vehicle-to-grid (V2G) systems. It also underscores the importance of advanced control strategies, such as Model Predictive Control (MPC) and hybrid AC/DC microgrids, for improving energy flow management and operational resilience. Despite these advancements, gaps remain in the comprehensive integration of ESS and EVs, particularly regarding interoperability between microgrid components and the lack of optimization frameworks that holistically address dynamic pricing, grid stability, and renewable energy integration. This paper synthesizes existing technologies and offers insights for future research aimed at advancing the sustainability, efficiency, and economic viability of microgrids.

Andrea Pugnana, Lorenzo Perini, Jesse Davis et al.

With the increasing deployment of machine learning models in many socially sensitive tasks, there is a growing demand for reliable and trustworthy predictions. One way to accomplish these requirements is to allow a model to abstain from making a prediction when there is a high risk of making an error. This requires adding a selection mechanism to the model, which selects those examples for which the model will provide a prediction. The selective classification framework aims to design a mechanism that balances the fraction of rejected predictions (i.e., the proportion of examples for which the model does not make a prediction) versus the improvement in predictive performance on the selected predictions. Multiple selective classification frameworks exist, most of which rely on deep neural network architectures. However, the empirical evaluation of the existing approaches is still limited to partial comparisons among methods and settings, providing practitioners with little insight into their relative merits. We fill this gap by benchmarking 18 baselines on a diverse set of 44 datasets that includes both image and tabular data. Moreover, there is a mix of binary and multiclass tasks. We evaluate these approaches using several criteria, including selective error rate, empirical coverage, distribution of rejected instance's classes, and performance on out-of-distribution instances. The results indicate that there is not a single clear winner among the surveyed baselines, and the best method depends on the users' objectives.

Yuzhi Xu, Haowei Ni, Qinhui Gao et al.

Computational molecular design -- the endeavor to design molecules, with various missions, aided by machine learning and molecular dynamics approaches, has been widely applied to create valuable new molecular entities, from small molecule therapeutics to protein biologics. In the small data regime, physics-based approaches model the interaction between the molecule being designed and proteins of key physiological functions, providing structural insights into the mechanism. When abundant data has been collected, a quantitative structure-activity relationship (QSAR) can be more directly constructed from experimental data, from which machine learning can distill key insights to guide the design of the next round of experiment design. Machine learning methodologies can also facilitate physical modeling, from improving the accuracy of force fields and extending them to unseen chemical spaces, to more directly enhancing the sampling on the conformational spaces. We argue that these techniques are mature enough to be applied to not just extend the longevity of life, but the beauty it manifests. In this perspective, we review the current frontiers in the research \& development of skin care products, as well as the statistical and physical toolbox applicable to addressing the challenges in this industry. Feasible interdisciplinary research projects are proposed to harness the power of machine learning tools to design innovative, effective, and inexpensive skin care products.

Candido Guido Muzio, Kaminski Paulo Carlos

Armoring civilian vehicles requires specialized knowledge and experience that many armoring companies lack as they are not direct or indirect suppliers of vehicle manufacturers. This limits their access to automotive quality and manufacturing certifications or detailed vehicle designs, which can result in loss or malfunctioning of automotive components during the armoring process. Therefore, this study aimed to investigate the challenges faced by Brazilian civilian armoring companies and identify opportunities for improvement in their production processes. Qualitative research was conducted using a questionnaire-based survey of eight specialized firms in Brazil, as well as literature related to DFMA, design for manufacturing and assembly, quality, automotive, and ballistic references. The study results include detailed armoring operation steps, qualitative survey reports, and helpful literature references for armoring practitioners to generate a standard armoring procedure for different vehicle models. Following best practices in automotive and armoring procedures collected in the survey responses can standardize and enhance ballistic protection operations while preserving the original vehicle systems' functionalities and warranties. This work provides valuable information for armoring companies to improve their operations and interfaces with automotive systems and follow automotive and ballistic references.

Ingrid J. Moreno, Dina Ouardani, Daniel Chaparro-Arce et al.

Reducing costs and time spent in experiments in the early development stages of vehicular technology such as off-road and agricultural semi-autonomous robots could help progress in this research area. In particular, evaluating path tracking strategies in the semi-autonomous operation of robots becomes challenging because of hardware costs, the time required for preparation and tests, and constraints associated with external aspects such as meteorological or weather conditions or limited space in research laboratories. This paper proposes a methodology for the real-time hardware-in-the-loop emulation of path tracking strategies in low-cost agricultural robots. This methodology enables the real-time validation of path tracking strategies before their implementation on the robot. To validate this, we propose implementing a path tracking strategy using only the information of motor’s angular speed and robot yaw velocity obtained from encoders and a low-cost inertial measurement unit (IMU), respectively. This paper provides a simulation with MATLAB/Simulink, hardware-in-the-loop with Qube-servo (Quanser), and experimental results with an Agribot platform to confirm its validity.

Al-Mutairi Nabeel Hasan, Al-Zubiedy Ali, Al-Zuhairi Ali J.

Compared to linear analogs, hyperbranched polymers (HBPs) have gotten much attention in the last decade because of their intrinsic globular topologies and distinctive features like low viscosity, high solubility, and a high degree of functionality. In this work, four types of hyperbranched polyester polymer HBPs have been synthesized using the A2+B3 polycondensation methodology. Firstly, the starting material B3 monomer (Pyrimidine-2,4,6-triol) has been synthesized using urea and malonic acid with the presence of sodium Na as the catalyst for the reaction. Secondly, four types of materials (tartaric acid TA, adipic acid AD, maleic acid MA, and phthalic anhydride PA) as A2 monomers were added to the starting material B3 monomer in an oil bath to prepare the four types of HBP. The chemical structure of HBPs was characterized by FTIR, and 1H-NMR. The molecular weight of the prepared HBPs was characterized by gel permeation chromatography GPC, and thermal properties were characterized by differential scanning calorimetry DSC and thermal gravimetric analysis TGA. FTIR results showed that there are new bands, such as C-O-C between A2 and B3 monomers, as indicated by 1H-NMR. In addition, GPC shows that the prepared polymers have a narrow molecular weight distribution with good thermal stability, as indicated by DSC and TGA.

Kevin Steinbach, Dominik Lechler, Peter Kraemer et al.

This paper presents a novel approach to address <i>noise, vibration, and harshness (NVH)</i> issues in electrically assisted bicycles (e-bikes) caused by the drive unit. By investigating and optimising the structural dynamics during early product development, <i>NVH</i> can decisively be improved and valuable resources can be saved, emphasising its significance for enhancing riding performance. The paper offers a comprehensive analysis of the e-bike drive unit’s mechanical interactions among relevant components, culminating—to the best of our knowledge—in the development of the first high-fidelity model of an entire e-bike drive unit. The proposed model uses the principles of <i>elastic multi body dynamics (eMBD)</i> to elucidate the structural dynamics in dynamic-transient calculations. Comparing power spectra between measured and simulated motion variables validates the chosen model assumptions. The measurements of physical samples utilise accelerometers, contactless <i>laser Doppler vibrometry (LDV)</i> and various test arrangements, which are replicated in simulations and provide accessibility to measure vibrations onto rotating shafts and stationary structures. In summary, this integrated system-level approach can serve as a viable starting point for comprehending and managing the <i>NVH</i> behaviour of e-bikes.

Jochen Hahne

In der vorliegenden Arbeit werden Langzeitmessungen über NH3-Emissionen an zwei zwangsbelüfteten Lege-Elterntierställen und wesentliche Einflussfaktoren auf die Höhe der Emissionen dargestellt. Die Emissionen beider Ställe mit je 432 bis 523 gehaltenen Tieren wurden im Zeitraum von 2017 bis 2021 über ein automatisch arbeitendes Online-Messsystem erfasst. Wie die Untersuchungen zeigen, hat die Entmistung maßgeblichen Einfluss auf die Höhe der NH3-Emissionen. Während bei früheren Untersuchungen an denselben Ställen bei einmaliger Entmistung in der Woche NH3-Emissionen von 148 ± 29 g NH3 a-1 TP-1 auftraten, waren es bei zweimaliger Entmistung im Mittel aller Messungen nur noch 35,2 g NH3 a-1 TP-1. Die Ergebnisse zeigen bei Betrachtung eines entmistungsfreien Zeitraums von bis zu 84 h ferner, dass sich die Emissionen im Mittel alle 24 h verdoppeln. Zur Minderung der NH3-Emissionen aus Lege-Elterntier-Ställen stellt daher die Verkürzung des Entmistungsintervalls eine sehr wirksame Option dar. Darüber hinaus kann bei gegebenem Entmistungsintervall die Teilstromreinigung mit anerkannten Abluftreinigungsverfahren, die nur 60 % der Auslegungsluftrate eines Stalles reinigen, eine NH3-Minderung von mindestens 40 % über das Jahr sicher gewährleisten.

Rostek Michaela

The aim of the article is to present the results of the productivity research of a manufacturing company with particular emphasis on logistics processes. The article presents another example of verification of the developed proprietary productivity method, with particular emphasis on logistic processes. An author's method is used to select indicators, measure productivity and development of processes improvement. The productivity research was carried out in a company in the automotive industry dealing in the production of semi-trailers. A productivity research procedure was developed for the company, it was measured and recommended for improvement of the selected process. The selection of the process to be improved was made on the basis of the forecasted values of the tested productivity indicators, also using econometric modelling. The results of the productivity indicators after the implementation of the improvement were also presented, which confirmed the validity of the applied method and the right choice of process improvement in the company.

King Ankobea-Ansah, Carrie Michele Hall

Estimation of combustion phasing and power production is essential to ensuring proper combustion and load control. However, archetypal control-oriented physics-based combustion models can become computationally expensive if highly accurate predictive capabilities are achieved. Artificial neural network (ANN) models, on the other hand, may provide superior predictive and computational capabilities. However, using classical ANNs for model-based prediction and control can be challenging, since their heuristic and deterministic black-box nature may make them intractable or create instabilities. In this paper, a hybridized modeling framework that leverages the advantages of both physics-based and stochastic neural network modeling approaches is utilized to capture CA50 (the timing when 50% of the fuel energy has been released) along with indicated mean effective pressure (IMEP). The performance of the hybridized framework is compared to a classical ANN and a physics-based-only framework in a stochastic environment. To ensure high robustness and low computational burden in the hybrid framework, the CA50 input parameters along with IMEP are captured with a Bayesian regularized ANN (BRANN) and then integrated into an overall physics-based 0D Wiebe model. The outputs of the hybridized CA50 and IMEP models are then successively fine-tuned with BRANN transfer learning models (TLMs). The study shows that in the presence of a Gaussian-distributed model uncertainty, the proposed hybridized model framework can achieve an RMSE of 1.3 × 10⁻⁵ CAD and 4.37 kPa with a 45.4 and 3.6 s total model runtime for CA50 and IMEP, respectively, for over 200 steady-state engine operating conditions. As such, this model framework may be a useful tool for real-time combustion control where in-cylinder feedback is limited.