05 Fakultät Informatik, Elektrotechnik und Informationstechnik

Permanent URI for this collectionhttps://elib.uni-stuttgart.de/handle/11682/6

Browse

Now showing 1 - 20 of 263

Open Access
A recommender approach to enable effective and efficient self-service analytics in data lakes
(2023) Stach, Christoph; Eichler, Rebecca; Schmidt, Simone
As a result of the paradigm shift away from rather rigid data warehouses to general-purpose data lakes, fully flexible self-service analytics is made possible. However, this also increases the complexity for domain experts who perform these analyses, since comprehensive data preparation tasks have to be implemented for each data access. For this reason, we developed BARENTS, a toolset that enables domain experts to specify data preparation tasks as ontology rules, which are then applied to the data involved. Although our evaluation of BARENTS showed that it is a valuable contribution to self-service analytics, a major drawback is that domain experts do not receive any semantic support when specifying the rules. In this paper, we therefore address how a recommender approach can provide additional support to domain experts by identifying supplementary datasets that might be relevant for their analyses or additional data processing steps to improve data refinement. This recommender operates on the set of data preparation rules specified in BARENT-i.e., the accumulated knowledge of all domain experts is factored into the data preparation for each new analysis. Evaluation results indicate that such a recommender approach further contributes to the practicality of BARENTS and thus represents a step towards effective and efficient self-service analytics in data lakes.
Open Access
Accelerated 3D FEA of an axial flux machine by exclusively using the magnetic scalar potential
(2023) Schäfer, Adrian; Pecha, Urs; Kaiser, Benedikt; Schmid, Martin; Parspour, Nejila
This article focuses on increasing the computational efficiency of 3D multi-static magnetic finite element analysis (FEA) for electrical machines (EMs), which have a magnetic field evolving in 3D space. Although 3D FEA is crucial for analyzing these machines and their operational behavior, it is computationally expensive. A novel approach is proposed in order to solve the magnetic field equations by exclusively using the magnetic scalar potential. For this purpose, virtual variable permanent magnets (vPMs) are introduced to model the impact of the machine’s coils. The effect on which this approach is based is derived from and explained by Maxwell’s equations. To validate the new approach, an axial flux machine (AFM) is simulated using both 2D and 3D FEA with the magnetic vector potential and current-carrying coils as a reference. The results demonstrate a high level of agreement between the new approach and the reference simulations as well as an acceleration of the computation by a factor of 15 or even more. Additionally, the research provides valuable insights into meshing techniques and torque calculation for EMs in FEA.
Open Access
Adaptive triple-fed antenna and thinned RF-chip integration into ultra thin flexible polymer foil
(2023) Fischer-Kennedy, Serafin B.; Özbek, Sefa; Wang, Shuo; Grözing, Markus; Hesselbarth, Jan; Berroth, Manfred; Burghartz, Joachim
Open Access
Additively manufactured transverse flux machine components with integrated slits for loss reduction
(2022) Kresse, Thomas; Schurr, Julian; Lanz, Maximilian; Kunert, Torsten; Schmid, Martin; Parspour, Nejila; Schneider, Gerhard; Goll, Dagmar
Laser powder bed fusion (L-PBF) was used to produce stator half-shells of a transverse flux machine from pure iron (99.9% Fe). In order to reduce iron losses in the bulk components, radially extending slits with a nominal width of 150 and 300 µm, respectively, were integrated during manufacturing. The components were subjected to a suitable heat treatment. In addition to a microscopic examination of the slit quality, the iron losses were also measured using both a commercial and a self-developed measurement setup. The investigations showed the iron losses can be reduced by up to 49% due to the integrated slits and the heat treatment.
Open Access
Adopting microservices and DevOps in the cyber‐physical systems domain : a rapid review and case study
(2022) Fritzsch, Jonas; Bogner, Justus; Haug, Markus; Franco da Silva, Ana Cristina; Rubner, Carolin; Saft, Matthias; Sauer, Horst; Wagner, Stefan
The domain of cyber‐physical systems (CPS) has recently seen strong growth, for example, due to the rise of the Internet of Things (IoT) in industrial domains, commonly referred to as “Industry 4.0.” However, CPS challenges like the strong hardware focus can impact modern software development practices, especially in the context of modernizing legacy systems. While microservices and DevOps have been widely studied for enterprise applications, there is insufficient coverage for the CPS domain. Our goal is therefore to analyze the peculiarities of such systems regarding challenges and practices for using and migrating towards microservices and DevOps. We conducted a rapid review based on 146 scientific papers, and subsequently validated our findings in an interview‐based case study with nine CPS professionals in different business units at Siemens AG. The combined results picture the specifics of microservices and DevOps in the CPS domain. While several differences were revealed that may require adapted methods, many challenges and practices are shared with typical enterprise applications. Our study supports CPS researchers and practitioners with a summary of challenges, practices to address them, and research opportunities.
Open Access
Advances in clinical voice quality analysis with VOXplot
(2023) Barsties von Latoszek, Ben; Mayer, Jörg; Watts, Christopher R.; Lehnert, Bernhard
Background: The assessment of voice quality can be evaluated perceptually with standard clinical practice, also including acoustic evaluation of digital voice recordings to validate and further interpret perceptual judgments. The goal of the present study was to determine the strongest acoustic voice quality parameters for perceived hoarseness and breathiness when analyzing the sustained vowel [a:] using a new clinical acoustic tool, the VOXplot software. Methods: A total of 218 voice samples of individuals with and without voice disorders were applied to perceptual and acoustic analyses. Overall, 13 single acoustic parameters were included to determine validity aspects in relation to perceptions of hoarseness and breathiness. Results: Four single acoustic measures could be clearly associated with perceptions of hoarseness or breathiness. For hoarseness, the harmonics-to-noise ratio (HNR) and pitch perturbation quotient with a smoothing factor of five periods (PPQ5), and, for breathiness, the smoothed cepstral peak prominence (CPPS) and the glottal-to-noise excitation ratio (GNE) were shown to be highly valid, with a significant difference being demonstrated for each of the other perceptual voice quality aspects. Conclusions: Two acoustic measures, the HNR and the PPQ5, were both strongly associated with perceptions of hoarseness and were able to discriminate hoarseness from breathiness with good confidence. Two other acoustic measures, the CPPS and the GNE, were both strongly associated with perceptions of breathiness and were able to discriminate breathiness from hoarseness with good confidence.
Open Access
All-in-memory brain-inspired computing using FeFET synapses
(2022) Thomann, Simon; Nguyen, Hong L. G.; Genssler, Paul R.; Amrouch, Hussam
The separation of computing units and memory in the computer architecture mandates energy-intensive data transfers creating the von Neumann bottleneck. This bottleneck is exposed at the application level by the steady growth of IoT and data-centric deep learning algorithms demanding extraordinary throughput. On the hardware level, analog Processing-in-Memory (PiM) schemes are used to build platforms that eliminate the compute-memory gap to overcome the von Neumann bottleneck. PiM can be efficiently implemented with ferroelectric transistors (FeFET), an emerging non-volatile memory technology. However, PiM and FeFET are heavily impacted by process variation, especially in sub 14 nm technology nodes, reducing the reliability and thus inducing errors. Brain-inspired Hyperdimensional Computing (HDC) is robust against such errors. Further, it is able to learn from very little data cutting energy-intensive transfers. Hence, HDC, in combination with PiM, tackles the von Neumann bottleneck at both levels. Nevertheless, the analog nature of PiM schemes necessitates the conversion of results to digital, which is often not considered. Yet, the conversion introduces large overheads and diminishes the PiM efficiency. In this paper, we propose an all-in-memory scheme performing computation and conversion at once, utilizing programmable FeFET synapses to build the comparator used for the conversion. Our experimental setup is first calibrated against Intel 14 nm FinFET technology for both transistor electrical characteristics and variability. Then, a physics-based model of ferroelectric is included to realize the Fe-FinFETs. Using this setup, we analyze the circuit’s susceptibility to process variation, derive a comprehensive error probability model, and inject it into the inference algorithm of HDC. The robustness of HDC against noise and errors is able to withstand the high error probabilities with a loss of merely 0.3% inference accuracy.
Open Access
All-inorganic CsPbI2Br perovskite solar cells with thermal stability at 250 °C and moisture-resilience via polymeric protection layers
(2025) Roy, Rajarshi; Byranvand, Mahdi Malekshahi; Zohdi, Mohamed Reza; Magorian Friedlmeier, Theresa; Das, Chittaranjan; Hempel, Wolfram; Zuo, Weiwei; Kedia, Mayank; Rendon, Jose Jeronimo; Boehringer, Stephan; Hailegnanw, Bekele; Vorochta, Michael; Mehl, Sascha; Rai, Monika; Kulkarni, Ashish; Mathur, Sanjay; Saliba, Michael
All-inorganic perovskites, such as CsPbI2Br, have emerged as promising compositions due to their enhanced thermal stability. However, they face significant challenges due to their susceptibility to humidity. In this work, CsPbI2Br perovskite is treated with poly(3-hexylthiophen-2,5-diyl) (P3HT) during the crystallization resulting in significant stability improvements against thermal, moisture and steady-state operation stressors. The perovskite solar cell retains ∼90% of the initial efficiency under relative humidity (RH) at ∼60% for 30 min, which is among the most stable all-inorganic perovskite devices to date under such harsh conditions. Furthermore, the P3HT treatment ensures high thermal stress tolerance at 250 °C for over 5 h. In addition to the stability enhancements, the champion P3HT-treated device shows a higher power conversion efficiency (PCE) of 13.5% compared to 12.7% (reference) with the stabilized power output (SPO) for 300 s. In addition, the P3HT-protected perovskite layer in ambient conditions shows ∼75% of the initial efficiency compared to the unprotected devices with ∼28% of their initial efficiency after 7 days of shelf life.
Open Access
All-perovskite tandem solar cells : from fundamentals to technological progress
(2024) Lim, Jaekeun; Park, Nam-Gyu; Seok, Sang Il; Saliba, Michael
Organic-inorganic perovskite materials have gradually progressed from single-junction solar cells to tandem (double) or even multi-junction (triple-junction) solar cells as all-perovskite tandem solar cells (APTSCs). Perovskites have numerous advantages: (1) tunable optical bandgaps, (2) low-cost, e.g. via solution-processing, inexpensive precursors, and compatibility with many thin-film processing technologies, (3) scalability and lightweight, and (4) eco-friendliness related to low CO2 emission. However, APTSCs face challenges regarding stability caused by Sn2+ oxidation in narrow bandgap perovskites, low performance due to Voc deficit in the wide bandgap range, non-standardisation of charge recombination layers, and challenging thin-film deposition as each layer must be nearly perfectly homogenous. Here, we discuss the fundamentals of APTSCs and technological progress in constructing each layer of the all-perovskite stacks. Furthermore, the theoretical power conversion efficiency (PCE) limitation of APTSCs is discussed using simulations.
Open Access
Alloy stability of Ge1-xSnx with Sn concentrations up to 17% utilizing low-temperature molecular beam epitaxy
(2020) Schwarz, Daniel; Funk, Hannes S.; Oehme, Michael; Schulze, Jörg
The binary alloy germanium tin has already been presented as a direct group IV semiconductor at high tin concentrations and specific strain. Therefore, it offers a promising approach for the monolithic integrated light source towards the optical on-chip communication on silicon. However, the main challenge faced by many researchers is the achievement of high tin concentrations and good crystal quality. The key issues are the lattice mismatch to silicon and germanium, as well as the limited solid solubility of tin in germanium of less than 1%. Therefore, this paper presents a systematic investigation of the epitaxial growth conditions of germanium tin with tin concentrations up to 17%. For this, we performed two growth experiments utilizing molecular beam epitaxy. In one experiment, we varied the growth temperature for the epitaxy of germanium tin with 8% tin to investigate the inter-growth temperature stability. In the second experiment, we focused on the strain-relaxation of germanium tin, depending on different tin concentrations and doping types. The results of subsequent material analysis with x-ray diffraction and scanning electron microscopy allow us to narrow the epitaxial window of germanium tin. Furthermore, we present a possible explanation for the unique relaxation mechanism of germanium tin, which is significantly different from the well-known relaxation mechanism of silicon germanium.
Open Access
The aluminum standard : using generative Artificial Intelligence tools to synthesize and annotate non-structured patient data
(2024) Diaz Ochoa, Juan G.; Mustafa, Faizan E.; Weil, Felix; Wang, Yi; Kama, Kudret; Knott, Markus
Background. Medical narratives are fundamental to the correct identification of a patient’s health condition. This is not only because it describes the patient’s situation. It also contains relevant information about the patient’s context and health state evolution. Narratives are usually vague and cannot be categorized easily. On the other hand, once the patient’s situation is correctly identified based on a narrative, it is then possible to map the patient’s situation into precise classification schemas and ontologies that are machine-readable. To this end, language models can be trained to read and extract elements from these narratives. However, the main problem is the lack of data for model identification and model training in languages other than English. First, gold standard annotations are usually not available due to the high level of data protection for patient data. Second, gold standard annotations (if available) are difficult to access. Alternative available data, like MIMIC (Sci Data 3:1, 2016) is written in English and for specific patient conditions like intensive care. Thus, when model training is required for other types of patients, like oncology (and not intensive care), this could lead to bias. To facilitate clinical narrative model training, a method for creating high-quality synthetic narratives is needed. Method. We devised workflows based on generative AI methods to synthesize narratives in the German language to avoid the disclosure of patient’s health data. Since we required highly realistic narratives, we generated prompts, written with high-quality medical terminology, asking for clinical narratives containing both a main and co-disease. The frequency of distribution of both the main and co-disease was extracted from the hospital’s structured data, such that the synthetic narratives reflect the disease distribution among the patient’s cohort. In order to validate the quality of the synthetic narratives, we annotated them to train a Named Entity Recognition (NER) algorithm. According to our assumptions, the validation of this system implies that the synthesized data used for its training are of acceptable quality. Result. We report precision, recall and F1 score for the NER model while also considering metrics that take into account both exact and partial entity matches. Trained models are cautious, with a precision up to 0.8 for Entity Type match metric and a F1 score of 0.3. Conclusion. Despite its inherent limitations, this technology has the potential to allow data interoperability by using encoded diseases across languages and regions without compromising data safety. Additionally, it facilitates the synthesis of unstructured patient data. In this way, the identification and training of models can be accelerated. We believe that this method may be able to generate discharge letters for any combination of main and co-diseases, which will significantly reduce the amount of time spent writing these letters by healthcare professionals.
Open Access
AmericasNLI : machine translation and natural language inference systems for Indigenous languages of the Americas
(2022) Kann, Katharina; Ebrahimi, Abteen; Mager, Manuel; Oncevay, Arturo; Ortega, John E.; Rios, Annette; Fan, Angela; Gutierrez-Vasques, Ximena; Chiruzzo, Luis; Giménez-Lugo, Gustavo A.; Ramos, Ricardo; Meza Ruiz, Ivan Vladimir; Mager, Elisabeth; Chaudhary, Vishrav; Neubig, Graham; Palmer, Alexis; Coto-Solano, Rolando; Vu, Ngoc Thang
Little attention has been paid to the development of human language technology for truly low-resource languages - i.e., languages with limited amounts of digitally available text data, such as Indigenous languages. However, it has been shown that pretrained multilingual models are able to perform crosslingual transfer in a zero-shot setting even for low-resource languages which are unseen during pretraining. Yet, prior work evaluating performance on unseen languages has largely been limited to shallow token-level tasks. It remains unclear if zero-shot learning of deeper semantic tasks is possible for unseen languages. To explore this question, we present AmericasNLI, a natural language inference dataset covering 10 Indigenous languages of the Americas. We conduct experiments with pretrained models, exploring zero-shot learning in combination with model adaptation. Furthermore, as AmericasNLI is a multiway parallel dataset, we use it to benchmark the performance of different machine translation models for those languages. Finally, using a standard transformer model, we explore translation-based approaches for natural language inference. We find that the zero-shot performance of pretrained models without adaptation is poor for all languages in AmericasNLI, but model adaptation via continued pretraining results in improvements. All machine translation models are rather weak, but, surprisingly, translation-based approaches to natural language inference outperform all other models on that task.
Open Access
Analysis of political debates through newspaper reports : methods and outcomes
(2020) Lapesa, Gabriella; Blessing, Andre; Blokker, Nico; Dayanik, Erenay; Haunss, Sebastian; Kuhn, Jonas; Padó, Sebastian
Discourse network analysis is an aspiring development in political science which analyzes political debates in terms of bipartite actor/claim networks. It aims at understanding the structure and temporal dynamics of major political debates as instances of politicized democratic decision making. We discuss how such networks can be constructed on the basis of large collections of unstructured text, namely newspaper reports. We sketch a hybrid methodology of manual analysis by domain experts complemented by machine learning and exemplify it on the case study of the German public debate on immigration in the year 2015. The first half of our article sketches the conceptual building blocks of discourse network analysis and demonstrates its application. The second half discusses the potential of the application of NLP methods to support the creation of discourse network datasets.
Open Access
Analytic free-energy expression for the 2D-Ising model and perspectives for battery modeling
(2023) Markthaler, Daniel; Birke, Kai Peter
Although originally developed to describe the magnetic behavior of matter, the Ising model represents one of the most widely used physical models, with applications in almost all scientific areas. Even after 100 years, the model still poses challenges and is the subject of active research. In this work, we address the question of whether it is possible to describe the free energy A of a finite-size 2D-Ising model of arbitrary size, based on a couple of analytically solvable 1D-Ising chains. The presented novel approach is based on rigorous statistical-thermodynamic principles and involves modeling the free energy contribution of an added inter-chain bond DAbond(b, N) as function of inverse temperature b and lattice size N. The identified simple analytic expression for DAbond is fitted to exact results of a series of finite-size quadratic N N-systems and enables straightforward and instantaneous calculation of thermodynamic quantities of interest, such as free energy and heat capacity for systems of an arbitrary size. This approach is not only interesting from a fundamental perspective with respect to the possible transfer to a 3D-Ising model, but also from an application-driven viewpoint in the context of (Li-ion) batteries where it could be applied to describe intercalation mechanisms.
Open Access
Analyzing the influence of hyper-parameters and regularizers of topic modeling in terms of Renyi entropy
(2020) Koltcov, Sergei; Ignatenko, Vera; Boukhers, Zeyd; Staab, Steffen
Topic modeling is a popular technique for clustering large collections of text documents. A variety of different types of regularization is implemented in topic modeling. In this paper, we propose a novel approach for analyzing the influence of different regularization types on results of topic modeling. Based on Renyi entropy, this approach is inspired by the concepts from statistical physics, where an inferred topical structure of a collection can be considered an information statistical system residing in a non-equilibrium state. By testing our approach on four models-Probabilistic Latent Semantic Analysis (pLSA), Additive Regularization of Topic Models (BigARTM), Latent Dirichlet Allocation (LDA) with Gibbs sampling, LDA with variational inference (VLDA)-we, first of all, show that the minimum of Renyi entropy coincides with the “true” number of topics, as determined in two labelled collections. Simultaneously, we find that Hierarchical Dirichlet Process (HDP) model as a well-known approach for topic number optimization fails to detect such optimum. Next, we demonstrate that large values of the regularization coefficient in BigARTM significantly shift the minimum of entropy from the topic number optimum, which effect is not observed for hyper-parameters in LDA with Gibbs sampling. We conclude that regularization may introduce unpredictable distortions into topic models that need further research.
Open Access
Application of pathfinding algorithms in partial discharge localization in power transformers
(2024) Beura, Chandra Prakash; Wolters, Jorim; Tenbohlen, Stefan
The introduction of artificial intelligence (AI) to ultra-high-frequency (UHF) partial discharge (PD) monitoring systems in power transformers for the localization of PD sources can help create a robust and reliable system with high usability and precision. However, training the AI with experimental data or data from electromagnetic simulation is costly and time-consuming. Furthermore, electromagnetic simulations often calculate more data than needed, whereas, for localization, the signal time-of-flight information is the most important. A tailored pathfinding algorithm can bypass the time-consuming and computationally expensive process of simulating or collecting data from experiments and be used to create the necessary training data for an AI-based monitoring system of partial discharges in power transformers. In this contribution, Dijkstra’s algorithm is used with additional line-of-sight propagation algorithms to determine the paths of the electromagnetic waves generated by PD sources in a three-dimensional (3D) computer-aided design (CAD) model of a 300 MVA power transformer. The time-of-flight information is compared with results from experiments and electromagnetic simulations, and it is found that the algorithm maintains accuracy similar to that of the electromagnetic simulation software, with some under/overestimations in specific scenarios, while being much faster at calculations.
Open Access
Artificial feature extraction for estimating state-of-temperature in lithium-ion-cells using various long short-term memory architectures
(2022) Kopp, Mike; Ströbel, Marco; Fill, Alexander; Pross-Brakhage, Julia; Birke, Kai Peter
The temperature in each cell of a battery system should be monitored to correctly track aging behavior and ensure safety requirements. To eliminate the need for additional hardware components, a software based prediction model is needed to track the temperature behavior. This study looks at machine learning algorithms that learn physical behavior of non-linear systems based on sample data. Here, it is shown how to improve the prediction accuracy using a new method called “artificial feature extraction” compared to classical time series approaches. We show its effectiveness on tracking the temperature behavior of a Li-ion cell with limited training data at one defined ambient temperature. A custom measuring system was created capable of tracking the cell temperature, by installing a temperature sensor into the cell wrap instead of attaching it to the cell housing. Additionally, a custom early stopping algorithm was developed to eliminate the need for further hyperparameters. This study manifests that artificially training sub models that extract features with high accuracy aids models in predicting more complex physical behavior. On average, the prediction accuracy has been improved by ΔTcell=0.01 °C for the training data and by ΔTcell=0.007 °C for the validation data compared to the base model. In the field of electrical energy storage systems, this could reduce costs, increase safety and improve knowledge about the aging progress in an individual cell to sort out for second life applications.
Open Access
Assessment of UHF frequency range for failure classification in power transformers
(2024) Schiewaldt, Karl; de Castro, Bruno Albuquerque; Ardila-Rey, Jorge Alfredo; Franchin, Marcelo Nicoletti; Andreoli, André Luiz; Tenbohlen, Stefan
Ultrahigh-frequency (UHF) sensing is one of the most promising techniques for assessing the quality of power transformer insulation systems due to its capability to identify failures like partial discharges (PDs) by detecting the emitted UHF signals. However, there are still uncertainties regarding the frequency range that should be evaluated in measurements. For example, most publications have stated that UHF emissions range up to 3 GHz. However, a Cigré brochure revealed that the optimal spectrum is between 100 MHz and 1 GHz, and more recently, a study indicated that the optimal frequency range is between 400 MHz and 900 MHz. Since different faults require different maintenance actions, both science and industry have been developing systems that allow for failure-type identification. Hence, it is important to note that bandwidth reduction may impair classification systems, especially those that are frequency-based. This article combines three operational conditions of a power transformer (healthy state, electric arc failure, and partial discharges on bushing) with three different self-organized maps to carry out failure classification: the chromatic technique (CT), principal component analysis (PCA), and the shape analysis clustering technique (SACT). For each case, the frequency content of UHF signals was selected at three frequency bands: the full spectrum, Cigré brochure range, and between 400 MHz and 900 MHz. Therefore, the contributions of this work are to assess how spectrum band limitation may alter failure classification and to evaluate the effectiveness of signal processing methodologies based on the frequency content of UHF signals. Additionally, an advantage of this work is that it does not rely on training as is the case for some machine learning-based methods. The results indicate that the reduced frequency range was not a limiting factor for classifying the state of the operation condition of the power transformer. Therefore, there is the possibility of using lower frequency ranges, such as from 400 MHz to 900 MHz, contributing to the development of less costly data acquisition systems. Additionally, PCA was found to be the most promising technique despite the reduction in frequency band information.
Open Access
AssistML : an approach to manage, recommend and reuse ML solutions
(2023) Villanueva Zacarias, Alejandro Gabriel; Reimann, Peter; Weber, Christian; Mitschang, Bernhard
The adoption of machine learning (ML) in organizations is characterized by the use of multiple ML software components. When building ML systems out of these software components, citizen data scientists face practical requirements which go beyond the known challenges of ML, e. g., data engineering or parameter optimization. They are expected to quickly identify ML system options that strike a suitable trade-off across multiple performance criteria. These options also need to be understandable for non-technical users. Addressing these practical requirements represents a problem for citizen data scientists with limited ML experience. This calls for a concept to help them identify suitable ML software combinations. Related work, e. g., AutoML systems, are not responsive enough or cannot balance different performance criteria. This paper explains how AssistML, a novel concept to recommend ML solutions, i. e., software systems with ML models, can be used as an alternative for predictive use cases. Our concept collects and preprocesses metadata of existing ML solutions to quickly identify the ML solutions that can be reused in a new use case. We implement AssistML and evaluate it with two exemplary use cases. Results show that AssistML can recommend ML solutions in line with users’ performance preferences in seconds. Compared to AutoML, AssistML offers citizen data scientists simpler, intuitively explained ML solutions in considerably less time. Moreover, these solutions perform similarly or even better than AutoML models.
Open Access
Audio guide for visually impaired people based on combination of stereo vision and musical tones
(2019) Simões, Walter C. S. S.; Silva, Yuri M. L. R.; Pio, José Luiz de S.; Jazdi, Nasser; F. de Lucena, Vicente
Indoor navigation systems offer many application possibilities for people who need information about the scenery and the possible fixed and mobile obstacles placed along the paths. In these systems, the main factors considered for their construction and evaluation are the level of accuracy and the delivery time of the information. However, it is necessary to notice obstacles placed above the user’s waistline to avoid accidents and collisions. In this paper, different methodologies are associated to define a hybrid navigation model called iterative pedestrian dead reckoning (i-PDR). i-PDR combines the PDR algorithm with a Kalman linear filter to correct the location, reducing the system’s margin of error iteratively. Obstacle perception was addressed through the use of stereo vision combined with a musical sounding scheme and spoken instructions that covered an angle of 120 degrees in front of the user. The results obtained in the margin of error and the maximum processing time are 0.70 m and 0.09 s, respectively, with obstacles at ground level and suspended with an accuracy equivalent to 90%.