I. Computing Methodologies
Refine
Document Type
- Doctoral Thesis (7)
- Master's Thesis (1)
Language
- English (8)
Has Fulltext
- yes (8)
Keywords
- Accelerometer (1)
- Datenfusion (1)
- Deep Learning (1)
- Human Pose (1)
- IMU (1)
- Künstliche Intelligenz (1)
- Machine Learning (1)
- Magnetometer (1)
- Matrix Completion (1)
- Microelectromechanical Systems (1)
Faculty / Organisational entity
Modeling and Simulation of Internet of Things Infrastructures for Cyber-Physical Energy Systems
(2024)
This dissertation presents a novel approach to the model-based development and simulation-based validation of Internet of Things (IoT) infrastructures within the context of Cyber-Physical Energy Systems (CPES). CPES represents an evolution in energy management, seamlessly blending physical and cyber components for efficient, secure, and dependable energy distribution. However, the intricate interplay of these components demands innovative modeling and simulation strategies.
The work begins by establishing a robust foundation, exploring essential background elements such as requirements engineering, model-based systems engineering, digitalization approaches, and the intricacies of IoT platforms. It introduces the novel concept of homomorphic encryption, a critical enabler for securing IoT data within CPES.
In the exploration of the state of the art, the dissertation delves into the multifaceted landscape of IoT simulation, emphasizing the significance of versatility, community support, scalability, and synchronization.
The core contribution emerges in the chapter on simulating IoT networks. It introduces a sophisticated framework that encompasses hardware-in-the-loop, software-in-the-loop, and human-in-the-loop simulation. This innovative framework extends the boundaries of conventional simulation, enabling holistic evaluations of IoT systems.
A practical case study on smart energy usage showcases the application of the framework. Detailed SysML models, including requirements, package diagrams, block definition diagrams, internal block diagrams, state machine diagrams, and activity diagrams, are meticulously examined. The performance evaluation encompasses diverse aspects, from hardware and software validation to human interaction.
In conclusion, this dissertation represents a significant leap forward in the integration of IoT infrastructures within CPES. Its contributions extend from a comprehensive understanding of foundational elements to the practical implementation of a holistic simulation framework. This work not only addresses the current challenges but also outlines a path for future research, shaping the landscape of IoT integration within the dynamic realm of CPES. It offers invaluable insights for researchers, engineers, and stakeholders working towards resilient, secure, and energy-efficient infrastructures.
3D joint angles based human pose is needed for applications like activity recognition, musculoskeletal health, sports biomechanics and ergonomics. The microelectromechanical systems (MEMS) based magnetic-inertial measurement units (MIMUs) can estimate 3D orientation. Due to small size, MIMUs can be attached to the body as wearable sensors for obtaining full 3D human pose and this system is termed as inertial motion capture (i-Mocap). But the MIMUs suffer from sensor errors and disturbances, due to which orientation estimated from individual MIMUs can be erroneous. Accurate sensor calibration is essential and subsequently alignment of these sensors to body segments must also be precisely known, which is called sensor-to-segment calibration. Sensor fusion is employed to address the disturbances and noise in MIMUs. Many state-of-art inertial motion capture approaches ignore the magnetometer and only use IMUs to reduce the error arising from inhomogeneous magnetic field. These algorithms rely on kinematic constraints and assumptions regarding joints and are based on IMUs located on the adjacent body segments. The full body coverage requires 13-17 such units and can be quite obtrusive. The setting up and calibration of so many wearable sensors also take time.
This thesis focuses on 3D human pose estimation from a reduced number of MIMUs and deals with this problem systematically. First we propose an accurate simultaneous calibration of multiple MIMUs, which also learns the uncertainty of individual sensors. We then describe a novel sensor fusion algorithm for robust orientation estimation from an MIMU and for updating sensors calibration online. The residual errors in both sensor calibration and fusion can result in drift error in the joint angles. Therefore, we present anatomical (sensor-to-segment) calibration in which an orientation offset correction term is updated and used for online correction of residual drift in individual joint angles. Subsequently we demonstrate that 3D human joint angle constraints can be learned using a data-driven approach in a high dimensional latent space. Owing to temporal and joint angle constraints, it is possible to use only a reduced set of sensors (as opposed to one sensor per segment) and still obtain 3D human pose. But the spatial and temporal prior learning from data is often limited due to finite set of movement patterns in most datasets. This introduces uncertainty while estimating 3D human pose from sparse MIMU sensors. We propose a magnetometer robust orientation parameterization and a data-driven deep learning framework to predict 3D human pose with associated uncertainty from sparse MIMUs. The model is evaluated on real MIMU data and we show that the uncertainty predicted by the trained model is well-correlated with actual error and ambiguity.
The rising demand for machine learning (ML) models has become a growing concern for stakeholders who depend on automatic decisions. In today's world, black-box solutions (in particular deep neural networks) are being continuously implemented for more and more high-stake scenarios like medical diagnosis or autonomous vehicles. Unfortunately, when these opaque models make predictions that do not align with our expectations, finding a valid justification is simply not possible.
Explainable Artificial Intelligence (XAI) has emerged in response to our need for finding reasons that justify what a machine sees, but we don't. However, contributions in this field are mostly centered around local structures such as individual neurons or single input samples. Global characteristics that govern the behavior of a model are still poorly understood or have not been explored yet. An aggravating factor is the lack of a standard terminology to contextualize and compare contributions in this field. Such lack of consensus is depriving the ML community from ultimately moving away from black-boxes, and start creating systematic methods to design models that are interpretable by design.
So, what are the global patterns that govern the behavior of modern neural networks, and what can we do to make these models more interpretable from the start?
This thesis delves into both issues, unveiling patterns about existing models, and establishing strategies that lead to more interpretable architectures. These include biases coming from imbalanced datasets, quantification of model capacity, and robustness against adversarial attacks. When looking for new models that are interpretable by design, this work proposes a strategy to add more structure to neural networks, based on auxiliary tasks that are semantically related to the main objective. This strategy is the result of applying a novel theoretical framework proposed as part of this work. The XAI framework is meant to contextualize and compare contributions in XAI by providing actionable definitions for terms like "explanation" and "interpretation."
Altogether, these contributions address dire demands for understanding more about the global behavior of modern deep neural networks. More importantly, they can be used as a blueprint for designing novel, and more interpretable architectures. By tackling issues from the present and the future of XAI, results from this work are a firm step towards more interpretable models for computer vision.
The aim of this thesis is to perform a case study to investigate the usability of SysMD in
industrial applications. The focus is on how well it can bridge the gap between requirement
specifications, modeling, and actual development.
SysMD is a new documentation and modeling language which aims to bring documentation
and modeling closer together while still not requiring the user to be an expert in modeling or
requirement specification. This differentiates SysMD from other tools which focus on either
documentation, modeling, or are aimed at modeling experts.
This thesis will show through the case study part that SysMD as a language has a good future
with potential of being used as a language bridging the gap between requirements,
documentation, and modeling without the user needing to be an expert within modeling. It
will also show that SysMD Notebook in its current state is not ready for primetime, and I give
recommendations on how to improve both the SysMD language as well as the SysMD
Notebook to make it usable for industrial projects in the future.
Sequence learning describes the process of understanding the spatio-temporal
relations in a sequence in order to classify it, label its elements or generate
new sequences. Due to the prevalence of structured sequences in nature
and everyday life, it has many practical applications including any language
related processing task. One particular such task that has seen recent success
using sequence learning techniques is the optical recognition of characters
(OCR).
State-of-the-art sequence learning solutions for OCR achieve high performance
through supervised training, which requires large amounts of transcribed
training data. On the other hand, few solutions have been proposed on how
to apply sequence learning in the absence of such data, which is especially
common for hard to transcribe historical documents. Rather than solving
the unsupervised training problem, research has focused on creating efficient
methods for collecting training data through smart annotation tools or generating
synthetic training data. These solutions come with various limitations
and do not solve all of the related problems.
In this work, first the use of erroneous transcriptions for supervised sequence
learning is introduced and it is described how this concept can be applied in
unsupervised training scenarios by collecting or generating such transcriptions.
The proposed OCR pipeline reduces the need of domain specific expertise
to apply OCR, with the goal of making it more accessible. Furthermore, an
approach for evaluating sequence learning OCR models in the absence of
reference transcriptions is presented and its different properties compared
to the standard method are discussed. In a second approach, unsupervised
OCR is treated as an alignment problem between the latent features of the
different language modalities. The outlined solution is to extract language
properties from both the text and image domain through adversarial training
and learn to align them by adding a cycle consistency constraint. The proposed
approach has some strict limitations on the input data, but the results
encourage future research into more widespread applications.
Recommender systems recommend items (e.g., movies, products, books) to users. In this thesis, we proposed two comprehensive and cluster-induced recommendation-based methods: Orthogonal Inductive Matrix Completion (OMIC) and Burst-induced Multi-armed Bandit (BMAB). Given the presence of side information, the first method is categorized as context-aware. OMIC is the first matrix completion method to approach the problem of incorporating biases, side information terms and a pure low-rank term into a single flexible framework with a well-principled optimization procedure. The second method, BMAB, is context-free. That is, it does not require any side data about users or items. Unlike previous context-free multi-armed bandit approaches, our method considers the temporal dynamics of human communication on the web and treats the problem in a continuous time setting. We built our models' assumptions under solid theoretical foundations. For OMIC, we provided theoretical guarantees in the form of generalization bounds by considering the distribution-free case: no assumptions about the sampling distribution are made. Additionally, we conducted a theoretical analysis of community side information when the sampling distribution is known and an adjusted nuclear norm regularization is applied. We showed that our method requires just a few entries to accurately recover the ratings matrix if the structure of the ground truth closely matches the cluster side information. For BMAB, we provided regret guarantees under mild conditions that demonstrate how the system's stability affects the expected reward. Furthermore, we conducted extensive experiments to validate our proposed methodologies. In a controlled environment, we implemented synthetic data generation techniques capable of replicating the domains for which OMIC and BMAB were designed. As a result, we were able to analyze our algorithms' performance across a broad spectrum of ground truth regimes. Finally, we replicated a real-world scenario by utilizing well-established recommender datasets. After comparing our approaches to several baselines, we observe that they achieved state-of-the-art results in terms of accuracy. Apart from being highly accurate, these methods improve interpretability by describing and quantifying features of the datasets they characterize.
Computational simulations run on large supercomputers balance their outputs with the need of the scientist and the capability of the machine. Persistent storage is typically expensive and slow, its peformance grows at a slower rate than the processing power of the machine. This forces scientists to be practical about the size and frequency of the simulation outputs that can be later analyzed to understand the simulation states. Flexibility in the trade-offs of flexibilty and accessibility of the outputs of the simulations are critical the success of scientists using the supercomputers to understand their science. In situ transformations of the simulation state to be persistently stored is the focus of this dissertation.
The extreme size and parallelism of simulations can cause challenges for visualization and data analysis. This is coupled with the need to accept pre partitioned data into the analysis algorithms, which is not always well oriented toward existing software infrastructures. The work in this dissertation is focused on improving current work flows and software to accept data as it is, and efficiently produce smaller, more information rich data, for persistent storage that is easily consumed by end-user scientists. I attack this problem from both a theoretical and practical basis, by managing completely raw data to quantities of information dense visualizations and study methods for managing both the creation and persistence of data products from large scale simulations.
Researchers and analysts in modern industrial and academic environments are faced with a daunting amount of multivariate data. While there has been significant development in the areas of data mining and knowledge
discovery, there is still the need for improved visualizations and generic solutions. The state-of-the-art in visual analytics and exploratory data visualization is to incorporate more profound analysis methods while focusing on improving interactive abilities, in order to support data analysts in gaining new insights through visual exploration and hypothesis building.
In the research field of exploratory data visualization, this thesis contributes new approaches in dimension reduction that tackle a number of shortcomings in state-of-the-art methods, such as interpretability and ambiguity. By combining methods from several disciplines, we describe how ambiguity can be countered effectively by visualizing coordinate values within a lower-dimensional embedding, thereby focusing on the display of the structural composition of high-dimensional data and on an intuitive depiction of inherent global relationships. We also describe how properties and alignment of high-dimensional manifolds can be analyzed in different levels of detail by means of a self-embedding hierarchy of local projections, each using full degree of freedom, while keeping the global context.
To the application field of air quality research, the thesis provides novel means for the research of aerosol source contributions. Triggered by this particularly challenging application problem, we instigate a new research direction in the area of visual analytics by describing a methodology to model-based visual analysis that (i) allows the scientist to be “in the loop” of computations and (ii) enables him to verify and control the analysis process, in order to steer computations towards physical meaning. Careful reflection of our work in this application has led us to derive key design choices that underlie and transcend beyond application-specific solutions. As a result, we describe a general design methodology to computing parameters of a pre-defined analytical model that map to multivariate data. Core applications areas that can benefit from our approach are within engineering disciplines, such as civil, chemical, electrical, and mechanical engineering, as well as in geology, physics, and biology.