Chalmers Open Digital Repository
Not a member yet
26247 research outputs found
Sort by
The Borsuk-Ulam Theorem in Synthetic Stone Duality
No full textThe recently introduced synthetic Stone duality [1] is an extension of homotopy type theory (HoTT), which is conjecturally modeled in the higher topos of light condensed anima as recently introduced by Clausen and Scholze [2]. As it turns out, synthetic Stone duality turns out to be an appropriate setting in which to develop a restricted form of point-set topology. We survey the development of point-set topology within synthetic Stone duality, recovering a working theory of second countable compact Hausdorff spaces and, in particular, a working interval whose topology is the usual metric topology. Using the interval, we are able to define topological paths and loops, as well as the topological fundamental group, in the standard way. We then turn to the theory of higher modalities in HoTT to relate topological spaces, as developed within synthetic Stone duality, to their homotopy types. After recalling the basics of the theory of higher modalities we are able to define the shape modality as the nullification of the interval. The shape modality, which is conjectured to more generally convert (topological) finite CW complexes to their homotopy types, converts the topological circle 1 into its higher inductive counterpart 1. Using shape, we are able to pass back and forth between reasoning about the topological and homotopical circle as convenient. As an application, we use the shape modality to provide a characterization of ℝ, long known as folklore, in terms of the modal unit : 1 → 1 for the topological circle. Similarly using the shape modality, we are able to show that topological paths in 1 lift through the standard covering map ℝ ↠ 1 while bypassing the standard covering space theory, and prove the two-dimensional Borsuk-Ulam theorem as an application
Förbättring av Overall Equipment Effectiveness i Bromsskivetillverkning Strategier för effektivare produktionsflöden
No full textThis thesis was conducted at AC Floby, a company that partly produces brake discs among other components. The study aimed to improve production efficiency on one of their production lines by analyzing the components of the OEE metric: Availability, Performance and Quality. Areas of improvement were identified based on a comprehensive assessment of the potential impact of proposed actions in relation to their feasibility within the company. These trade-offs guided the selection of which aspects of the current state and result that were further discussed. By combining internal data, observations, and interviews with operators, maintenance staff and production engineers, the current state was mapped to help identify areas of improvement. The theoretical framework includes concepts such as Lean Production, Value Stream Mapping (VSM), Total Productive Maintenance (TPM), Value-Driven Maintenance (VDM), as well as models for root cause analysis, data-driven decision-making, and operator
engagement. The results indicate that availability is the OEE component with the greatest potential for improvement. Key challenges identified through root cause analysis include poor data quality,
lack of standardized downtime registration, long and inconsistent changeover times and recurring technical failures. Additional areas of improvement include the flow of information between departments, the lack of a standardized practice for downtime registration, and insufficient follow-up on preventive maintenance activities.
Based on these findings, the main improvement proposals are: measures to enhance data quality, such as more measurement points, standardized downtime registration, and updated cycle times. Furthermore, improvements to buffer dimensioning, development of operator driven maintenance, and application of the SMED methodology are recommended to reduce availability losses on the production line. To ensure the long-term value and feasibility of the proposed improvements, they were also evaluated from the perspective of the four dimensions of sustainability: economic, social, ecological and ethical
Prototypframtagning av moduler för utomhusgym. Utomhusträning som förenar estetik och funktionalitet
No full textThe purpose of the project was to explore the possibility to integrate training modules in Invencon’s pergola. To therefore create a balance between aesthetics and functionality while enabling full-body training. The goal was to create concepts for training modules through sketches, physical prototypes and cad prototypes. A user study that aimed towards active adults, 30 years or older that owned a back yard, showed a need of accessible, aesthetically pleasing exercise equipment. A concurrent analysis showed that outdoor gyms made of metal often were affected of rust, wooden ones required
more space and indoor gyms typically shared similar design. Ergonomic analyses identified some risks of injuries, often connected to incorrect technic and difficulty detecting the error.
Three concepts were chosen for further development and prototyping, because they together could solve full body training:
1. Multi barbell: A simplified smith-machine where a barbell runs between wooden beams and is fixed using pins.
2. Kettlepot: A kettlebell designed to function as a flowerpot when it is not used for exercising, with integrated weigh bowls that enables weight adjustment.
3. Cvt-cable machine: A simplified cable machine placed in a cabinet with weight adjustment through a cvt-gear.
The design was adjusted based on risk analysis. A sematic analysis showed that the concepts were self-instructive. The conclusion is that the three concepts is considered to fulfill the goal by enabling full body training, that balances aesthetics and functionality. This report is written in Swedish
Identification of Wi-Fi-enabled drones via software-defined radio (SDR) and artificial intelligence (AI)
No full textThis thesis presents a method for detecting Wi-Fi-controlled drones using software-defined radio (SDR) technology combined with artificial intelligence (AI). Radio frequency (RF) signals in the 2.4 GHz band were captured and analyzed to distinguish drone transmissions from conventional wireless activity. A rule-based bandwidth analysis was first implemented, followed by a convolutional neural network (CNN) classifier trained on power spectral density (PSD) features. The system successfully identified drone signals in real time under test conditions, demonstrating that SDR combined with AI provides a cost-effective and extensible framework for RF-based drone detection
A Method for Learning Automated Driving Systems and Data Consistency Analysis
No full textIn this thesis a method for developing a deep learning network to predict signals in
automated driving systems is presented. The goal is to incorporate the network in a
tool aimed to find discrepancies in vehicle testing data. To this end, a neural network
was constructed and trained on simulations of vehicle dynamics with the purpose of
predicting the dynamics of a real test vehicle. A parallel goal was to compare the
properties of the network to a software for simulating vehicle dynamics. The thesis
was made in collaboration with Zenseact, an AI and software company developing
the complete software stack for automated driving and advanced driver-assistance
systems (ADS and ADAS). The network did predictions replicating a simulation
software’s Controllers, Vehicle model and the two combined. To evaluate which
type of training data gave the best result, three datasets with different driving
schemes were generated by the simulation software. On a test set constructed from
21 log files from real test drives, the network got the best results predicting the
Controllers and Vehicle model combined. In this case, for only predicting a vehicle’s
lateral motion, a training set consisting of constant longitudinal velocity generated
the best results. If both lateral and longitudinal quantities are desired, a dataset
with sinusoidally changing acceleration performed best overall. The other purpose
was to compare the time aspects of the network to the simulation software. This
was done by evaluating the time for data generation, training and prediction on the
”Constant” dataset used in this thesis. This dataset is comprised by 84,528 time
steps of ADS signals. Data generation took 8.5 to 9 minutes over all tests. The
training time was consistently 61 to 62 seconds in total and the prediction time on
the 84,495 batches was between 6 to 7 seconds, or an average of 75.9 μs per batch. It
was concluded that, apart for the intended goal of being a core in a data consistency
tool, the main use case for the network could be in applications where fast data
generation times are crucial, for example in near real time predictions during test
drives
Cross-Language Dependency Analysis for VS Code Extension Ecosystem
No full textVisual Studio Code (VS Code) is currently the most popular integrated development environment (IDE), primarily due to its highly modular architecture facilitated by third-party extensions. These extensions can rely on dependencies spanning multiple programming languages, notably JavaScript and native languages such as C and C++. Such cross-language interactions introduce complexity and potential security vulnerabilities due to differences in memory management, type safety, and crash resilience between languages. While previous research has identified the inherent security risks in cross-language bindings within individual packages in the npm ecosystem, the implications of such vulnerabilities within the VS Code extension ecosystem have yet to be explored.
This thesis investigates cross-language dependencies in VS Code extensions, specifically focusing on the interactions between JavaScript and native code. A methodology is presented to systematically discover, construct, and analyse the dependency tree from an extension to native code. The study uncovers patterns, characteristics, and potential security risks associated with native dependencies in VS Code extensions. This research provides insights into the lack of security practices within the VS Code ecosystem by addressing the gap between current knowledge about cross-language vulnerabilities and VS Code extensions. The results show that 455 (14.7%) out of the investigated 3,078 extensions either implemented native code directly or depend on a package including cross-language cooperation. While only two extensions had direct production code in a native language, they amassed 171 potential vulnerabilities. Additionally, 211 extensions depended on 228 dependencies containing native code that amassed 8,732 potential vulnerabilities in total, showing the potential risks of using such packages
Assessment of Hydrogen Infused In-situ Methanation Systems: Modeling and Analyzing from a Techno-Economic and Climate Perspective
No full textThe transition away from fossil fuels requires the development of alternative technologies, as no single solution can meet all the requirements for full decarbonization. Hydrogen is a flexible energy carrier, with one potential application being methanation. A methanation process uses hydrogen to convert carbon dioxide from biogas production to bio-e-methane. This thesis offers a comparative assessment
of hydrogen-infused in-situ methanation systems operating under different conditions, from a techno-economic and climate perspective, applied to a biogas plant in
Linköping operated by Tekniska Verken.
The evaluation was carried out through simulations using Odyssey software, considering three cases: (1) hydrogen used for the methanation process solely, (2) hydrogen used for methanation combined with direct sales, and (3) hydrogen used solely for direct sales. The outcomes were influenced by electrolyzer characteristics, system size, and fluctuating electricity prices. Results indicated that the cost-optimal configuration was highly dependent on the operational strategy and electrolyzer capacity.
Furthermore, Case 2 showed strong sensitivity to hydrogen pricing, while Case 3
showed that both methanation and direct hydrogen sales could be more profitable,
depending on the specific setup. The key findings of the climate analysis showed
that the emissions from the systems were small compared to if the system were
placed in another country, due to the difference in grid emission factors.
Uncertainties regarding future electricity prices and market values of hydrogen and
bio-e-methane pose challenges for assessing long-term feasibility. Nevertheless, this
thesis highlights the importance of resource efficiency by focusing on hydrogeninfused in-situ methanation systems, which increase biogas production without additional substrate, thereby helping to reduce fossil fuel us
Robust Model-Based Clustering Techniques for Non-Uniform LiDAR Point Clouds via Range Image Transformations
No full textAbstract
Autonomous driving is a rapidly advancing field that is reshaping the future of transportation by promising improved traffic flow, reduced energy consumption, safer roads, and the elimination of human intervention, thereby saving time. An essential part of autonomous driving is a strong perception system such cameras and radar, which are key for safety features like lane keeping, adaptive cruise control, and emergency braking. However, LiDAR technology stands out being a central part of the perception system due to its ability to provide high-quality 3D spatial information even during poor lighting conditions. A specific type of modern LiDAR sensor (e.g., Luminar Iris) introduces non-uniformity in point cloud distribution
by offering adaptive resolution. This feature provides more details for the area of interest, while reducing unnecessary data in less relevant areas such as the sky. However, this non-uniformity of the LiDAR data introduces challenges for classical clustering algorithms, which work under the assumption of a uniformly distributed LiDAR point cloud (i.e., fixed sensor angular resolution). Consequently, these algorithms will struggle with over and under-segmentation problems when the distance between objects varies. In this thesis, two robust model-based clustering algorithms are developed and designed to cluster non-uniform LiDAR point clouds. The methods build on classical breakpoint detection and range image-based clustering approaches to handle varying point densities while still being computationally efficient. Furthermore, two ground removal approaches are developed and evaluated along with a suggestion of two evaluation methods, visual and quantitative evaluation on a dataset of Luminar point clouds. The results demonstrate better clustering performance of the
two developed algorithms compared to a baseline algorithm, thus resulting in fewer over-segmentation and under-segmentation errors. Moreover, a thorough analysis of computational is provided. In summary, this research resulted in a stable and reliable clustering performance across diverse scenarios, which helps to realize safer and more predictable autonomous driving without relying on large-scale labeled data for training, as in deep learning models