Chalmers Open Digital Repository
Not a member yet
26247 research outputs found
Sort by
Automated Satellite Tracking for Continuous Internet Access - Design and Implementation of a Dynamic Antenna Control System for Vehicular Internet Connectivity
No full textThis thesis presents the design of a 2-DoF gimbal control system for vehicle-mounted antennas to maintain stable satellite alignment for uninterrupted Internet connectivity. Addressing challenges posed by vehicle movement, the work encompasses developing a simulation environment incorporating mathematical model of an antenna system, sensor fusion algorithms, and advanced control strategies, including
complementary filter, Kalman filter, PID controller, and LQR. The results revealed that while the complementary filter responded more quickly to changes in system dynamics, it suffered from drift, whereas the Kalman filter provided more stable estimations at the cost of a delayed response due to its reliance on model dynamics. The study adopted a hybrid control strategy, employing an LQI controller on the outer gimbal angle and a PID controller on the inner gimbal angle. Future work may integrate signal strength sensors, and adopt quaternion-based orientation to overcome gimbal lock. Conducting a robustness analysis of the controllers can further enhance system reliability. This research provides valuable insights into filtering and control techniques, offering a robust framework for applications requiring precise orientation estimation and efficient control in dynamic systems
Utvärdering och tillämpning av Automatiserade maskininlärningsramverk för finansiell dataanalys
No full textMaskininlärning används idag av många olika sektorer med målet att förutsäga framtiden med hjälp av tekniska databaserade prognoser. Finanssektorn har under en lång period varit ett område som utvecklat teknik för att nyttja möjligheten att prediktera utfall för aktiekurser eller handelsbeslut. Under utvecklingens gång har denna typ av maskininlärning även övergått till det automatiserade med hjälp av automatiserade maskininlärningsramverk. Detta projekt syftar på att tillämpa dessa ramverk och sedan utvärdera resultaten för diverse förutbestämda faktorer.
En stor samling ramverk utvärderades under projektets testfas där de mest lämpade följde med för djupare analys och tillämpning. Faktorer som användarvänlighet, komplexitet och prestanda noterades och de kvarlevande ramverken testades med verklig data. Kod för datamärkning med regressions- och klassifikationsuppgifter utvecklades för ramverken i förberedelse för utfallstester.
Resultaten från utfallstesterna noterades för varje individuellt ramverk och presenterades med hjälp av tabeller och grafer. Hädanefter jämfördes ramverkens resultat för varje förutbestämd faktor och därefter redovisades ramverkens lämplighet inom respektive kategori samt i dess helhet
Makroalger som resurs En sammanställning av näringsåtervinning, vattenrening och möjligheter inom livsmedelsindustrin
No full textThe aim of this paper is to provide an overview of macroalgae (sugar kelp and sea
lettuce) growing along the Swedish west coast and to explore its potential applications
as a protein source and a biofilter for wastewater treatment as well as an integrated
part of a fish farm. The methods used include a comprehensive meta-analysis, study
visits and interviews with experts in the field. Different case studies were conducted,
with the primary aim of calculating the nitrogen and phosphorus uptake of sea lettuce
to assess its potential for filtering wastewater from various sources. Although the aim
was to conduct research on both sugar kelp and sea lettuce, the primary focus and case
studies were centered around sea lettuce due to the limited existing research on sugar
kelp. The results showed that the protein content of macroalgae could be increased
when cultivated in wastewater with high concentrations of nitrogen and phosphorus.
However, sugar kelp did not survive under such conditions. The elevated protein content
may position macroalgae as a potential competitor to established protein sources such as
fish. When combining wastewater treatment with macroalgae cultivation, results showed
that the water could be effectively filtered. This approach could involve integrating
macroalgae cultivation into the effluent stream of a wastewater treatment plant located
in the harbor in Gothenburg. In conclusion, integrating macroalgae cultivation with
wastewater treatment plants could enhance water filtration and improve the quality
of the effluent released into the sea. This approach may contribute to mitigating the
problem of eutrophication. Regarding its potential as a plant-based protein source,
macroalgae cultivation shows great promise, particularly when integrated with the fish
industry. However, the practical implementation in society and industry presents several
challenges. Further research, as well as investments, is needed to enable large-scale
adoption
Wireless Helm Navigation for a Small Passenger Ferry
No full textThis thesis investigates the design and development of a wireless mobile phone application intended to facilitate the safe and efficient navigation of a small passenger ferry. Conducted in collaboration with Cstrider AB, the project applies participatory design and Research Through Design (RTD) methodologies to address cognitive and usability challenges faced by ferry operators. Through interviews, thematic analysis, and iterative prototyping in Figma and Flutter, the study identifies key user requirements, such as simplicity, intuitive controls, traditional visual metaphors, and minimal attentional demand. The final high-fidelity prototype emphasizes skeuomorphic design and streamlined functionality support situational awareness and reduce cognitive workload. While evaluations suggest that the application is generally perceived as intuitive, concerns persist around trust and reliability in safety-critical contexts. This work contributes both a validated interaction prototype and actionable insights into the design of mobile interfaces for maritime control systems, with implications for future development in wireless and remote vessel operation
Frihamnskyrkan - Main model
No full textThe model shows a section through the whole building with a main focus of the church hall and facade. You can see all five different stories of the Frihamnskyrkan starting with the big entrance on the ground floor, followed by its church hall and three stories around a small atrium that host conference rooms, educational facilities and much more
A softer Skutskär; a design methodology on developing public life in a small town centre
No full textSkutskär, the central town in Älvkarleby municipality
with around 6000 inhabitants, is currently developing
its centre since there is a will from the municipality to
increase the attractiveness in the centre and in turn
improve the public life; in other words the social life
which occurs in the open public spaces, by increasing
it. The size and scale of Skutskär places it into a rurban
context which isn’t either the countryside or a city. This in
turn gives special conditions and circumstances.
The thesis ”A softer Skutskär” aims to develop a design
methodology on how to work with increasing public
life in a small town centre. The work is divided into
answering the questions ”What, how, and where?” within
the framework of finding interventions and ways of
working to increase public life in a rurban context centre.
This gives the research question ”How can public life be
increased in the rurban context of Skutskär centre?”.
The design methodology summarizes the research
on public life into five key factors which are needed
for increasing public life: streets, public and private,
microclimate, activity, and scale. The key factors continue
throughout the methodology by looking at different
references, mapping analysis methods, and design
tools based on the key factors. All of this information is
tested on the specific site, Skutskär centre, to explore
the rurban preconditions. Is there a difference compared
to increasing public life in a city? The thesis shows that
the key factors and what you need to achieve might not
differ in the rurban scale, but instead what is needed is
to broaden the perspective of what these key factors
could look like. The metrics for the amount of people
that is seen as a street full of public life might not be the
same in a city compared to a rurban centre, and the main
meeting place could be outside the food shop or by the
bus stop.
A discussion summarizes the findings of the thesis,
and in conclusion the material does not present a
finalized answer, but instead it offers ways of building
an argumentation and presents another way of thinking
when planning in the rurban context
Aging in place within historical heritage; timeworn foundation timeless care
No full textChina’s rapid urbanization has intensified tensions
between modernization and architectural heritage
preservation, particularly in rural areas. Meanwhile, the
aging population faces growing challenges in securing
sustainable, elderly-friendly living environments.
Traditional Chinese architecture holds cultural and
environmental value, yet there is a gap in adapting
these structures for contemporary elderly care while
preserving their heritage. Many historical dwellings
also reflect outdated social hierarchies, including rigid
gender divisions in space.
This study explores the renovation of a historic Huizhou
residential building in rural Anhui Province, integrating
conservation, adaptive reuse, and sustainability to
enhance elderly well-being. Through case studies,
field research, interviews, and theoretical analysis, it
evaluates key factors such as daylight, ventilation,
accessibility, and cultural continuity. Based on these
findings, a sustainable renovation design is proposed,
balancing historical authenticity with modern
functionality.
Grounded in theories of adaptive reuse, aging-inplace,
and sustainable architecture, this study provides
a model for transforming vernacular buildings into
elderly-friendly spaces. It also considers how spatial
modifications can challenge outdated hierarchies
embedded in traditional architecture, making historic
spaces more inclusive while retaining their cultural
essenc
Stochastic Methods for Beam Weight Optimization for Large Antenna Arrays
No full textTraditional methods for beam weight optimization suffer from poor convergence
and long run times for large antenna arrays and are limited to ideal antenna element
models. To overcome these limitations, more efficient optimization algorithms
are needed. Previous research has demonstrated that stochastic approaches such as
genetic algorithms and particle swarm optimization are promising alternatives for
solving this problem. This thesis investigates the tailoring of these stochastic optimization
algorithms for beam weight optimization using high-dimensional, realistic,
simulated data.
We propose the use of an absolute radiation pattern scale as opposed to a relative
one, which allows us to simplify the objective function and promote a more
efficient energy usage. We decrease runtimes by restricting the optimization to a
low-rank search space, implementing warm starts, and leveraging GPU acceleration.
We find that these tailored methods show robust performance and converge significantly
faster compared to current methods. The resulting beams also consistently
outperform those currently in use. We demonstrate the effectiveness of our approach
on antenna arrays of up to 1,152 subarrays, where antennas of 288 subarrays or more
are considered large
Kvantfelskorrektion av bitfel i repetitionskoder med maskininlärning
No full textOne of the biggest challenges with quantum computers is minimizing and correcting errors
that arise due to the instability of quantum systems. The qubits in the system are sensitive
to external factors and disturbances, which can lead to various types of quantum errors,
including bit-flip errors. To address this, there are different methods for predicting and
correcting quantum errors. Two such methods are Minimum Weight Perfect Matching
(MWPM) and machine learning. This report investigates the effectiveness of a trained
graph neural network (GNN) as a decoder by comparing the rate of bit-flip errors in re petition codes to MWPM on IBM’s quantum computers.
The method is based on using Qiskit to construct a quantum circuit capable of detecting
bit-flip errors without collapsing the state of the qubits, thereby obtaining the error rate
in the form of syndromes. Data collected from runs on the quantum computer is used to
train a GNN and is also applied to MWPM for code distances d ∈ [3, 21]. The performance
of the decoders is then compared.
The results showed that a larger GNN with seven graph convolutional layers generally
performed better than both MWPM and a smaller network with three graph convolutio nal layers for all investigated code distances, except for d = 7. MWPM also outperformed
the smaller network, which generally had the lowest performance among the decoders
studied. The method and results are analyzed based on theory and the research question
to provide a broader understanding of how a trained GNN and MWPM behave for diffe rent code distances. Relevant sources of error that may have affected the results are also
presented, along with a discussion of societal and ethical aspects.
In conclusion, the findings indicate that machine learning is an effective method to apply
for decoding in quantum error correction, given sufficiently large and well-trained GNNs
