Chalmers Open Digital Repository
Not a member yet
26247 research outputs found
Sort by
Investigating the Potential Effects of Intelligent Access
No full textThe rapid development of freight transport has introduced challenges related to infrastructure capacity, environmental sustainability, and logistical efficiency. The
purpose of this thesis is to investigate contemporary and future aspects of autonomous driving, high capacity transport, smart roads and smart parking to achieve
IA for road freight transport and port- and terminal logistics. To the knowledge of
the authors, no research that focused on the holistic view of all these areas, encompassing both contemporary and future aspects in relation to Intelligent Access, have
been found. The study uses a qualitative research methodology, which combines
literature review and expert interviews to analyze current and future frameworks
and concepts related to aforementioned areas. It identifies how frameworks like
SAE levels for AVs and Performance-Based Standards (PBS) for HCT can be incorporated into IA systems to improve route planning, reduce emissions, and increase
infrastructure longevity. Key findings reveal serious barriers to implementation including fragmented regulations, limited infrastructure readiness and challenges in
data governance. Nonetheless, the research outlines how coordinated deployment of
IA can improve efficiency, reduce environmental impact, and enhance the resilience
of freight transport systems. By aligning technical innovations with coherent policy frameworks, this thesis outlines pathways toward a smarter, safer, and more
sustainable freight transport ecosystem. The thesis contributes to a general understanding of how digital and automated innovations can support the transition
toward sustainable and intelligent logistics networks
Design of a superconducting Bias-Tee
No full textThe tight integration of qubits is a significant challenge for the development of
modern quantum computers. Closely placed qubits will be subjected to frequency
crowding. Flux tuning of qubits can reduce the frequency crowding by changing the
resonance frequency of a qubit by changing the magnetic flux, which requires the
coupling of DC and RF signals. A bias-tee implements the diplexing of DC power
and RF signals.
In this thesis, We report findings on a microwave bias-tee designed to be fabricated
as an integrated passive device in a superconducting process on silicon or quartz.
A bias-tee requires an inductive branch and a capacitive branch to diplex DC signals
and RF signals. Therefore, passive components such as inductors and capacitors
have been modelled and simulated. Moreover, the effect of the wafer, dielectrics, and
packaging techniques are studied to describe the non-ideal performance of passive
components.
One of the challenges of simulating thin film devices in the Finite Element Method
(FEM) solvers is the issue of dielectric thickness compared to the rest of the structure.
The large difference in dimensions leads to erroneous mesh elements. This
thesis also introduces a novel detailed explanation of the meshing of thin films in
FEM solvers that has not been thoroughly covered in the existing literature.
The simulation results show capacitors with a quality factor of 7000 realised in a
silicon process and a quality factor of 4000 realized in a quartz process. We also
report the design of planar coils, which have a quality factor of 2400 for the silicon
process and 2000 for the quartz process
Phantom limb with Variable Stiffness Actuator
No full textTechnology and healthcare are improving faster than ever, and with it comes an increased demand to be able to test new innovations before they get approved for use. However, with strict guidelines on how these tests are allowed to be performed, it is not always that easy. One of the areas affected by these guidelines is the continued development of better exoskeletal arms for people with impaired movement. Due to the risk of potential injury when testing these products in a patient who does not have full control over their arm, and who might experience rapid and spasmodic movement in it, testing on people is highly limited [1].
Therefore, this thesis presents the design and development of a robotic phantom limb with a VSA (Variable Stiffness Actuator), intended as a test platform for exoskeletal devices used in rehabilitation. The motivation comes from the need to test exoskeletons safely without relying on human subjects, particularly for conditions involving involuntary or spasmodic movements. The project combines mechanical, electrical and control system engineering to create an anthropomorphic arm capable of replicating realistic joint dynamics and variable stiffness behaviors.
Key objectives included designing the VSA using nonlinear springs to mimic human muscle behavior, implementing a control system using an STM32 microcontroller and ROS2 architecture, and integrating sensors such as IMUs, hall sensors, and force-torque sensors to monitor system performance. The system was modeled and simulated in Simulink to validate control strategies and mechanical performance prior to physical implementation. The experimental results demonstrated that the prototype could achieve the target stiffness levels, torque outputs, and angular positions necessary to simulate human arm movement under various conditions. This work contributes a valuable tool to the advancement of rehabilitation robotics by enabling more efficient and safer testing of assistive devices, ultimately aiming to accelerate the development and clinical deployment of user-centered robotic systems
Event Detection in Smart Meter Data with Complex Event Processing
No full textSmart Grids with their accompanying Smart Meters are increasingly taking over
from conventional electrical grids and manually inspected electricity meters. This
trend gives rise to large amounts of data being collected every day by electricity
service providers. Each Smart Meter may emit several readings each hour and can
be located at both customers and producers, as well as throughout the infrastructure
in locations such as substations.
Through analysing this data, a service provider can respond more swiftly to changes
in supply and demand, as well as detect anomalies in the grid and at meters. But
the large amount of data that is generated quickly exceeds what could be manually
inspected and requires the use of techniques made to analyse large quantities of
data. One approach is to analyse the data as it arrives in the form of a data stream,
without the need to first save it to permanent memory. Two prominent techniques for
analysing data streams are those of stream processing and complex event processing.
This thesis is conducted in collaboration with Göteborg Energi. It investigates
the performance difference of stream processing and complex event processing for
pattern detection in Smart Meter data. The findings are further used to guide
the implementation of a pattern that combines both techniques and to support the
creation of pattern templates. The tests are conducted on three patterns, with a
primary focus on comparing the effects on latency and throughput under different
levels of source parallelism in the jobs. The results show that while stream processing
has a performance advantage over complex event processing on Smart Grid data,
combining the two techniques can achieve comparable performance. Using stream
processing as an aggregation step before complex event processing can maintain
high performance while offering potential reusability and simplifying the creation of
future patterns
Visualization of CI/CD Flows Using Eiffel Events: An interactive dashboard for viewing pipelines of Eiffel Events
No full textThe Eiffel protocol was developed by Ericsson to enable technology agnostic communication between continuous integration and continuous delivery/deployment (CI/CD) ecosystems. In the Eiffel protocol, every action taken in a CI/CD pipeline is an event. Currently, there is no open-source tool available to get a clear overview of events or view a pipeline flow that is intuitive and easy to understand. The goal was to create a web application that uses the data from Eiffel events and displays the sequence of events and their contents with a dashboard. The application was built using Next.js, Express.js, and MongoDB. It is possible to browse through repositories and branches, and select pipelines to view in their entirety. The visually accessible page showcasing the event pipeline helps the user understand how different events relate to each other and how one sequence led to another. The resulting web application has been validated through positive feedback by stakeholders at Nexer and has helped lay the foundation for how an event-visualizing web application could look and function. An application similar to the one created in this project can in the future help developers and stakeholders follow and analyze Eiffel event pipelines with ease
Designing Corporate Incubators for Strategic Value Creation
No full textAs market dynamics accelerate, companies are increasingly using open innovation
to complement internal R&D. Corporate incubators have become a popular way to
access new technologies and foster entrepreneurship, though there is still limited
insight into how they should be designed to create value for the company. This
master’s thesis explores how corporate incubators can be designed to create value
for the parent company. The study focuses on the strategic, structural, and
operational design choices involved in establishing and running a corporate
incubator.
This study is based on 23 semi-structured interviews with people working in
Swedish corporate incubators, startups, venture capital firms, and public
organizations. Even though the incubators are set up and run in different ways,
several common patterns and challenges were found. By combining insights from
the interviews with earlier research, the study presents eight key components that
companies should think about when designing a corporate incubator: strategic
focus, source of ideas, governance, location, equipment, access to firm’s resources,
funding, and networks. These areas are important for building an effective
incubator. The study also identifies four main types of values that incubators can
bring to the parent company: helping with technology development, promoting an
entrepreneurial culture, spotting new trends, and improving the company’s brand.
The discussion explains which components are especially important for creating
each type of value.
The findings underline the importance of tailoring the incubator’s design to the
parent company’s strategic goals while remaining open to external influences.
Different startups, objectives of the incubators and cultural challenges affect how
the incubator should be designed. The study provides insights for firms aiming to
launch or refine a corporate incubator and guidance on how design choices
influence the type and extent of value generated
Dirty preservation: An experimental counterpractice
No full textDue to pressure of economic growth and a
drive towards newness, the built environment
is suffering from an accelerated process
of breaking, going out of style, and being
replaced by something new – a broken system
which will result in the demolition of 2 billion
square meters of built space in Europe by 2050
(HouseEurope!, 2025). This thesis departs
from a frustration with the destructive cycles
of redevelopment and takes the stance that
even dirty and overlooked buildings must be
preserved. This statement requires a profound
shift within the preservation practice, which
usually includes objects of undisputed cultural
and historical significance. The question
is then, what happens when we declare an
“insignificant” building to be an object worthy
of preservation, care and affection?
To test out an alternative method of
preservation, an abandoned boiler plant from
the 1950s, located in a Stockholm suburb, was
chosen as subject of interest for this thesis.
The building awaits demolition and shows
apparent signs of neglect and decay. Elevating
the building and looking at it as a cultural
heritage object, conventional methods of
monument documentation and evaluation
are applied. The process is driven forward
by smudging the conventional practice with
critical theory, messing with its norms and
expected outcomes, all with the aim to develop
a morphed and dirty preservation method
which can generate new perspectives on value,
as well as new modes to preserve.
Acting as dirty preservationists, our objective
has not been to reprogram or transform the
building, but to remain with the uncertainty of
evaluation and care. By applying preservation
methods typically reserved for undisputed
heritage to a neglected boiler plant, we
question what qualifies as worth preserving.
The resulting instructions – based on repair,
reconstruction, and site-sourced materials –
are shaped by care rather than economic gain.
This thesis argues for a broader, more inclusive
understanding of heritage, one that treats
the overlooked as valuable and preservation
as a creative, reparative force rather than a
conservative one
Advanced thermal regulation of muffler
No full textTighter legislation on emissions, such as the upcoming EURO-7, puts increased pressure on manufacturers to further decrease emissions from their vehicles. This study address Nitrogen Oxides (NOx) emission from heavy-duty trucks. For the selective catalytic reduction catalyst, an important component of the exhaust after-treatment system to properly react NOx into pure nitrogen and water, is that the temperature
needs to be sufficiently high in the muffler. During extended periods of low engine load, such as downhills or idling, the muffler cools down as colder exhaust enters the muffler, as a result NOx emissions increases.
To address this, the study investigates a thermal property of the muffler’s dynamics that allows the use of time normalization, which simplifies the dynamics. Time normalization was then utilized in the design of the resulting model predictive controller. Furthermore, the impact of incorporating preview information was investigated to determine what benefits could be gained by using a model predictive controller with
preview, compared to one without.
The results demonstrated that it was possible to utilize the structure of the dynamics to construct a controller in a linear time-invariant environment, to control the linear time-varying system. Additionally, incorporating preview information improved the performance in cases where the exhaust temperature drops fell outside the region of which the controller could compensate. In such cases the controller pre-heated the muffler to maintain a sufficient temperature, thus extending the duration of constraint satisfaction but at the cost of increased energy consumption. However, when the temperature drop was within the region of what the controller could supply, the controller without preview performed equally well
Contrastive Learning For Molecular Representation
No full textThis thesis explores the integration of contrastive learning into REINVENT, AstraZeneca’s in-house generative model for molecular design, with the aim of improving the model’s understanding of chemical equivalence between different SMILES
representations of the same compound. To this end, a contrastive learning framework was developed, incorporating SMILES-based data augmentation techniques such as enumeration and subgraphing. The framework was evaluated on three datasets: a proprietary baseline derived from ChEMBL35, and the publicly available MOSES and GuacaMol datasets. To assess the impact of architectural design on performance, multiple model architectures were investigated, including a newly introduced intermediate architecture. Results indicate that the intermediate architecture consistently achieves higher validity across all datasets, but tends to reduce novelty. Furthermore, using multiple augmentation strategies improved the model’s ability to generate chemically diverse and novel compounds, as measured by metrics such as novelty and Fréchet ChemNet Distance (FCD). These findings suggest that contrastive learning can offer measurable benefits in de novo molecule generation, although its effectiveness may depend heavily on architecture and dataset-specific tuning