Chalmers Open Digital Repository

Välkommen till Chalmers öppna digitala arkiv!

Här hittar du:

Studentarbeten utgivna på lärosätet, såväl kandidatarbeten som examensarbeten på grund- och masternivå
Digitala specialsamlingar, som t ex Chalmers modellkammare
Utvalda projektrapporter

Forskningspublikationer, rapporter och avhandlingar hittar du i research.chalmers.se

Enheter i Chalmers ODR

Välj en enhet för att se alla samlingar.

Visar 1 - 2 av 2

Senast publicerade

Kompakt mikrovågssystem för detektering av intrakraniella blödningar
(2026) Agfors, Lovisa; Bergman, Naëmi; Ek, Andreas; Ingels, Virve; Murén, Izabella; Myllykangas, Lina
Stroke is one of the leading causes of death in Sweden and worldwide. Despite the fact that the time between symptom onset, diagnosis and treatment has a major impact on a patient’s chances of survival and recovery, there are currently no effective and reliable methods for prehospital stroke diagnosis. Microwave imaging has been proposed as a potential solution for detecting intracranial hemorrhages, such as stroke, already in the ambulance. However, these measurement systems are limited by factors such as high cost and bulky design. The aim of this project was to further develop and evaluate a compact and cost-effective measurement system designed to achieve performance comparable to that of a Vector Network Analyzer (VNA). To enable the measurement of weaker signals, the system was modified to reduce noise levels and crosstalk between system components. Different configurations using different components were tested and evaluated with respect to signal-to-noise ratio, isolation, and the ability to detect weaker signals. To evaluate the performance of the system and to develop a simple machine learning algorithm, a phantom model of a human head both with and without hemorrhage was constructed. The measurement results showed that the isolation of the measurement system had increased from 30 dB to 100 dB, and that its ability to detect weaker signals had improved. The measurement system and the VNA produced comparable measurement results for phantoms with and without hemorrhage. However, the measurement system still exhibited higher noise levels and lower precision compared to the VNA, indicating that further development is necessary. Although it was difficult to visually distinguish between measurement results from phantoms with and without hemorrhage, the classification algorithm achieved a high accuracy (81%) and an AUC value of 0.963.
AI/ML Applications to Identify Tank Cleaning Operations & Quantify Slop Discharge
(2026) Alabdalla, Omar; Alreda, Alamin
Chemical tankers discharge contaminated washwater from tank-cleaning operations under conditions regulated by MARPOL Annex II and the IBC Code, yet compliance cannot generally be verified externally at the time of discharge. Vessels are not required to report when, where or in what volume tank cleaning occurs and current monitoring relies on aerial surveillance that is constrained by weather, daylight and geographic coverage. This leaves a gap in the environmental oversight of chemical tanker traffic in sensitive regions such as the Baltic Sea. Crucially, no reliable labelled record of confirmed tank-cleaning events exists, which precludes supervised machine learning approaches. This thesis develops and evaluates a reproducible, label-free pipeline for identifying vessel trajectories whose navigational behaviour is potentially consistent with tank cleaning activity, using historical AIS data. The work is conducted in collaboration with Scanjet AB and is based on AIS records for the Baltic Sea from January to March 2021. Segmented vessel trajectories are enriched with engineered behavioural features; speed variation, turning-angle variation and drift, and compressed before being rendered as multi-channel image tensors. A self-supervised representation learning framework (BYOL) with a custom convolutional encoder learns compact behavioural embeddings from these tensors without labels, and HDBSCAN cluster ing is applied to isolate behaviourally uncommon trajectories for expert-informed interpretation. Two trajectory compression methods, DP and TDKC, are system atically compared and the contribution of the engineered behavioural features is assessed through ablation. Applied to 91,431 learned trajectory embeddings, the framework organised vessel movement into a small number of coherent behavioural clusters while isolating a mi nority of sparse, behaviourally distinct trajectories consistent with irregular manoeu vring, looping and offshore-deviation patterns. Behavioural feature engineering was found to be decisive: removing the drift channel fragmented the embedding space and increased the proportion of unassignable trajectories from 36% to 69%. The compression comparison showed that, in the configurations tested, DP supported more coherent representation learning than TDKC; this difference is attributable primarily to the amount of trajectory structure retained during compression rather than to the compression criterion itself. Detected anomalies were combined with operational washing-system metadata from Scanjet AB to produce scenario-based estimates of potential washwater discharge volumes. The pipeline is presented as a detection-support tool rather than an autonomous classifier: its outputs are candidate trajectories requiring expert review, not con firmed evidence of discharge. The thesis demonstrates that self-supervised repre sentation learning provides a viable foundation for large-scale behavioural anomaly analysis in label-scarce maritime monitoring contexts and offers a transferable basis for future environmental-monitoring systems.
Autonomous Navigation In Real-Time Endovascular Simulation
(2026) Magnusson, Niklas; Pettersson, Lukas
This Master’s thesis investigates the use of deep reinforcement learning for autonomous navigation in a simulated endovascular environment. Specifically, a Soft Actor-Critic (SAC) algorithm is employed to train an agent to control a micro guidewire and a micro catheter for path-following tasks. The simulation environment is based on the VIST simulator, where the agent observes a 23-dimensional state representation capturing relevant path-following information. The agent was rewarded for path following, with the reward function incorporating vessel centerline alignment, penalization of deviation from the path, and progression toward the target. The agent produces continuous translation and rotation commands for the respective tools. Training was conducted on four anatomies over 1600 episodes, each consisting of up to 500 time steps depending on task completion. The best performing model emerged after 800 training episodes, which achieved a success rate of 94 % on validation data. During testing, it achieved a 96 % success rate on an unseen cerebral anatomy and an 88 % success rate on an unseen liver anatomy. This indicates that the agent learned transferable navigation strategies rather than anatomy specific memorization, which demonstrates that deep reinforcement learning is a viable approach for endovascular navigation in a simulated environment.
Towards Reliable Retrieval Systems: Design and Evaluation of an Agentic GraphRAG Pipeline
(2026) Spreitz, Adam
Large language models are increasingly used for knowledge-intensive question answering, but their reliability remains limited when answers must be grounded in domain-specific documents. This limitation is particularly important in high-assurance environments, where systems must be controllable, traceable, and deployable without relying on external infrastructure. This thesis investigates the design and evaluation of an agentic GraphRAG system for document-grounded question answering, using scientific literature as a controlled proxy for technical internal documentation. The implemented system combines document ingestion, section-aware chunking, knowledge graph construction, vector indexing, graph traversal, reranking, and languagemodel- based answer generation. Two retrieval architectures are compared under shared conditions: a VectorRAG baseline using iterative hybrid search and a GraphRAG system using community-first hierarchical traversal over an explicit knowledge graph. The systems are evaluated across multiple retrieval configurations, generation models, and query sets using automated RAG evaluation metrics, statistical tests, and pairwise LLM-as-judge comparisons. The results show a clear divergence between retrieval-oriented metrics and answerlevel evaluation. VectorRAG achieves stronger automated retrieval scores, particularly on context precision and recall, while GraphRAG is preferred in holistic answer comparisons and manual validation. This suggests that chunk-level retrieval metrics do not always capture the usefulness of structurally retrieved evidence for downstream answer generation. The findings indicate that graph-based retrieval can improve answer quality and interpretability in agentic RAG systems, but also highlight important limitations related to dataset construction, evaluator dependence, corpus scale, graph quality, and agentic control.
Konceptutveckling av ett belastningsgränssnitt - En användarcentrerad produktutvecklingsprocess för en optimerad squat
(2026) Hansson, Alicia; Sandell, Linnéa
This report documents the development of a load interface for belt-squat machines, conducted on behalf of a fitness equipment company. The purpose of the project was to investigate and develop a concept that maximizes comfort and usability for a broad target group for usage in commercial gym environments. Through a user-centered design process, including market analyses and user testing of existing products, the areas for pain at pressure points, instability, and a lack of user-friendliness were identified as objects for development. The work resulted in three conceptual solutions that were evaluated against a requirement specification, where the V-band concept was selected as the final design proposal. The result is a detailed-designed belt-squat belt that implements a user-friendly interface design. Through a combination of a stable backplate and a unique, foldable geometry, the pressure distribution over the hips and lower back is optimized without restricting the user's range of motion. To lower the barrier to use, the belt features a numbered adjustment system for specific settings and a permanent anchoring to the machine, which minimizes the risk of misuse and misplaced equipment. The concept has been dimensioned based on anthropometric data to include users within the 5th to 95th percentiles of the Swedish population. The conclusion of the project is that by prioritizing user comfort, mechanical simplicity, and an intuitive visual interface, the user experience of a belt-squat belt can be significantly improved, thereby making the execution of the exercise more accessible and safe for a universal target group.