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

Physical fault detection on the Controller Area Network - An investigation into the feasibility of detecting physical faults on the CAN bus with machine learning
(2026) Engström, Pontus; Roos, Rikard
Detecting physical faults on a Controller Area Network is a complex task due to unpredictable signal behavior under normal circumstances, and signal behavior unique to the configuration of the specific network and its components. Furthermore, physical faults can lead to disruptive signal behavior, which are not always detected by the protocol’s error handling. If it does not activate the fault handling of the protocol, the disruptive behavior is only seen on the physical layer, potentially resembling unpredictable behavior under normal states. Such physical faults can imply a marginal system, leading to unpredictable signal behavior over time. If the fault is detected by the protocol, the source of the fault is not detected, and further investigation of how and where the fault occurred is required. Traditional fault detection methods include manual analysis, which requires time and expertise. This thesis investigates the feasibility of applying machine learning methods to achieve physical fault detection on a CAN bus. The aim was to determine if, and to which extent, this could be achieved. A physical test bench was configured to enable CAN traffic under normal and faulty configurations. By calculating the offsets of the expected position of recessive signal edges, minimal data was used to distinguish physical faults from normal conditions. A range of machine learning classifiers were trained and evaluated to exhaust the possibilities of achieving high accuracy on this task. The problem was narrowed down to detect termination faults, and in turn, to localize the faults. The results indicates that it is feasible to differentiate termination faults with data-driven methods. In particular, Gradient Boosting, a decision tree ensemble architecture, achieved a 99.98% accuracy on this task. However, localizing the faults was best achieved with Gradient Boosting with an accuracy of 93.74 %. Two distinct fault classes were of too similar shape to be separable to a perfect degree. The main limitation of this study included the lack of generalization capabilities. The signal data proved to be tightly connected to the physical setup of the CAN bus, and applying a model to different network than it was trained on is very likely to fail. Furthermore, the data was generated under stable, indoors conditions, with high-quality hardware. A further investigation of the real-world practical use is required.
Chalmershindret 2026
(2026) Carlsson, Julia; Thylander, Nellie; Tinghall, Julia; Torstensson, Johanna; Wallh, Magdalena
Assessing potential causality between blood metabolites and primary sclerosing cholangitis using two-sample Mendelian randomization
ROSENGREN, OLA
Primary sclerosing cholangitis (PSC) is a chronic cholestatic liver disease with a poorly understood aetiology. An observational study by Molinaro et al. have suggested an association between blood circulating metabolites and PSC, however, causal inference is limited by confounding and reverse causation. The aim of this study was to assess potential causal effects of blood circulating metabolites on PSC risk using Mendelian randomization (MR), and to compare the findings to the associations found in the observational study. Atwo-sample MR framework was applied using summary statistics from genome-wide association studies (GWAS) of PSC-associated metabolites and PSC. Instrumental variables (IVs) were selected based on genome-wide significance and linkage disequilibrium pruning. Causal effects were estimated using the inverse variance weighted method, with sensitivity analysis performed to assess heterogeneity and horizontal pleiotropy. MR analysis were conducted for seven metabolites previously associated with PSC and seven control metabolites with no association. Overall, the selected IVs explained only a small proportion of variance in metabolite levels, as reflected by low F-statistics, indicating weak instrument strength and a potential bias of causal effect estimates towards the null. No evidence of a causal effect was detected for either PSC-associated metabolites nor the control metabolites. Sensitivity analysis indicated moderate heterogeneity with little evidence of horizontal pleiotropy. These findings suggest that currently available IVs for the blood circulating metabolites are insufficient to robustly assess causality in PSC. While no causal effects were detected, this study highlights important limitations in metabolite-based MR and underscores the need for larger and more powerful metabolite GWAS to enable reliable causal inference.
App-Operated Robot with Live Streaming and environmental sensor
(2026) Firuzian, Reza
Being in inaccessible or dangerous environments requires remote controlled observation and control systems. This report describes the steps necessary for connecting various hardware components to each other and the software required to combine the hardware and software into a functioning system. This project presents the construction and implementation of an app-operated mobile robot capable of navigating different terrains, providing live video and image capture, and measuring ambient temperature. The system is built around a Jetson Orin Nano and integrates multiple hardware components, including a motor driver, motors, cameras, an environmental sensor and a 4G modem for remote communication. A server running on the robot handles video streaming, motor control commands, sensor data transmission, while a customdeveloped mobile application enables real-time monitoring and remote control. The robot was developed in a modular manner, allowing individual features to be implemented and tested incrementally. The results show successful remote control, live video streaming, and temperature monitoring, although communication instability due to the 4G modem affected latency and control responsiveness. Overall, the project demonstrates the feasibility of integrating hardware and software into a functional remotely operated robotic platform and provides a foundation for future enhancements.
Återtagning av små fastvingsdrönare i sjöräddningsmiljö
Bergman, Andreas
This thesis work was conducted in collaboration with Sjöräddningssällskapet (SSRS), which utilizes unmanned aerial vehicles (UAVs) to support maritime rescue operations. In coastal environments with small landing space and variable wind conditions, conventional landing methods for fixed-wing UAVs are not always suitable. The objective of this project was therefore to develop a conceptual system for safe mid-air capture and retrieval of a small fixed-wing UAV adapted to SSRS operational requirements. The work followed a structured product development process. After establishing system requirements and performing a functional analysis, solution principles were generated using a morphological matrix. Partial solutions and overall concepts were reduced through a systematic elimination process, excluding technically unsuitable or impractical alternatives. The remaining concepts were further analysed with respect to energy absorption, feasibility, and operational applicability. The final concept consists of a portable, modular capture system based on mechanical energy absorption through a tensioned line between two interconnected units. The system is dimensioned according to the UAV’s kinetic energy at approach, enabling a compact design compared to large-scale commercial systems. Analytical calculations indicate that the concept is feasible at a conceptual level and capable of handling the expected loads. The study demonstrates that an established capture principle can be scaled and adapted to maritime rescue applications through systematic concept development. Future work should include prototype development and experimental validation under realistic operating conditions.