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.
Senast inlagda
Post
Virtual Verification Framework for Vehicle Motion Systems
(2024) Blakqori, Albijon; Kotur, Mille; Chalmers tekniska högskola / Institutionen för mekanik och maritima vetenskaper; Chalmers University of Technology / Department of Mechanics and Maritime Sciences; Bruzelius, Fredrik; Gröndahl, Albin; Mark, Hans
The automotive industry continually strives to enhance vehicle development processes to be faster, more cost-effective, and of higher quality. This thesis presents
the development of a Virtual Verification Framework (VVF) to improve the Computer Aided Engineering (CAE) verification process for vehicle motion systems.
The focus is on the initial stages of vehicle development, specifically replacing traditional Vehicle-in-the-loop (ViL) testing with more efficient Software-in-the-loop
(SiL) methods.
The framework is developed using IPG CarMaker, a widely adopted simulation
software, and Simulink, allowing detailed subsystem simulations such as braking
systems. The objective is to create a correlated CAE environment that can perform high-fidelity simulations and provide reliable data for system verification. This
involves implementing accurate simulation models, selecting relevant verification
scenarios, and analyzing both simulations’ and real-world data’s performance and
accuracy.
Key research questions addressed include the analysis of output data reliability for
correlation studies between SiL and ViL and the potential expansion of the SiL
stage to replace some aspects of ViL in system verification. The thesis demonstrates that while a complete VVF is not yet realized, significant progress has been
made, particularly in implementing system-specific models and functional testing
within CarMaker for Simulink (CM4SL).
Challenges identified include simulated and real-world data discrepancies, particularly with tire modeling and sensor frequency differences. Despite these, the framework shows promise for future scalability and application, aiming to reduce reliance
on physical prototypes, enhance safety in early-stage testing, and streamline the
vehicle development process. The work concludes that a more robust and trustworthy virtual verification environment can be established, significantly benefiting the
automotive industry’s development cycles.
Post
Invariant Feature Extraction for Power Quality Disturbances Using Deep Learning - Compact and Interpretable Features Enhance Performance on Downstream Tasks
(2024) Stenhede Johansson, Elias; Schütz, Valter; Chalmers tekniska högskola / Institutionen för elektroteknik; Hammarstrand, Lars; Olsson, Viktor
ABSTRACT
Feature extraction is a crucial step for tasks such as classification, clustering and predictive risk estimation. In power quality analysis, feature selection is often performed by domain experts. However, crafting a small yet sufficiently informative feature set is often challenging, particularly when evaluation data is scarce. In this thesis, we present an autoencoder-based feature extraction method tailored for spectrograms of three-phase current and voltage waveforms. The extracted feature set is invariant with respect to time translation and sample length, and we demonstrate that sufficient information is retained to reconstruct the spectrogram with high accuracy. By applying Uniform Manifold Approximation and Projection (UMAP) with a specific pseudometric, invariance with respect to phase permutations can be obtained while further reducing the dimensionality of the extracted features. The interpretability of the extracted feature sets is evaluated by observing changes in reconstructed signals when the latent variables are perturbed. The usefulness of the features is measured through three tasks: clustering, fault prediction and root cause disturbance classification. For the fault prediction task, we demonstrate that training an LSTM model with UMAP features significantly increases the AUC value compared to training the same model with manually selected features by a domain expert. In the classification task, both autoencoder features and UMAP features result in a higher macro 𝐹1-score than manual features when training a neural network classifier, regardless of the training set size. The improvements are particularly notable for the smallest training sets. Additionally, using a model-free label propagation method on these features further enhances performance. Interestingly, we find that pretraining the autoencoder improves reconstruction fidelity, even when the pretraining dataset consists of audio recordings that are quite different from power quality measurements.
Post
Autonomous Driving via Imitation Learning in a Small-Scale Automotive Platform
(2024) Wellander, Johan; Petersén, Arvid; Chalmers tekniska högskola / Institutionen för elektroteknik; Hammarstrand, Lars; Ebadi, Hamid
Abstract
In recent years, the advancement of autonomous driving (AD) technology has garnered significant interest. Traditionally, AD systems have relied on multiple submodules, each handling specific tasks such as perception, path planning, and vehicle control. However, an emerging alternative is the implementation of end-to-end systems, which directly process sensor input to predict vehicle control. While both reinforcement learning (RL) and imitation learning (IL) are utilized
in end-to-end AD systems, RL often finds its strength in simulated environments, where agents learn through exploration and failure. In contrast, IL, learning from an expert model or human, proves more suitable for real-world applications, requiring substantially less data. This thesis presents an implementation of IL for achieving autonomous driving on a go-kart platform. Leveraging both behavioral cloning (BC) and Human Gated Dataset Aggregation (HG-DAgger), we compare the impact of using an interactive IL algorithm HG-DAgger compared to BC. Additionally, our research explores the use of different inputs, including color camera, stereo depth camera, IMU, and the position of ORB features. We also detail the development of a comprehensive software pipeline encompassing data collection, data formatting, model training, and go-kart control. For evaluation, the go-kart was driven around a track for three laps using the trained
BC and HG-DAgger models, and assessed based on number of interventions required per lap, distance without accident, lap time, lap time deviation. The results from the evaluation indicate an improvement in performance from using HG-DAgger over BC as well as an improvement from using a stereo depth camera or the position of ORB features as supplementary inputs to the color camera and IMU.
Post
Exploring opportunities and challenges of increased circularity within the textile industry - A case study at Sätila of Sweden
(2024) Gamble, Julia; Hultfeldt, Matilda; Chalmers tekniska högskola / Institutionen för teknikens ekonomi och organisation; Chalmers University of Technology / Department of Technology Management and Economics; Hellström, Andreas; Hellström, Andreas
Improving circularity and closing the loops in resource use is gaining more and more attention from scholars, practitioners, policymakers and governments. Adopting circular economy practices is pointed out as a necessity in the move towards an increased sustainability throughout different industries. The textile industry is currently under pressure to change the way it uses resources. Hence, new regulations and way of practice is therefore being promoted. To reduce environmental impact, businesses in the textile industry can rethink the traditional linear business models towards circularity, e. g extended product life. Circular economy applications are accompanied by several types of challenges but also opportunities, which is what this thesis aims to explore. Examining these aspects by a conducted data collection, through literature reviews and interviews with experienced people in the business. The findings are put into the context of a small clothing manufacturer, Sätila of Sweden, examining their possible route of becoming more circular. Transitioning from a linear economy to a circular economy requires efforts to decrease the use of virgin materials and slowing down the loop of resource usage. Encouraging longer uses, recycling, and reuse of already existing textile products are actions to proceed with in order to slow down the loops.
Post
PEMFC catalyst dissolution and degradation measurements with EQCM-D
(2024) Bahraminasab, Mina; Kothala, Gnana Lahari; Viriyaparp, Intiporn; Chalmers tekniska högskola; Chalmers University of Technolog; Wickman, Björn; Rieger, Nils; Wickman, Björn