Hide menu

Examensarbeten och uppsatser / Final Theses

Framläggningar på IDA / Presentations at IDA


Se även framläggningar annonserade hos ISY och ITN i Norrköping / See also presentations announced at ISY and ITN in Norrköping (in Swedish)

If nothing is stated about the presentation language then the presentation is in Swedish.

WExUpp - kommande framläggningar
2024-05-13 - AIICS
Identifiera beteendemönster hos truckförare för att förutspå risk för olycka
Victoria Winqvist, Unn Zachrison
Avancerad (30hp)
kl 14:15, Alan Turing (På svenska)
[Abstract]
This thesis explores the possibility of identifying risk behaviour patterns among forklift drivers through the analysis of telemetry data using unsupervised clustering algorithms. The objective is to predict whether certain behaviour patterns increase the risk of accidents. With the increasing accessibility of Internet of Things technology, data from forklifts has become more available, allowing for the study of driver behaviour. The telemetry data utilised is sourced from Toyota Material Handling Manufacturer Sweden’s internal database, collected from Data Handling Units that are installed on forklifts across Europe. This data, known as "shock data," is triggered when a force is applied to the forklift, such as a collision. The thesis investigates combinations of various clustering algorithms and dataset modifications. The evaluation of the results is conducted using several quantitative measures and visualisation, along with analysis of time distribution, geographical placement, comparison of forklift models, and comparison with "no shock data." The evaluation yields K-Prototypes and K-Means as the best performing algorithms, while indicating that soft clustering and density-based clustering are not well-suited for the data. The identified best performing algorithms reveal two recurring driver behaviour patterns: the first one being driving forward at high speed with the lift motor idle, and the second pattern being driving backward at low speed while lowering the forks. However, a majority of the data points remain unclassified into specific behaviour patterns, suggesting that the dataset or methods used may not be sufficient enough. This prompts further exploration into the inclusion of additional features such as steering angle and forklift height. The thesis demonstrates the feasibility of identifying risk behaviour patterns, with potential for future research expanding on the findings to further contribute to the prevention of workplace accidents involving forklifts.
2024-05-16 - ADIT
Analys, design och utvärdering av databasscheman i Azure Data Explorer
Angelica Ferlin, Linn Petersson
Avancerad (30hp)
kl 13:00, Babbage (In English)
[Abstract]
Data warehouses are today used to store large amounts of data. This thesis investigates the impact different database schema designs have on query execution time within the cloud platform Azure Data Explorer. It is a relatively new platform, and limited research exists on how the database schema should be designed in Azure Data Explorer. Further, the design of the database schema has a direct impact on the query execution times. The design should also align with the use case of the data warehouse. This thesis conducts a requirements analysis, determines the use case, and designs three database schemas. The three database schemas are implemented and evaluated through a performance test. Schema 1 is designed utilizing results tables from stored functions, while schema 2 utilizes sub-functions divided by different departments or products aimed to minimize the data accessed per query. Finally, schema 3 uses the results tables from the sub-functions found in schema 2. The conclusion from the performance test shows that schema 3 has the best overall improvement in terms of query execution time compared to the other designs and the original design. The findings emphasize the critical role of database schema design in influencing query performance. Additionally, a conclusion is drawn that using more than one approach to enhance query performance is increasing the potential query performance.
2024-05-20 - HCS
Damage Assessment on Remote Sensing Imagery with Foundation Models
Gustaf Lindgren
Avancerad (30hp)
kl 08:15, Alan Turing (In English)
[Abstract]
There is currently an ongoing paradigm shift in machine learning; instead of training task-specific models from scratch, foundation models i.e., large pre-trained models are adapted for various downstream tasks. Foundation models excel in zero- and few-shot learning, ideal for domains with limited labeled data, such as disaster assessment on remote sensing imagery (RSI).

This thesis explores how the foundation models CLIP and SAM can be utilized to classify RSI affected by natural disasters and segment intact and damaged infrastructure without extensive retraining. For the scene classifications, various text prompt techniques are tested as well as zero-shot prompting with images. Moreover, few-shot learning methods such as linear probing and prompt learning are explored. For the open vocabulary semantic segmentation task, "pipelines" are implemented that leverage the open vocabulary classification abilities of CLIP and zero-shot image segmentation capabilities of SAM.

This work demonstrates that foundation models can be used effectively for detecting flooding on RSI and there were promising results on other disaster types as well. While handcrafted text prompts yielded the best accuracy, the zero- and few-shot learning methods with images offered a better trade-off between accuracy and consistency. Although the performance of the zero-shot segmentation pipelines was generally poor, they showcased the potential of SAM for accurate segmentations on disaster imagery when being provided with prompts of sufficient quality.

Page responsible: Final Thesis Coordinator
Last updated: 2022-06-03