University of Twente Student Theses
Crowdsensing road quality using inertial measurement units on bicycles
Oude, S.J.M. de (2024) Crowdsensing road quality using inertial measurement units on bicycles.
|
PDF
969kB |
| Abstract: | This work examines methods to enhance crowdsensing of road quality using Inertial Measurement Units (IMUs) on bicycles. The focus of this research has been on three main areas. Firstly, a centralized machine learning model was developed to classify road types based on different pavements. Secondly, a classical machine learning model was developed to classify participants and identify unique features of participants. Lastly, a federated learning model has been created to explore its potential to classify road quality while mitigating privacy issues. Through this study, significant personalized factors were found that impact the accuracy and generalizability of road quality classification models. Whereas prior studies have often overlooked these personalized biases, this research highlights their importance in developing robust and universally applicable models. Although, the federated learning approach did not fully mitigate these biases, it offers promising direction for future research to achieve more universally applicable road quality insights. |
| Item Type: | Essay (Bachelor) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science |
| Programme: | Computer Science BSc (56964) |
| Link to this item: | https://purl.utwente.nl/essays/101929 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page