Apply now »

Thesis Work - Deep Learning-based Multi-Modal Fusion and Alignment strategies for Real-World Driving

Thesis Worker at Volvo Cars
Welcome to explore the world of Volvo Cars by writing your thesis with us! As a thesis worker in our organization you are supported by a supervisor who follows you during your project. Through your thesis work you will be able to contribute to our company purpose – providing freedom to move in a safe, sustainable and personal way – from day one! 

About this opportunity - Background

To build the driving automation features of the future, it is important to combine multiple sensor observations. However, it is not always trivial what is the most effective approach to combine different modalities. Therefore, the purpose of this thesis project is to explore the combined usage of different sensor observations (e.g., video, time-series data). The scope of the thesis can be adjusted to suit your master’s program and interests, and you will get to formulate the final research questions in cooperation with us. 

Scope of the thesis work

To motivate the choices of your research, we will start with a literature review to understand what methods and procedures have been applied in previous studies to extend or align different modalities (e.g., Imagebind or XFusion). After that, we will refine together what kind of experiments could be interesting on two open-source naturalistic driving datasets. An example approach could be to first evaluate pre-trained foundation models, followed by evaluating (self-supervised) modeling approaches from scratch, and lastly evaluating how to combine the various modalities most effectively for multiple downstream tasks. 

What you'll bring

You are currently enrolled as a student in the final year of a master’s degree in computer science, artificial intelligence, data science, or a related quantitative field. You must be able to program in Python (PyTorch and Transformers), and experience in machine learning is desired. As you will become part of the analytics and artificial intelligence team within the research and development department of Volvo Cars, we expect you to work a few days per week at our office in Gothenburg. 

Duration

The thesis work will start in January 2026 for 20 weeks, conducted by a pair of two students, and in agreement with your university supervisor, the project should be 30 ECTS (academic credits).  


Volvo Cars. For Life.
For nearly a century, Volvo Cars has empowered people to move freely in a personal, sustainable and safe way. Today, we are driving bold advancements in electrification, sustainability and automotive safety. To realise our ambitious vision, we are seeking innovative minds who are ready to tackle the challenges of tomorrow – today.

At Volvo Cars, we believe extraordinary things are achieved by ordinary people with a passion for making a difference. If you’re inspired by the opportunity to help redefine the future of mobility, we invite you to be part of our journey.

Ready to take the next step?
Selection will be ongoing during the application period, so do not hesitate to send in your application. Please attach your resume, a motivation letter where you briefly describe how you would approach this project (max ½ a page), and credentials. You apply individually, but if you have a preferred partner to collaborate with, you can let us know so we can consider that.  
 
Apply as soon as possible but no later than the 31st of October. Please note that applications via email will not be accepted.  For questions about the thesis work please contact Koen Vellenga at koen.vellenga@volvocars.com.  

As part of the recruitment process, the final candidates might undergo a background check.

Welcome with your application!

Gothenburg, SE, 40531

Job requisition ID:  77768

Apply now »