Master Thesis - Towards Robust Deep Learning SD map localization
Join us in advancing autonomous driving by leveraging deep learning and sensor fusion to enhance vehicle localization.
Why This Matters
Being able to localize the vehicle within a map is a crucial task for bringing robustness to an automated driving assistance system and to design safe autonomous driving as it enables the car to extract information from the map that is beyond its sensing range. This could be for example the number of drivable lanes, current and future speed limits, upcoming splits, upcoming curvature of the road, etc... Such information is really important for planning and controlling the vehicle in a smooth manner.
SD (Standard Definition) map localization has been studied and developed for some time and is widely used in satellite navigation systems, but due to the limited information being used, it is hard to model every situation properly and to handle them correctly. And while an incorrect localization in a satellite navigation device might cause a slight detour, it can have severe consequences if the information is used to steer the vehicle.
With deep learning it is now possible to process a wider range of measurement sources in a more unified way in order to form a better understanding of the environment surrounding the vehicle.
Building upon a previously developed deep learning SD localization framework we would like to investigate the use of new sensor input which would bring semantic information and assess how such information would improve the localization performance.
Your Role in the Project
In this master thesis project, you will:
- Evaluate and analyze the limitations of the current deep learning SD localization framework.
- Identify and build a dataset featuring challenging driving scenarios.
- Investigate the use of additional sensor inputs, such as:
- Lane marking types
- Road edges
- Traffic signs
- Full-frame images
- Assess how much these new inputs improve the localization performance.
What’s in It for You?
- Gain hands-on experience working with Graph Neural Networks and real-world autonomous driving challenges.
- The opportunity to contribute to patent applications and/or academic publications.
- Collaborate with industry experts and directly impact cutting-edge technologies.
- Be part of a supportive and inclusive team, where diverse perspectives are valued.
- Expand your professional network in a global and innovative company.
- Flexible start dates to accommodate your schedule.
What We’re Looking For
We’re looking for two students to work as a team, ideally with strong knowledge in:
- Machine learning and deep learning (experience with Graph Neural Networks is a plus)
- Statistics and sensor fusion
- Python programming, including ML/DL libraries such as Pandas, PyTorch, and Scikit-learn
How to Apply & Important Details
Please submit your individual applications, including your CV, motivational letter, and grade transcripts. We encourage you to apply as a pair and include the name of your partner in your application.
- Planned start: January 2025 (flexible).
- Final application date: November 15, 2024 (applications are reviewed on a rolling basis).
- Duration: 30 ECTS
For more information or questions, feel free to contact us at:
More about Zenseact
Our software makes a difference.
Using AI-based technology to create the ultimate driver support, we’re fighting to end car accidents and make roads safe for everyone. Around 1,4 million people die in traffic yearly while approximately 50 million people get injured. Many get disabled as a result of their injury. We can do better.
One purpose, one product.
We’re a software company dedicated to revolutionizing car safety. By designing the complete software stack for autonomous driving and advanced driver-assistance systems, we’re fighting to end car accidents and make roads safe for everyone. Zenseact was founded by Volvo Cars, and the teams are based in Gothenburg, Sweden, and Shanghai, China.When we aim for zero accidents faster, we strive to speed up the transition to safe automation. This is essentially achieved by making cars updatable – like a computer or a phone. With regular software updates, a vehicle can be made safer long after its production. By accelerating improvement loops, shortening development cycles, and deploying high-capacity software quickly, we can make cars safer, faster.
Culture with people at heart
To achieve our mission of saving lives and ending traffic accidents is to go where nobody has before. It requires us to venture into the unknown, pioneering new technology and pushing the frontier of autonomous driving. While there’s no denying our determination and expertise, we must stand united to succeed. By fostering a culture of support and enablement – a place of psychological safety where all of us can thrive – everything else will follow. We call this a people-at-heart culture. This culture means caring. It means the company cares about me, and we care about one another. It means sharing, so we give each other energy and have fun together. Our culture is also about belonging. It’s important to feel at home and that we can be ourselves at work. Finally, a people-at-heart culture means well-being. So, we enjoy the flexibility needed to be and do our best – at work and in life.
Zenseact works proactively to create a culture of diversity and inclusion, where individual differences are appreciated and respected. To drive innovation we see diversity as an asset, which means we value and respect differences in gender, race, ethnicity, religion or other belief, disability, sexual orientation or age etc.
Interviews are held on a continuous basis, so we highly recommend that you submit your application at your earliest convenience.
- Competence area
- Opportunities for Students, Graduates & Innovators
- Locations
- Gothenburg, Sweden
- Remote status
- Hybrid
Gothenburg, Sweden
About Zenseact
One purpose, one product
We are a software company focused on transforming car safety. By developing a complete software stack for autonomous driving and advanced driver-assistance systems, we aim to eliminate car accidents and make roads safer for all. Founded by Volvo Cars, Zenseact operates globally, with teams in Gothenburg and Lund, Sweden; Munich, Germany; and Shanghai, China.
Master Thesis - Towards Robust Deep Learning SD map localization
Join us in advancing autonomous driving by leveraging deep learning and sensor fusion to enhance vehicle localization.
Loading application form
Already working at Zenseact?
Let’s recruit together and find your next colleague.