Thesis Project: 30 hp Learning based HD map Change Detection and Update

We hope you are a young talent on the look-out for an exciting Master thesis to push the boundaries of autonomous driving! Join us for a thesis opportunity in revolutionizing HD-map change detection and updates with cutting-edge learning-based methods!

Background
Have you ever wondered what is helping autonomous vehicles to drive safely in challenging traffic situations? The answer is High Definition (HD) maps: detailed, digital maps that help autonomous vehicles navigate the world with precision. But the world we all live in constantly changes, and so do our roads! In this thesis, you will leverage latest advances in machine learning techniques to quickly and accurately identify these changes, ensuring that our autonomous vehicles can always operate with the most up-to-date HD-map information.

During your time at Scania, you’ll dive into the forefront of technology, draw inspiration and apply your innovative ideas to state-of-the-art change detection methods. You’ll tackle intricate problems and refine techniques to enhance the accuracy and robustness of HD-map maintenance. This is a unique opportunity to engage with a leading company in the automotive industry and gain valuable experience in autonomous vehicle technology.

If this sounds like an exciting challenge and you’re eager to dive into groundbreaking research while learning new skills, we’d love to hear from you. Join us and be a key player in helping autonomous vehicles find their way!

Description of the assignment
The focus will be to extend an existing end-to-end HD-map Change Detection and Update framework with novel machine learning techniques to increase robustness and accuracy. You will then test the improved pipeline on a large-scale dataset of both real-world and synthetically generated HD-map changes and evaluate the performance gain of the system.

Education/line/direction
Applied mathematics, computer science, machine learning, robotics, engineering physics, electrical engineering, or a related technical science or engineering subject. Experience with deep learning frameworks and architectures, and strong coding skills in python.

Number of students: 1
Start date for the Thesis project: 2025-01
Estimated timescale: 6 months

Contact person and supervisors
Natalie Richardson, EEARD, natalie.richardson@scania.com
Lena Wild, lena.wild@scania.com
Rafael Valencia Carreño, rafael.valencia.carreno@scania.com

Application
Your application should contain CV, motivation letter and copies of grades.

A background check might be conducted for this position. We are conducting interviews continuously and may close the recruitment earlier than the date specified.
 

Thesis worker