Master Thesis - Machine learning approach to road geometry estimation
Master Thesis & Research Opportunities - Gothenburg, Sweden
A vital part of several ADAS features is the road geometry estimation. The road geometry for ADAS systems have typically been done by utilising lane markers, target vehicles and map data and the rules for how this data should be combined have typically been to a large extent via hand-coded heuristics. However, as there typically is a lot of valuable data with proper ground truth available it would be of great interest to see how far purely data driven approaches could reach to achieve similar or possibly even superior results compared to the hand-crafted.
In this master thesis project, you will focus on:
- Large scale data with ground truth tools available
- Utilising existing but also to some extent construct cost functions to describe optimal road geometry estimation
- Utilise machine learning methods for optimal road geometry estimation
- Train machine learning algorithms with huge amount of data via computational clusters
We are looking for two students, preferably with strong skills in:
- Programming and scripting (C/C++, Matlab, Python)
- Mathematics and statistical modelling
- Machine learning and big data
- Computational clusters and parallel computing
Final application date: 2019-11-30. Please send in individual applications with CV, motivational letter and grade transcripts. If you wish to partner with someone, simply note that in your application.
Planned start: 2020-02-01, with some flexibility.
Duration: 30 ECTS
For questions regarding the project, please Contact: email@example.com