Machine learning (ML), deep learning (DL) and reinforcement learning (RL) and other artificial intelligence (AI) methods have achieved considerable success in various application fields in the last years by learning from given data implicitly rather than relying on theoretical models based on empirical evidence. Typical applications are classification of images, speech recognition and time-series analysis.
In terms of positioning and navigation, data-driven methods have achieved remarkable results, regularly outperforming classical model-based estimation. Data-driven approaches have great potential to effectively cope with typical challenges of localization, including complex radio propagation scenarios (non-line-of-sight), device heterogeneity, optimal sensor data fusion, environmental variations and different movement types. Non-linear function approximations like deep neural networks (DNNs) can capture those relationships and model them implicitly.
In the »Precise Localization and Analytics« department, our research focus is on AI methods in a variety of applications, including indoor navigation, fitness/sports tracking, robot navigation and industrial surveillance, augmented and virtual reality.
We are looking for students interested in writing a thesis on one of a variety of topics, including:
- New data-driven methods for supervised, semi-supervised and unsupervised learning for navigation and tracking with RF, IMU and optical sensors
- Combination of data-driven methods and conventional tracking systems such as Bayesian tracking filters, including uncertainty modeling
- Modelling of complex and dynamic movements
- Performance optimization and embedded AI
- Adaptive and self-learning systems
- You are studying engineering, computer science or a related field
- You have experience with or interest in different DL methods such as DNN and generative models
- You have knowledge in Python programming using toolboxes like sklearn, PyTorch, TensorFlow and Keras
- You are interested in statistics, positioning, sensor data fusion or Bayesian tracking filters
- An open and cooperative working environment
- Collaboration in interesting and innovative projects
- Many opportunities to gain practical experience and attend seminars
- Flexible hours that allow you to balance your studies and on-the-job experience
- Sufficient opportunity to develop your interests and skills
The thesis will be assigned and carried out in accordance with the rules of your university. For this reason, please discuss the thesis with a professor who can advise you over the course of the project.
Please submit your application (PDF: including a cover letter, your CV and your latest transcripts of records) to: Nina Wörlein, via our career portal, quoting reference number
57917. We look forward to receiving your application in English or German!
Please let us know how you learned about this job opportunity.