Selfsupervised IMU-Denoising for Visual-Inertial SLAM
Selfsupervised Learning, IMU denoising
Beschreibung
In Visual-Inertial SLAM (Simultaneous Localization and Mapping), inertial measurement units (IMUs) are crucial for estimating motion. However, IMU data often contains accumulative noise, which degrades SLAM performance. Self-supervised machine learning techniques can automatically denoise IMU data without requiring labeled datasets. By leveraging self-supervised training, the project aim to explore neural networks distinguish useful IMU signal patterns from noise, improving the accuracy of motion estimation and robustness of Visual-Inertial SLAM systems.
Voraussetzungen
- Knowledge in Machine Learning and Transformer.
- Motivation to learn and research.
- Good coding skills in C++ and Python.
- Project experience in Machine Learning (PyTorch) is a plus.
Kontakt
xin.su@tum.de