Quaternion

TE-PINN: Quaternion-Based Orientation Estimation using Transformer-Enhanced Physics-Informed Neural Networks
Novel transformer-enhanced physics-informed neural network (TE-PINN) achieving 36.8% error reduction in quaternion-based attitude estimation from IMU data. Integrates multi-head attention mechanisms with physics-based constraints (quaternion kinematics, rigid body dynamics) for robust real-time orientation estimation in high-noise, dynamic conditions.
Arman Asgharpoor Golroudbari
PDF Code
Generalizable end-to-end deep learning frameworks for real-time attitude estimation using 6DoF inertial measurement units
End-to-end deep learning framework for real-time inertial attitude estimation using 6DoF IMU. Achieves 40% accuracy improvement over traditional methods, evaluated on 7 datasets (120+ hours, 200+ km of IMU data). Generalizable across various sampling rates and motion patterns.
Arman Asgharpoor Golroudbari, Mohammad H. Sabour
PDF Cite Code Slides DOI