Constrained Sensing and Reliable State Estimation with Shallow Recurrent Decoders on a TRIGA Mark II Reactor
The episode discusses a paper detailing SHRED, an architecture for reliable state estimation in complex systems like a TRIGA Mark II reactor. The method uses time series analysis and SVD compression to map sparse sensor data to the full system state. Key findings show that the model performs accurately even when sensors are poorly placed, allowing it to learn from real-world data and correct its predictions.