Xiaoli Bai
Rutgers University
Calibration ensures that the measurement data are used appropriately and is critical for space missions which demand high-level accuracy, precision, and resolution over a long-term use. This project aims to enable autonomous space sensor calibration, both individually and collectively as a sensing network. We are particularly interested in applications of planetary observations and explorations when satellite constellations are used. The algorithms developed in this research are fundamental and will lead to computationally efficient, adaptive, and robust learning methods that can be used for many space applications. The proposed principled Bayesian framework to learn the models, fuse the information with uncertainties, and generate optimal decisions will advance the state of the art in uncertainty quantification, numerical analysis, and control theory.