• Corpus ID: 237532741

Machine-Learned HASDM Model with Uncertainty Quantification

  title={Machine-Learned HASDM Model with Uncertainty Quantification},
  author={Richard Joseph Licata and Piyush M. Mehta and W. Kent Tobiska and Snehalata V. Huzurbazar},
The first thermospheric neutral mass density model with robust and reliable uncertainty estimates is developed based on the SET HASDM density database. This database, created by Space Environment Technologies (SET), contains 20 years of outputs from the U.S. Space Force’s High Accuracy Satellite Drag Model (HASDM), which represents the state-of-the-art for density and drag modeling. We utilize principal component analysis (PCA) for dimensionality reduction, creating the coefficients upon which… 

Figures and Tables from this paper

Uncertainty Quantification Techniques for Space Weather Modeling: Thermospheric Density Application

Two techniques to develop nonlinear ML regression models to predict thermospheric density while providing robust and reliable uncertainty estimates are proposed: Monte Carlo (MC) dropout and direct prediction of the probability distribution, both using the negative logarithm of predictive density (NLPD) loss function.

A Framework to Estimate Local Atmospheric Densities With Reduced Drag‐Coefficient Biases

An accurate estimation of upper atmospheric densities is crucial for precise orbit determination (POD), prediction of low Earth orbit satellites, and scientific studies of the Earth's atmosphere. But



Improved Neutral Density Predictions Through Machine Learning Enabled Exospheric Temperature Model

This work utilizes derived temperature data and a nonlinear machine‐learned (ML) regression model to improve upon the performance of the linear EXospheric TEMPeratures on a PoLyhedrAl gRid (EXTEMPLAR) model and provides performance improvements over its predecessor.

A methodology for reduced order modeling and calibration of the upper atmosphere

Atmospheric drag is the largest source of uncertainty in accurately predicting the orbit of satellites in low Earth orbit (LEO). Accurately predicting drag for objects that traverse LEO is critical

High Accuracy Satellite Drag Model (HASDM)

Calibrated Probabilistic Forecasting Using Ensemble Model Output Statistics and Minimum CRPS Estimation

This work proposes the use of ensemble model output statistics (EMOS), an easy-to-implement postprocessing technique that addresses both forecast bias and underdispersion and takes into account the spread-skill relationship.

Thermosphere modeling capabilities assessment: geomagnetic storms

The specification and prediction of density fluctuations in the thermosphere, especially during geomagnetic storms, is a key challenge for space weather observations and modeling. It is of great

Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning

A new theoretical framework is developed casting dropout training in deep neural networks (NNs) as approximate Bayesian inference in deep Gaussian processes, which mitigates the problem of representing uncertainty in deep learning without sacrificing either computational complexity or test accuracy.

A Quasi‐Physical Dynamic Reduced Order Model for Thermospheric Mass Density via Hermitian Space‐Dynamic Mode Decomposition

Thermospheric mass density is a major driver of satellite drag, the largest source of uncertainty in accurately predicting the orbit of satellites in low Earth orbit (LEO) pertinent to space

New modes and mechanisms of thermospheric mass density variations from GRACE accelerometers

Monitoring and understanding the upper atmosphere processes is important for orbital decay and space physics. Nowadays, Low Earth Orbit (LEO) accelerometers provide a unique opportunity to study

New density estimates derived using accelerometers on board the CHAMP and GRACE satellites

Atmospheric mass density estimates derived from accelerometers onboard satellites such as CHAllenging Minisatellite Payload (CHAMP) and Gravity Recovery and Climate Experiment (GRACE) are crucial in

Total Densities Derived from Accelerometer Data

The German satellite Challenging Minisatellite Payload (CHAMP), carrying the STAR accelerometer onboard, was launched in July 2000. The CHAMP mission profile is compatible with studies of the