Deep Learning Method for Martian Atmosphere Reconstruction
- 1 October 2021
- journal article
- research article
- Published by American Institute of Aeronautics and Astronautics (AIAA) in Journal of Aerospace Information Systems
- Vol. 18 (10), 728-738
- https://doi.org/10.2514/1.i010922
Abstract
The reconstruction of atmospheric properties encountered during Mars entry trajectories is a crucial element of postflight mission analysis. This paper proposes a deep learning architecture using a long short-term memory (LSTM) network for the reconstruction of Martian density and wind profiles from inertial measurements and guidance commands. The LSTM is trained on a large set of Mars entry trajectories controlled through the fully numerical predictor-corrector entry guidance (FNPEG) algorithm, with density and wind from the Mars Global Reference Atmospheric Model (GRAM) 2010. The training of the network is examined, ensuring that the LSTM generalizes well to samples not present in the training set, and the performance of the network is assessed on a separate training set. The errors of the reconstructed density and wind profiles are, respectively, within 0.54 and 1.9%. Larger wind errors take place at high altitudes due to the decreased sensitivity of the trajectory in regions of low dynamic pressure. The LSTM architecture reliably reproduces the atmospheric density and wind encountered during descent.Funding Information
- National Aeronautics and Space Administration (80NSSC18K1510)
This publication has 23 references indexed in Scilit:
- Reconstruction of Atmospheric Properties from Mars Science Laboratory Entry, Descent, and LandingJournal of Spacecraft and Rockets, 2014
- Neural-Network-Inspired Machine Learning for Autonomous Lunar TargetingJournal of Aerospace Information Systems, 2014
- Entry Guidance: A Unified MethodJournal of Guidance, Control, and Dynamics, 2014
- The Hypersonic Inflatable Aerodynamic Decelerator (HIAD) Mission Applications StudyPublished by American Institute of Aeronautics and Astronautics (AIAA) ,2013
- Atmospheric Modeling Using Accelerometer Data During Mars Reconnaissance Orbiter Aerobraking OperationsJournal of Spacecraft and Rockets, 2008
- Recurrent Neural Networks Are Universal ApproximatorsLecture Notes in Computer Science, 2006
- Selection of the Mars Exploration Rover landing sitesJournal of Geophysical Research, 2003
- Mars Pathfinder Entry, Descent, and Landing ReconstructionJournal of Spacecraft and Rockets, 1999
- Long Short-Term MemoryNeural Computation, 1997
- Multilayer feedforward networks are universal approximatorsNeural Networks, 1989