@inproceedings{7622fe5421e842a4bb83b68d2df8abb8,
title = "IMAGE-BASED OPTIMAL POWERED DESCENT GUIDANCE VIA DEEP RECURRENT IMITATION LEARNING",
abstract = "Future missions to the Moon and Mars will require autonomous landers/rovers to perform successful landing manoeuvres. In order to accomplish this task, reliable, fast and autonomous Guidance, Navigation, and Control (GNC) algorithms are necessary. In recent years, the strong capabilities of modern hardware have allowed employing deep learning models for space applications. In this paper, we present an image-based powered descent guidance via deep learning to control the command acceleration along the three axes. In particular, a hybrid architecture, composed of a Convolutional Neural Network and a Long Short Term Memory (CNN-LSTM), is trained using, as inputs, sequences of images taken during the descent. Hence, the neural network maps the sequences of images into the values of the command acceleration. The images are generated within a simulated environment with physically based ray-tracing capabilities.",
author = "Luca Ghilardi and Andrea D{\textquoteright}ambrosio and Andrea Scorsoglio and Roberto Furfaro and Richard Linares and Fabio Curti",
note = "Publisher Copyright: {\textcopyright} 2021, Univelt Inc. All rights reserved.; AAS/AIAA Astrodynamics Specialist Conference, 2020 ; Conference date: 09-08-2020 Through 12-08-2020",
year = "2021",
language = "English (US)",
isbn = "9780877036753",
series = "Advances in the Astronautical Sciences",
publisher = "Univelt Inc.",
pages = "2691--2706",
editor = "Wilson, {Roby S.} and Jinjun Shan and Howell, {Kathleen C.} and Hoots, {Felix R.}",
booktitle = "ASTRODYNAMICS 2020",
}