IEEE - Aerospace and Electronic Systems - June 2022 - 4

DOI. No. 10.1109/MAES.2022.3170740
From the Editors of the Special Issue on Current
Applications and Innovations of Artificial
Intelligence and Machine Learning in Aerospace
Sabah Mohammed , Lakehead University, Thunder Bay, ON P7B 5E1, Canada
Ruay-Shiung Chang , National Taipei University of Business, Taipei City
100, Taiwan
Carlos Ramos, Polytechnic of Porto, 4200-465 Porto, Portugal
Tai-Hoon Kim
, University of Tasmania, Hobart, TAS 7005, Australia
Sabah Mohammed
Artificial intelligence (AI) and machine learning (ML) play an increasingly important role in aerospace
applications and serve various military, commercial aviation, and space exploration sectors
to ensure safety, dependability, and customer loyalty. AI/ML contributes to provide various automated
systems used in aviation, such as fuel efficiency, smart maintenance, smart air traffic management,
pilot training, passenger identification, threat identification, remote sensing, and fully
autonomous aerial vehicles among other systems. AI/ML is concerned with algorithms and techniques
that allow systems to " learn " and " reason " based on algorithms and techniques employing
computational and statistical methods. It can significantly enhance speed, efficiency, workload,
and safety to enable the integrating of more complex technologies, such as autonomous visionbased
navigation and data ecosystems. Recently advanced data analytics provided the aviation
industry a way to respond to COVID and advise airlines on when to swap aircraft for bigger or
smaller planes and how the global health restrictions may change flight schedules. While there are
many other innovative use cases ofAI/ML in aviation and aerospace, the overarching conclusion is
that the implementation must be driven by safety. Achieving such high-quality contributions for
this Special Issue would have been impossible without the undertaking ofthe Editorial Board members
and external reviewers. We take this opportunity to thank them for their great support and
cooperation.
The first paper " Using MLSTM and multioutput convolutional LSTM algorithms for detecting
Ruay-Shiung Chang
anomalous patterns in streamed data ofunmanned aerial vehicles, " by Ahmad Also and Zouhair Dahrouj,
compares two deep learning tools to detect anomalies in the values ofthe UAV attributes. These
tools were used in a novel way to predict the potentially defective sensors of the UAV. The first
approach is the multiple LSTM networks, where each LSTM tries to detect anomalies in the consecutive
instances of one attribute. The second approach is the multioutput convolutional LSTM, which
reshapes the input data as a sequence of3D tensors to consider the relationships among the different
attributes and uses one ConvLSTM network to analyze the reshaped input. ConvLSTMhas the advantage
of preserving the convolutional architecture and the benefits of the LSTM. This combination is
suitable for analyzing multivariate temporal data. The two approaches are compared by using reallife
datasets and well-known evaluation indicators, such as detection rate, false alarm rate, precision,
and F1 score.
In " GPS spoofing detection by neural network machine learning, " Sam C. Bose proposes an
alternate antispoofing method using neural networks (NN) that can be incorporated into GPS
receivers as an add-on software module. The supervised ML technique uses readily available GPS
observables, such as pseudorange, carrier phase, Doppler shift, and carrier-to-noise density ratio to
distinguish between authentic and spoofed signals. Hardware and software details of training and
testing the NN are presented. NN architecture, relative contributions, database size, number of hidden
layers, and distribution of neurons between hidden layers are examined.
4
IEEE A&E SYSTEMS MAGAZINE
JUNE 2022

https://orcid.org/0000-0002-7639-0696 https://orcid.org/0000-0001-8013-5578 https://orcid.org/0000-0003-0117-8102

IEEE - Aerospace and Electronic Systems - June 2022

Table of Contents for the Digital Edition of IEEE - Aerospace and Electronic Systems - June 2022