AUSTRALIA IN SPACE
AI, EO, and you By Shane Keating, University of New South Wales
A
rtificial Intelligence (AI) is seemingly everywhere
says Jessica Cartwright, a senior scientific programmer and
these days, disrupting industries from web search
remote sensing expert at Spire Global. Spire is a leading global
to healthcare and finance to fine arts. The space
provider of space-based data, analytics and space services,
industry is no different, and a host of innovative companies
operating a constellation of more than 100 multipurpose
are finding novel ways to exploit the AI revolution.
satellites for monitoring maritime and aviation data, soil
Broadly speaking, artificial intelligence aims to create
models by leveraging the massive amounts of data we have
that normally require “thought” --- tasks like recognising a
from these satellites,” says Cartwright.
face, understanding speech, or deciding whether to bring an
Cartwright is leading a project at Spire that uses AI to
umbrella on your morning commute. A neural network is a
measure sea ice from space. Sea ice is a crucial component
machine that mimics the brain itself, taking in information
of the climate system and improved measurements will
(“let’s look outside”) and passing it through a network of
help us better understand the impact of climate change.
artificial neurons or nodes. Machine learning is way of
But measuring sea ice is challenging, given the geographic
developing artificial intelligence by looking for patterns in
remoteness and vast scale of the Arctic and Antarctic and the
data (“It’s cloudy; it might rain.”). And deep learning is a type
costs and challenges of polar research. “If we can tell what's
of machine learning that uses many neural networks stacked
happening with the ice, we can understand not only how the
on top of each other to produce an output (“better bring a
climate is changing, but we can also understand who it might
brolly”).
affect when it does change,” Cartwright told me.
AI research started in the 1950’s, but progress surged in
Cartwright uses a technique called GNSS-Reflectometry,
the mid-2010’s with the increased availability of powerful
which uses reflected signals from Global Navigation Satellite
computers, more sophisticated algorithms, and, above all,
Systems (such as GPS in the US and Galileo in Europe) to cost-
data. Over the same period, there has been an explosion
effectively collect large quantities of information about Earth’s
in the availability of Earth Observation (EO) data --- NASA
surface. As the GNSS signal is bounced off Earth’s surface, it
estimates that their constellation of spacecraft alone
carries with it a signature of the surface that is measured by
produces 2 Gb of data every 15 seconds, with only a fraction
Spire’s satellites. Using that data to measure sea ice requires
of that amount being analysed. AI is now a crucial tool for
a model for how the signal is modified by the roughness and
making sense of the deluge of EO data from space.
conductivity of the ice. That’s where AI comes in.
“We're using AI to get fast results from our satellites,” 48 | Australia in Space Magazine
moisture and weather patterns. “AI enables us to create
machines that mimic human intelligence by performing tasks
Spire partnered with our group at the UNSW Data