For most people, backyard space exploration is about peering up at the skies through a telescope, but for Abigail Azari, it means using artificial intelligence to study the planets in Earth\u2019s solar system \u2014 and keep a watchful eye on what the heavens can throw at us.<\/p>\n
Working to improve the application of AI in interpreting huge amounts of data gathered by satellites, Azari<\/a>, an assistant professor in the University of Alberta\u2019s Faculty of Science<\/a> and Faculty of Engineering<\/a>, wants to help answer important scientific questions about what can affect our planetary neighbourhood, like space weather.<\/p>\n Azari is one of 25 new faculty members focused on advancing AI research, announced last week<\/a> at the Upper Bound conference<\/a> hosted by the Alberta Machine Intelligence Institute (Amii)<\/a>. Thanks to a $30-million investment from the Canadian Institute for Advanced Research<\/a> through Amii, the goal is to bring together some of the world\u2019s best minds in artificial intelligence from a wide range of fields including engineering, environmental science, education and health.\u00a0<\/p>\n \u201cApplying machine learning to the observations spacecraft return can help us better understand planetary systems and extend our physical theories,\u201d she says.\u00a0<\/p>\n Using a \u201cphysics-informed\u201d machine learning approach can help researchers narrow down and identify the most relevant observations and pick the best conditions that improve our understanding of questions they\u2019re exploring, Azari explains. \u201cIf a model suggests a density or temperature that is physically impossible, for example, we can use physics-informed AI to make a decision to discard it.<\/p>\n \u201cIt can help weigh the possible solutions, to say one is more probable than another.\u201d<\/p>\n Having a more comprehensive understanding of space environments and their dynamics can help more accurately predict the timing of space weather phenomena that can affect the Earth, including auroras<\/a>, as well as more disruptive events like radiation storms and geomagnetic storms.<\/p>\n \u201cOn a global scale, these types of events can have large repercussions on infrastructure. They can disrupt satellite communications or impact electrical grids to cause power outages,\u201d Azari says, noting events like a mass blackout in Quebec in 1989<\/a> that was sparked by a series of coronal mass ejections<\/a>.<\/p>\n \u201cYou want to make sure you\u2019re ready for events like these when they arrive, through measures like space weather alerts that can help warn power networks or delay satellite operations.\u201d<\/p>\n The work of Azari and her team<\/a> is also important to exploring other planetary environments. For example, she leads machine learning research relevant to studying the space environment of Mars. Her team is looking at how magnetic fields play a role in this history and evolution of the planetary space environment, including atmospheric retention or loss.\u00a0<\/p>\n Working with several other U of A researchers in areas such as computing science<\/a> and earth and atmospheric sciences<\/a>, Azari is also applying her research to help prioritize what data to collect from space. \u201cIf you’re sending a satellite to another planet, you can’t get all the data back that you\u2019d ideally like to have, for a number of reasons. So we\u2019re working on how to choose what observations to take in an automated way that still uses scientific reasoning.\u201d\u00a0<\/p>\n Through her research, Azari ultimately hopes to enable a new approach scientists can use in their discovery work.<\/p>\n \u201cI\u2019d like to improve the use of machine learning and statistics for answering science questions. Machine learning can help us answer questions fundamental in physics that we haven\u2019t been able to before this point, especially if we combine these methods with our physical understanding of the worlds around us.\u201d<\/p>\n","protected":false},"excerpt":{"rendered":"For most people, backyard space exploration is about peering up at the skies through a telescope, but for…\n","protected":false},"author":2,"featured_media":51412,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[30458,24,30459,25,30460,372,28165,50,30461,160,30462,30463],"class_list":["post-51411","post","type-post","status-publish","format-standard","has-post-thumbnail","category-ai","tag-abigail-azari","tag-ai","tag-amii","tag-artificial-intelligence","tag-canada-cifar-ai-chairs","tag-engineering","tag-machine-intelligence","tag-machine-learning","tag-planetary-science","tag-science","tag-space-environment","tag-space-physics"],"_links":{"self":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/51411","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/comments?post=51411"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/51411\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media\/51412"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media?parent=51411"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/categories?post=51411"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/tags?post=51411"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}