Associate ProfessorShannon Ryan

As an engineer, what do you do when the system you are trying to design/characterise is so complex that conventional methods won't work? What about if experiments are so difficult and/or expensive to conduct that you'll never be able to sufficiently explore the effects of individual variables?

These questions have directed my research towards applied machine learning (ML) and artificial intelligence (AI) for solving complex engineering problems. My early career research into the engineering of physical protection systems took me from Germany to the NASA Johnson Space Centre, and later the US Army Research Laboratory, Fraunhofer EMI, and the Defence Science and Technology Group in Australia. Applied AI/ML can be transformative in the domains of complex engineering research - either protecting spacecraft against debris impact or protecting military vehicles against ballistic and hypervelocity threats (as well as many others).

In addition to actively performing research and publishing in these areas I am trying to broaden the understanding of, and appreciation for, applied AI/ML in hard engineering disciplines where there can be well-justified resistance and scepticism. A recent objective of mine has been to collate and release engineering datasets that can be used to train and evaluate ML systems for impact protection, including the world’s largest open-source database of hypervelocity impact data, collected from over 15 years of Whipple shield (a type of debris shielding used on spacecraft) impact testing.

I am an internationally recognised expert in spacecraft physical protection against space debris impacts, terminal ballistics, and armour systems. I leverage my international network and diverse research experience to help formulate and supervise under- and post-graduate student projects.