Abolfazl Shirazi will defend his doctoral thesis on Thursday, March 11th
- Due to the restrictions caused by the COVID-19 pandemic, the defense will be held online and users will be able to follow it live
Abolfazl Shirazi received a Bachelor’s degree in Aerospace Engineering in 2010 and in 2012 he obtained a Master’s degree in the same field.
He joined the Basque Center for Applied Mathematics – BCAM as a PhD student in 2016 within Machine Learning research line at Data Science group in framework of La Caixa Fellowship.
His PhD thesis, Efficient Meta-heuristics for Spacecraft Trajectory Optimization, has been supervised by Prof. Josu Ceberio (UPV/EHU) and Prof. Jose Antonio Lozano (BCAM).
Due to the COVID-19 pandemic, the defense will be held online, through the platform BBCollaborate. It will take place on Thursday, March 11th at 12:00, and users will be able to follow it live using the following link: https://eu.bbcollab.com/guest/b4716b035afe4c6980035afec493440b
On behalf of all BCAM members, we would like to wish Abolfazl the best of luck in his upcoming thesis defense.
PhD thesis Title:
Efficient Meta-heuristics for Spacecraft Trajectory Optimization
Abstract:
Spacecraft trajectory optimization is one of the challenging subjects in astrodynamics. Cutting-edge technology for orbital maneuvers of the space vehicles turns this subject into a matter of interest for the aerospace community, as it has a major impact in space mission analysis and design. At the same time, rapid development of computational capabilities allows computer science areas such as artificial intelligence to get involved with real-world problems. This dissertation was an effort to make this connection and to develop novel meta-heuristics with high efficiency, concentrating on the adaptive behavior of the algorithms relative to the characteristics that describe the space orbit transfer mission. To this end, first, a comprehensive review was carried out, evaluating all state-of-art techniques and methods to deal with spacecraft trajectory optimization problems. Then, long-range and short-range space rendezvous transfers were chosen as two general target problems to be tackled in this dissertation. Consequently, two algorithms were developed for long-range and short-range space rendezvous missions respectively. In the development of these algorithms, special attention has been paid to the discovery of the features of the problems and connecting them to the parameters of the algorithms. The algorithms were mainly developed relying on probabilistic basis. Conducted experiments confirm the robustness and the efficiency of the proposed algorithms in tackling different problems. Besides the algorithms, a simulation software is also developed to provide a framework for the visualizing of space trajectories.
