Project Description:

• Spaceborne Synthetic Aperture Radar (SAR) systems are powerful Earth/Planetary Observation (EPO) instruments for all condition imaging. Due to active sensing nature of these instruments such instruments are among the most power-hungry platforms. As a result, the SAR imaging, which is a computationally expensive process, has been always done on ground stations. This makes such instruments less agile and responsive to the dynamics of observation scenarios. The earth observation (EO) SAR systems work in a time-scheduled observation framework, i.e. in the contrast with continuous monitoring, due to the limited downlink bandwidth, which is needed to transfer back buffered raw-data to earth.

  The aim of this internship is to demonstrate if we can apply AI based solutions on the SAR raw data, to avoid computationally expensive SAR focusing algorithm, and evaluate the computational complexity and accuracy of “deployed machine learning” methods. While the evaluation will be done for a generic SAR framework, the C-band SAR data from Sentinel 1 will be used for the task of vessels detection and results will be compared with the ML applied to the focussed SAR images. In this study, the open datasets will be used to acquire necessary data for learning and validation. The final stage of this internship is to reduce the ML complexity to run in (close to) real-time on a reference AL enabled board equipped with Movidius Myriad2 VPU. The results of this study may be published in the form a scientific report in a conference or a journal.

• Find out more about SPIN and explore FAQs here.

Applicant Specification:

• Being independent and have good time-management.

• Minimum second year of undergraduate on a relevant field, including Engineering, Geo-science, Computer Science, Mathematics and Physics.

Minimum requirements:

• Passion and enthusiasm

• Interests in spaceborne on-board processing solutions

• Some programming skills, preferably Python.

• Some experience in machine learning and its programming tools, and synthetic aperture radar are desirable, but not necessary.

Process details:

• This internship offers a 10 weeks fixed term contract, starting from the 3rd of June 2024. The corresponding annual salary is £21,978, pro rata for this internship period (£1,831.50 per calendar month gross), Grade 03/1 of The University of Edinburgh pay scales.

• School of Engineering Induction Event to be held on 3/6/24

• 6 days holiday to be taken during the placement

This job description set outs the skills and experience we believe are needed to be able to do this job but, research also tells us women are much more likely than men to take this list of requirements as absolute and self-select out of the process. If you think you can deliver this role then we want to hear from you, regardless of the boxes you didn’t tick.