Research

Undergraduate wins Top Presenter at American Physical Society

CAP3 Undergraduate Capstone and PPTP student Ms. Sana Elgamal was awarded Top Presenter for her presentation on "No Catch-22 for Fuzzy Dark Matter: testing substructure counts and core sizes via high resolution cosmological simulations" at the American Physical Society (APS) meeting. The virtual meeting was from April 24-26, 2023. Students presented on a breadth of physics topics to the academic physics community.

How NYUAD Hunted an Asteroid

How can humanity defend themselves from a potential asteroid strike on Earth? On September 26, NASA intentionally collided their DART — Double Asteroid Redirection Testspace — probe into the non-threatening asteroid Didymos. In this first-ever asteroid-deflection space mission, the asteroid and probe paths were calculated with impressive accuracy, and the probe successfully made impact with the center of the asteroid and deflected it out of its original orbit.

In preparation for this test, a mixed group of researchers from NYUAD and the US went to a Dubai desert on September 21 to observe the asteroid pass over a bright star visible from the emirate. Another team of researchers performed the same observation from Oman. Their goal was to get better estimates on the asteroid’s orbit, by determining the time of the occultation with a precision better than 0.1 seconds, and help NASA determine the precise location of the asteroid to assist with the spacecraft’s guidance. However, there was a low chance of successfully observing the occultation from the UAE and Oman.

A researcher from UC Santa Cruz, Paul Dalba, came to the UAE bringing along the equipment required to observe this event. Uncertain about their ability to capture the event, the UAE team used six separate sets of eVscope telescopes and video cameras simultaneously. A simultaneous effort was made by the researchers on the Oman side, who used additional five eVscope telescopes. With these digital, compact, and smart devices, the team divided themselves into different groups, covering possible/estimated asteroid paths, and anticipated the passing of Didymos 43 minutes past midnight.

All UAE located telescopes performed to standard, and all UAE telescopes had clear non-detections. Despite the clear non-detection of all telescopes, the high quality of the data speaks to the flawless execution of the difficult observation by the UAE research team, helps exclude the UAE locations as the asteroid paths, and tells us with confidence where the asteroid was not located

Five days after this desert astronomical adventure, astronomers worldwide were left stunned as Didymos and its debris brightened up in the sky after its impact with DART.

Mohamad Ali-Dib wins in the 5th Annual NYUAD Graduate and Postdoctoral Research Showcase

Mohamad Ali-Dib works as a Research Scientist at CAP3. An abstract of his winning presentation on Craters Identification with Artificial Intelligence is below.

Impact craters are the dominant morphological structures on most solar system planets and moons. Their numbers can be used as a diagnostic tool to estimate the surface age of objects, while their shapes and sizes encode valuable information on the impactors that created these craters. Finding new craters and retrieving their sizes has, however, generally been a manual process, and as such is rather extremely time-consuming. Mohamad's research focuses on using modern Artificial Intelligence techniques to detect craters in space probes imagery data. The algorithms he developed are currently being deployed to help plan the upcoming Emirates Lunar Rover's path on the Moon.

Image: Lunar surface (left) and the craters detected by the machine learning algorithm Mohamad developed (right)

Writing in the Stars

Using an AI technique (a neural network architecture called cycleGAN) Mario Pasquato turned simulated star clusters into the Center for Astro, Particle, and Planetary Physic's abbreviation CAP3. The neural net learns to translate images from one class (e.g., pictures taken in winter) to another (e.g., pictures taken in summer) by being shown examples. The result is very general and can be applied to translating a distribution of black points on a white background, as would be obtained by a dynamical simulation of a star cluster, into APOD-like pictures (on which the net was actually trained). Effectively, Mario generated these images that appear as if he is "writing with stars.”

The images points_X.png are semi-random distributions of points in the shapes of letters, and the resulting output is in the pictures named X.png. 

Download the PDF in
English or Arabic