For years, scientists have used networks of cameras in order to observe meteors. Until now, various fireball networks have recorded approximately 800 trajectories of meteoroids significant enough to have dropped meteorites on the ground. In only 3% of cases have meteorites been recovered. In order to increase the recovery yield of the more frequent smaller falls, we need a better way in searching for freshly fallen meteorites. This summer, as an intern with NASA’s SETI program, my team used drones to go to the site of impact to survey the area for fallen meteorites. The images collected from the drone are analyzed by a meteorite recognition algorithm, which is supported by a deep learning neural network. In this session the audience will learn how asteroids are a threat to the Earth and how machine learning can be applied to problems related to Planetary Defense. They will also learn how, with the use of drones and deep learning, the fresh meteorite finds can be greatly improved.
Sravanthi Sinha is a Full Stack Software Engineer. She is really excited about the future of AI research. Sravanthi was one of the data scientists who worked on a project for the NASA Frontier Development Lab program which was hosted at SETI Institute. She is one of the 32 students selected from all around the world to be part of the first class Holberton School. She was also a Student Intern at National Resource for Network Biology (NRNB) in 2012 and is a member of NRNB Academy Alumni. Sravanthi successfully completed Google Summer Of Code (GSOC) 2013 and GSOC 2014 first as a student and then as a mentor. Published WikiPathways: capturing the full diversity of pathway knowledge (2015) in Oxford Journals. Sravanthi earned a Bachelor’s degree in Electronics & Communication Engineering from Jawaharlal Nehru Technological University in Hyderabad, India.