High-Level Project Summary
We have developed a way to identify and distinguish space debris . The project will help solve the problem of collisions between spacecraft and other space objects. The neural network will determine the position and calculate the orbital motion. Knowing the speed and orbit, you can prevent a catastrophe that will cost enormous costs and lives.
Link to Project "Demo"
Link to Final Project
Detailed Project Description
The neural network analyzes the dataset and identifies space debris on it. In accordance with this moment of the neural network, the object is whether it is space debris. We used the aicowd API, Python Libraries: Torch, Detectron, pandas, numpy, tqdm, matplotlib, open CV
Space Agency Data
https://orbitaldebris.jsc.nasa.gov/quarterly-news/
https://www.esa.int/Safety_Security/Space_Debris/Analysis_and_prediction
https://cyberleninka.ru/article/n/metod-i-algoritm-raspoznavaniya-iskusstvennyh-okolozemnyh-orbitalnyh-obektov-i-musora-dlya-obespecheniya-bezopasnosti-kosmicheskih
Hackathon Journey
The hackathon is a great place to test your knowledge and skills, both technical and social, teamwork is also crucial. It was thanks to the task that we were able to make sure of the well-coordinated work of our team. We were inspired by the interest in developing AI in space to prevent disasters such as the 2009 Iridium 33 collision with space debris.
Thanks to the hackathon, we have practical skills in AI / ML
References
https://pytorch.org/
https://pandas.pydata.org/
https://numpy.org/
https://matplotlib.org/
https://www.tensorflow.org/
https://github.com/facebookresearch/detectron2
Tags
#AI #ML #debris #space