Trash Team

https://drive.google.com/file/d/1z9DnnBVGLXVMDHSswzZX9Ap4qxwU2lua/view?usp=sharing

https://gitlab.com/kristapsb52/nasaspaceapps2021/-/tree/master

What exactly does it do? How does it work?

As the challenge requires, we have developed a non-commercial open-source software ( meaning it’s accessible for use and modification by anyone) that performs the visualization and calculation.

Our solution utilizes the Two-line element format to determine the orbit of every tracked object we could find from NASA provided sources and some others such as Space Track and Celestrak, displaying it in 3D space using Unreal Engine 4. 

To make us different from other tools in the same vein, we have in addition performed collision visualizations between various types of our modeled objects and also, using statistics, predicted where the small untracked debris might be located. (See example here)

Using accurate models of satellites, we are able to determine the weak points in which the object may fracture, meaning it would be possible to estimate the amount of new trash added and their movement vectors.

What benefits does it have?

• Our demo version has collected ~22 000 tracked space objects.
• Unique 3D models for each object type.
• In addition, performing object destructive collision upon contact in real-time and future.
• Generation of previously estimated small, untracked space debris using statistics.
• We are using advanced physics engine for visualization which performs better than web tools, offers more functionality and different ways to integrate the program into future projects.

What do we hope to achieve? Future possibilities.

The program would be useful primarily for scientists and space machinery developers especially for its ability to predict the potential risks and likelihood of a collision, but it serves an educating purpose as well for students and the general public.

We have big plans for the future of this project since there is a lot of potential. 

In case we get to collaborate with some of the agencies that create satellites, we would be able to access their satellite models and determine the potential breakage points, which would allow us to model fractures better. We experimented with stress testing already for our models, here’s our replica of Venta 1 getting hit by a couple of pounds of force and getting damaged.

Collision modeling in space is very complicated, as the result is dependent on many factors such as object velocity, rotation, mass, weak points and others. Getting the trajectory of each produced fraction that is true to a real-life event takes a more in-depth mathematical model.

Since the tiny rubble we have generated around the planet is a rough estimate of where it could be located, a machine learning algorithm could be used to better distribute the objects. We could also use machine learning to determine the dangerous areas in which collisions are more likely to happen in the future.

Used tools, coding languages, hardware, or software

Software: Unreal Engine 4, Blender, Fusion 360, Solid Works, Canva.com, Inkscape.

Coding languages: C++, Python - Pandas, Numpy, Skyfield, ephem.

Hardware: Intel i9 - 10 980XE CPU@3GHz, 64GB DD4 memory, 2x RTX Quadro 4000, SteamVR set (for testing).

Data used:

Two line elements - Celestrak [1], Space-track [2], stuffin.space [3] keeptrack.space [4]

Crossreferenced calculated positions with stuffin.space and celestrak.com object position.

Informative:

ESA (statistics for space debris) [5]

NASA (Orbital debris) [6]

NASA (Orbits catalog) [7]

Our Space Apps experience was wonderful and interesting.

We learned a lot of things about ourselves, teammates and how to work in a specific team, because each of us could do one thing at a time that he/she is good at and even perform tasks that they had never done before or had basic knowledge in the field.

We learned that computer programming takes a lot of time and it needs knowledge and occasionally it's important to rethink our ideas that didn’t work the first time. Also we gathered information and skills from each other, which inspired us to learn new skills, for instance, Unreal Engine App or even Blender App. Two of our teammates learned what really a hackathon is, as it was their first time in such an event.

Our approach was simple on paper, but it took hard work and a lot of time to develop a project with many programming failures in between and a small amount of sleep. We looked at all the challenges, made a chart with some of them that we could accomplish, measured each challenge in the difficulty of scale 1 to 5 and wrote the first ideas for each challenge and then chose one which we agreed upon with each other. 

To resolve setbacks and challenges, we pushed and helped each other, because we understood that our idea and the project itself was great, that we can win, and we were even doing it after the final pitch, just because we wanted to keep working and fix the problems and mistakes.

We would like to thank Kirilo Vasiļiskovs and especially Linda Gulbe for mentoring and telling to us nice words just to keep moving forward and of course we would like to say thank you to each other for being amazing in this team, helpful and inspiring for new ideas and projects.

References:

[1] https://celestrak.com/NORAD/elements/

[2] https://www.space-track.org/

[3] http://www.stuffin.space

[4] https://www.esa.int/Safety_Security/Space_Debris/Space_debris_by_the_numbers

[5] https://www.keeptrack.space/tle/

[6] https://www.nasa.gov/mission_pages/station/news/orbital_debris.html

[7] https://earthobservatory.nasa.gov/features/OrbitsCatalog/page1.php

#debris, #SpaceDebris, #SpaceTrash #UnrealEngine, #unreal, #KesslerSyndrome, #Collisions, #StressSimulation

This project has been submitted for consideration during the Judging process.