Awards & Nominations
Team 7 has received the following awards and nominations. Way to go!
Global Nominee
CATRINGTON
High-Level Project Summary
The Carrington Event was the most intense geomagnetic storm in recorded history. It created strong auroral displays, sparking and even fires in multiple telegraph stationsIf such event is faced by earth today it will take a month to recover from blackouts and technological damage that will cause fall of economy and life lossCATRINGTON is a mobile application that is served by a ML model and server. ML model predicts the happening of next Carrington event and server in response send the alerts priorly through app and IOT device targeting all sectors of population(rural , urban). App also provides interactive 3D models to educate people about the Carrington events and precautionary measures.
Link to Final Project
Link to Project "Demo"
Detailed Project Description
WHAT EXACTY DOES IT DO :
CATRINGTON is a mobile application that is being served by a Machine Learning algorithm and a backend server. The application provides prior alerts in form of In-App notifications, SMS and E-Mails. To ensure the delivery of alerts a High Pitch sound is played from the device after 5 minutes if notification is not being seen by users and will play until users click on the notification. App is capable to work without Internet facilities. The app have feature to provide on ground information to escape agencies and provides precautions to save lives in the form of interactive 3-D models.
HOW DOES IT WORK:
- ML model receives data from the satellite and Generates Solar Wind Ion Density, Thermal Speed, Velocity as output.
- Output is processed, if found dangerous signal is sent to server.
- Server then pushes out alert messages through bulk messaging, triggering Android App to produce high pitch sound to ensure delivery.
- Android app work without internet.
- To Alert remote areas radio waves will Trigger the IOT devices to produce Smoke, Sound and etc.
- Integrated 3D-models and Augmented Reality to educate students and people about the whole event.
- Web portal to know the on ground information.
BENEFITS:
1) App is fully functional without the internet connectivity that is beneficial for people living in rural areas.
2) People without a phone can also be alerted with the help of IOT devices using distress smokes and flares.
3) Interactive 3-D models can help to teach people how to take necessary precautions in case of emergency.
4) These steps can help reduce the effects of upcoming Carrington events.
Our Hopes:
We hope our solution can minimize the effects of the Carrington events and reduce the life loss due to such events to 10 percent. Our ultimate goal is to save lives.
Tech Stack :
- Prediction : Machine Learning(Tensorflow, Keras, SpacePy)
- Server : Node.js , Heroku
- Application : Android SDK, java, retrofit
- IOT : ARDUINO
- Database : MongoDb
- 3D-Animation: UNITY
- Web Portal : React.js
- Messaging : Twilio
Space Agency Data
The magnetic field dataset was used as parameters from DSCOVR: https://cdaweb.gsfc.nasa.gov/pub/data/dscovr/h0/mag/2022/
The wind ion parameters as targets were used to train the machine learning model: https://cdaweb.gsfc.nasa.gov/pub/data/wind/swe/swe_h1/2022/
Both of these datasets were used to train the machine learning model which was the centre of the problem.
Hackathon Journey
TEAM 7, under the leadership of Shivam Kundra, managed to complete the project in the given time frame.
Every member of the team contributed to making this project function and meet the requirements of the NASA Space Apps Challenge.
Two members were able to create an Android Application that will be used by the end users as a part of an early warning.
The backend hosted server was built by Shivam which handle request from Android Application and requests from Dashboard.
The 3D rendition of the model was also built so as to give a clear picture of the project and how each step will execute at the time of the Carrington Event.
The Machine Learning Model was trained and implemented to give an early insight that will be sent to the API to the user.
Every member of the team participated and gave their best to make this project a victory. The functionalities that are promised, are successfully implemented along with the documentation and video presentation.
References
pip CDF module: https://pypi.org/project/cdflib/
CDF lib for C: https://spdf.gsfc.nasa.gov/pub/software/cdf/dist/cdf38_0/
SpacePy module for reading CDFs: https://spacepy.github.io/
DTW pip module: https://pypi.org/project/dtw-python/
DTW paper: https://arxiv.org/abs/2109.03742
Tags
#SocialWelfare #software