Inspiration

Our project is inspired from the fact that cars currently don’t have any system in place that is similar to an airplane’s blackbox to record the final moments of what happened before a crash. So, we decided to attempt at making one, and now we have CrashBox Crash Detection System (or CDS)

What it does

The CrashBox detects a crash using one of the three sensors, accelerometer, gyro sensor or the temperature sensor and sends an SMS to 911 giving the location of the crash and what has happened in the crash: If there is a fire or if the car has flipped over, what the make and model of the vehicle are and who owns it, and underlying health conditions that the driver may have.

How we built it

This project was built using an Arduino mega and a mpu6050 microcontroller. It uses python to display an image corresponding to the direction the mpu6050 microcontroller is facing. After the python script senses a spike in deceleration, temperature or the car has flipped over, it stops and sends an SMS through twilio to the paramedics and takes the data before the crash putting it into a CSV file that gets put on the website.

Challenges we ran into

We ran into a couple of challenges, none of us were very good with HTML or Javascript so we were left wondering how to add a CSV file automatically to the website. This took us some time even leading us to work on it through the night. We had other problems with some of the python code as text messages would send automatically and populate someone's phone. Lastly we had to find good threshold values to detect the crash so that it doesn’t detect one if it hasn't happened and if a crash happened but it doesn’t get detected.

Accomplishments that we're proud of

We are proud of the fact that we were able to connect the 2 separate sections of our project, the website and the python scripts. We thought this would be a difficult task but we ended up able to accomplish it and link all the parts of our project together. We were also able to get a full model of the gyro sensor working with nice graphics to help visualize the way it works and send a text to a given phone number with all that data.

What we learned

We learned how to take data from the arduino sensor and put that data onto a csv. We also learned how to use Twilio SMS API to send SMS messages and how to use OpenCage API to track a phone based on its number. We also learned a little bit about web development to construct our website, specifically the file uploading part of the website was something none of us had experience with before.

What's next for Crash Box Crash Detection System (CDS)

In the future we hope to find a way to make the update system for the file on our website automatic rather than manual so that the website will be updated every time a new crash occurs. We also would like to implement a computational neural network to analyze the data that we get from the CrashBox and use it to determine the most common conditions that accidents occur in, such as oversteering or drunkenness, so that we may prevent further crashes in the future. We also want to add more sensors to our CrashBox so that we may gather more data on different types of crashes.

Built With

Share this project:

Updates