Inspiration
Both of us have had experiences where we have been in the car while a driver was not practicing the best driving and found it very distressing. We wanted to prevent situations like this from occurring in the future by encouraging safer driving through our website.
What it does
Tracks when users are getting sleepy to assist drivers when driving at night. It prevents drunk driving by asking users to complete a questionnaire to see if they are mentally fit to drive. Sends the driver a text message as a warning with facts about the harms of speeding if they are well above the speed limit.
How we built it
We originally wanted to have an AR simulation for the driver but found that it was not feasible to finish that in the given timeline. We used Python flask and javascript for the backend and used css and html to make the frontend. CV vision was used to add the feature to detect drowsiness and we are in the process of utilizing javascript API to track speed and speed limits using an IP address.
Challenges we ran into
Spent 6 hours trying to understand the documentation behind glitch and webGLR so we could simulate speed in our AR model but were unable to identify the component that dealt with scroll speed. Had made a 3D model with a blender so we implemented that along with a speed sign and moved on to implementing other functionalities in our website. We struggled with understanding how API's worked and which API's to use to add the necessary functionality to our website. We realized that we would want to use IP geolocation API's to get latitude and longitude coordinates from an ip address. Additionally, we then had to use those coordinates with the reverse geocoding api to get the speed limits and then had to use a JSON response to get the current speed. We had issues with a 403 error that we were unable to resolve. Integrating the code written separately was also challenging and required us to look at the relationship between Python, CSS, and HTML more closely.
Accomplishments that we're proud of
We have a better understanding of how API's work. Got the feature to count the number of blinks using CV to work.
What we learned
The best way to learn can sometimes be to fail continuously. Applications of coding languages and concepts to solve world problems.
What's next for Drive Safer
Debugging all functionalities and making sure that they are functioning. Measuring through quantitive data the impact of this website. Improving the website based on user input. Further developing the AR feature.
Log in or sign up for Devpost to join the conversation.