-
Hazard Map
-
The Actual Device
-
The post-ride display after completion
-
After ride recap
-
Start your ride button
-
List of all-time rides
-
stored info
Inspiration
Our inspiration came from Strava, an app for tracking running, and the amount of safety risks we see while biking during the summer
What it does
Our product is a hardware device known as BlindSpot that clips onto handlebars with a Raspberry Pi computer and a camera. The device identifies road risks relating to biking, including pothole detection and safe bike lane detection. This then transmits to a software iPhone app that tracks GPS, navigation, and movement during rides, gathering information from the hardware device. Using the captured images, it calculates a safety rating of the street or area for cyclists. The BlindSpot device additionally includes a button that allows the rider to photograph their ride without using their phone, supporting a safer and more focused ride.
How we built it
BlindSpot was built using multiple programming languages and hardware components. Specifically, using Swift for the iPhone App. Python for the ML and the Neural Network we utilized. We used Claude, Codex, and Antigravity for simple code assistance to help us advance conceptually on a project.
Challenges we ran into
We were challenged by difficulties working with the 3d printer, with sizing errors made. Hardware also challenged us with connectivity issues between the Raspberry Pi and the iPhone. On the software side, we faced issues with authentication due to the database losing storage and not working as we wished at the start.
Accomplishments that we're proud of
We are proud of using a neural network to be utilized for the detection of obstacles and road lines.
What we learned
We learned to be better prepared, idea-wise, with better mutual agreement, and to spend time on ideas beforehand to perfect a winning product
What's next for BlindSpot
Next, we would like to collect more data to make this a community read product and increase the efficiency of the bike attachment computation for better tracking of obstacles and safety hazards.
Log in or sign up for Devpost to join the conversation.