Bicycle Accident Detection System
Software as a Service
NoCategory
Thesis DesignClients
Mapua University
Techstack
PythonArduinoscikit-learnAndroid
Purpose
Cyclists who crash alone may be unable to call for help. This Mapua University prototype monitors a rider's orientation and movement to detect accidents automatically and trigger an alert for faster emergency response.
Description
An Arduino-based unit mounted on the bicycle reads a MARG orientation sensor to track the rider's tilt and motion in real time. Sensor data is classified in Python with a support vector machine trained in scikit-learn, letting the system distinguish genuine falls from normal riding events such as leaning through turns. A companion Android app completes the alerting chain, so a detected accident can reach someone who can respond.
