Smart Vehicle Safety System with Overload Detection and ML-based Driver Monitoring

Road safety is a crucial concern for this society nowadays because accidents could arise from any number of reasons-like vehicle overloading, driver drowsiness, alcohol consumption, and environmental hazards. In turn, this research aims to target all of the stated issues via an integrated vehicle monitoring system using the latest technology. In the system, the load cell is capable of detecting vehicle overload and reducing the vehicle motor and stopping it whenever the load crosses the threshold. Driver drowsiness is detected through a CNN-based machine learning model trained on a custom dataset with 92% accuracy. This system alerts once fatigue is detected and disables the vehicle altogether in a safe manner. Alcohol detection is done in real time using a gas sensor, wherein the motor is prohibited from starting when alcohol is sensed. Accident detection is done with the help of a vibration sensor that shares the live location of the vehicle on the cloud IoT dashboard. There is also a temperature monitoring system to avoid failures due to overheating. This ensures real-time availability of data, remote monitoring, and automatic safety controls, thus minimizing human error and promoting road safety.