AI-Powered Driver Behavior Prediction, Drunk Driving Prevention, Accident Detection, and Insurance Integration
Nandini G. Iyer, M. Arulmozhi, P Sivakumar, S. Sudharsan, Jeny Sophia S, R. K. Kavitha · 2023
According to CADD (The Community Against Drunk and Drive), 70% of road accidents in India occur due to drinking and driving. This novel design ensures that the vehicle does not start unless the rider wears the helmet and passes the alcohol test. The accident detection system placed in the bike, senses the occurrence of an accident using a vibration sensor and passes the signal to Arduino and then the controller extracts the location and time of accident and sends to the cloud and also to the emergency number. If the rider is detected with alcohol the buzzer goes on and with the help of PWM, the speed is reduced slowly and then only the bike stops. The additional feature of the proposed model is that speed, time stamp and location of the accident that is uploaded to the cloud can be accessed by the concerned insurance company to decide if the victim is eligible for insurance. A core enhancement in the proposed design is real-time driver behavior prediction using machine learning analytics. It analyzes speed, acceleration and distance to vehicles to detect risky behavior patterns, issuing timely warnings to promote safer practices and minimize accident risks.