The problem Accident Response System solves

Most deaths due to accidents is due to lack of a quick responce system. Our project uses a machine learning model to predict the severity of the accident and respond accordingly. If the severity parameter is more than a particular value(2) alert message is sent to the nearest control room as well as the emergency contact numbers. If severity parameter is less than 2, alert message is sent to only the nearest control room. The control room must be equiped with ambulances and necessary basic medical application. This software cuts down the time wasted by nearby people who do not to respond to the accident fearing the legal procedures. This even makes the task of the hospitals easier because the control room can alert the nearest hospital about the victim's state. We can further improve this model by alerting the nearest blood donors about the accident by asking each user of this software to mandatorily agree to the condition that he/she will also donate blood to other needy users. This will be thefuture scope of the project

Challenges we ran into

The bigest challenge was getting input to the ML model. This requires various sensors on the vehicles. So we just tried to simulate the accident by giving random inputs within a correct range of values. To alert the model that an accident has happened we will be requiring a thermostat,airbag sensors etc. Therefore we used a simple button click of which implies that the accident has happened. We used this idea becuase our concentration is on the responce system but not on the hardware aspects(sensors) of the problem. Developing the hardware will be the future scope of this project.