Abstract:
A road accident occurs unpredictably and involuntarily and often causes damage, injury, or death. The purpose of this project is to reduce the mortality rates after an accident by eliminating the time delay between the accident’s occurrence and the first medical emergency. This research deploys a Machine Learning based intelligent system that can detect an accident, to notify the hospital and the police station. By using an Internet of Things system, sensors are embedding in the vehicle, such as speed sensor, gravitational force sensor, sound sensor, Nodemcu-ESP8266 with Global Positioning System (GPS). The sensors collect the data, which is then processed and analyzed using artificial neural networks. Once an accident happens, the sensors detect the accident, the sensors send the data to the Nodemcu-ESP8266, Nodemcu-ESP8266 collect data from sensors and transmit this data to the data engine. The prototype experiment results used 49 input data sensors, - with 35 accidents, -27 accidents were properly detected, while other 8 accidents were not detected; the percentage detection rate was 77.14% with this experiment. It was possible to detect the occurrence of accidents, communicate with the hospitals and police station. The GPS coordinates were provided by the GPS to identify the accident location. Based on Machine Learning with sensors can help people in an accident to get first aid on time.
