UAV for Disaster Management and Response

Abstract

Natural disasters such as earthquake, flood, wildfire etc. can have adverse effects on communities and the environment. They are known for the severity they cause, leaving communities without access to basic life necessities, and often taking several lives. Human caused disasters such as fire breakouts, chemical explosions and leakages also cause serious damages and lead to the loss of lives. Disaster Management is a framework using which we can deal with the impacts of a disaster, and prepare for, respond to and learn from the effects of major failures. In order to prevent, mitigate and recover from the impacts of the disaster, there arises a need to develop better disaster management models. When a disaster occurs, the primary challenge faced by the rescue and search teams is to locate victims at the earliest and provide the necessary help and aid. However, rescue teams are unable to locate or reach those in need of help in due time. It is also difficult to assess the extent of damage occurred. The proposal of this project has therefore, been laid down in the light of all these challenges. This Project implemented using a UAV (Unmanned Aerial Vehicle), also called a drone will enable the first responders to overcome many of the challenges faced while rescuing people. The drone uses an image sensor to perform aerial surveillance to assess the extent of damage and identify victims who are in dire circumstances and bring them to safety. It utilizes computer vision and machine learning algorithms (Voila-jones) for training models to identify humans and other objects using pattern recognition.