Assisting an SAR mission for lost people using a UAV

Abstract

Search and Rescue (SAR) missions are nowadays still being executed by on-foot person searchers or by manned aircrafts including planes and helicopters, both of which are time and cost inefficient. The main challenge is the time needed to react to take the required action. New developments in Unmanned Aerial Vehicle (UAV) technology are able to tackle such a problem with the use of drones and aerial photography. This study aims to exploit such a technology alongside shortest path and object detection algorithms to reduce both mission duration and the risk of the patient getting injured. Before proceeding with a solution for the aforementioned problem, research on UAV/drone technology and SAR missions for lost people in society was carefully studied. As a case-study for this project, the research was focused on lost individuals living with dementia. Based on this research, models and prototypes were formulated prior to development. An Android mobile application was developed to simplify the communication between the DJI (Da-Jiang Innovations) drone and the operator. This was done by making use of the DJI Mobile Software Development Kit (SDK). An object detection algorithm received the images captured by the drone to detect persons at multiple points throughout the route. A separate Android mobile application was developed to efficiently gather data on the SAR mission, including personal information and possible locations that are vital throughout such a mission. This mobile application used the Firebase Real-time Database to allow for the collection and usage of the mission information through multiple devices. Given the time constraint to search for a lost individual with dementia during such an SAR mission, two shortest path algorithms were implemented to aid the drone operator in the drone navigation from one location to another, depending on prioritisation and probability of finding the lost person.