Novel Mobile Robot Concept for Human Detection in Fire Smoke Indoor Environments using Deep Learning

Abstract

Human victim detection in smoky indoor environments during search and rescue missions is still challenging. This situation is due to the fact that fire fighters are on the one hand exposed to thermal radiation and unstable building structures. On the other hand, their cognitive fatigue, due to long search and rescue missions, reduce the efficient victim detection in such hazardous environments with limited visibility. In this paper, a novel concept of a remote-controlled and heat protected unmanned ground vehicle with victim detection system is presented, which detects missing victims in real time in smoky indoor environments and display its localization with detection rate to an operator outside the danger zone. The victim detection system, based on a trained deep learning model, is designed to address the specific properties of victims, which are for instance characterised by a lying position. The novel concept provides a framework for the design and the validation of the mobile robot with the victim detection system.