Real Time Victim Detection in Smoky Environments with Mobile Robot and Multi-sensor Unit Using Deep Learning

ORCID

https://orcid.org/0000-0002-4237-6327

Document Type

Book Part

Disciplines

Electrical and Computer Engineering | Engineering | Mechanical Engineering

Publication Details

Robot Intelligence Technology and Applications 7. RiTA 2022. Lecture Notes in Networks and Systems, vol 642. © 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG.

Abstract

Victim detection in smoky indoor environments during search and rescue missions is still challenging and a critical situation. This situation is because firefighters are on the one hand exposed to unstable building structures and on the other hand their cognitive fatigue, due to long search missions, reduces the efficient victim detection in these hazardous environments. In this paper, an approach to detect a victim in real time with an optical and low resolution thermal camera assisting firefighters in their missions is presented. Thereby, the multi-sensor unit is mounted on a remote-controlled mobile robot with a trained victim detector using deep learning and display the detection in real time to an operator outside the scene. Experiments show that this approach enables an efficient detection in smoky indoor environments. The victim detection model achieves an average detection rate above 75% in real time with a low resolution thermal camera in dense smoke.

This document is currently not available here.

Share

COinS