Interoperability of Sensors in Buildings for Monitoring the Search for Live Victims after Earthquakes

Abstract

This paper describes Rescue in Collapsed Building (RICB) a telemetry system for recovering victims inside a collapsed building due to an earthquake. The use of Artificial Neural Networks (ANN) with Raspberry Pi and sensors are being used in several fields, for example: to recognize pedestrians on streets, to identify human activities such as running, walking and sitting, and to collect garbage collection. RICB’s field is human support and rescue. RICB is proposed as an Artificial Intelligence tool to improve the identification of human patterns according to the readings of sensors such as movement, sound and temperature. These measurements will be collected from a Raspberry Pi Zero W device inside a transparent acrylic sphere with a Human Radiating Sensor (HRS) that is our own elaboration in this paper. Thereafter, it will be analyzed by the ANN to determine if there is a body of data in the set of transmitted data to a server. That application lets an expert monitor and manipulate RICB to read the measurements of sensors. When RICB processes this data, the user can see the probability of the victim being alive to enable communication with him/her and local rescue teams.