Research and Conceptual Design of Sensor Fusion for Object Detection in Dense Smoke Environments

Abstract

In this paper, we propose a conceptual framework for a sensor fusion system that can detect objects in a dense smoke environment with a visibility of less than 1 m. Based on the review of several articles, we determined that by using a single thermal IR camera, a single Frequency-Modulated Continuous-Wave (FMCW) radar, and multiple ultrasonic sensors simultaneously, the system can overcome the challenges of detecting objects in dense smoke. The four detailed methods proposed are as follows: First, a 3D ultrasonic sensor system that detects the 3D position of an object at a short distance and is not affected by temperature change/gradient; Second, detecting and classifying objects such as walls, stairs, and other obstacles using a thermal IR camera; Third, a 2D radial distance measurement method for a distant object using an FMCW radar; Fourth, sensor fusion for 3D position visualization of multiple objects using a thermal IR camera, 3D ultrasonic sensor system, and FMCW radar. Finally, a conceptual design is presented based on the proposed methodologies, and their theoretical usefulness is discussed. The framework is intended to motivate future research on the development of a sensor fusion system for object detection in dense smoke environments.