An adaptive neuro-fuzzy inference system applied for the design of a firefighting robot using a photovoltaic panel

Abstract

Firefighting robots are one of the most effective tools for extinguishing early fires that cause loss of life and property, as well as permanent disability to the affected victims. These robots are used to save the lives of engineers in industrial sites with hazardous conditions. This paper aims to design and implement a mobile robot that detects and extinguishes fires using artificial intelligence and solar energy techniques. A mobile robot capable of moving is designed using a rotary motor, a flame sensor, a pump and a solar panel to supply power to the electronic components, all controlled by an Arduino Uno microcontroller and programmed using Adaptive
Neuro-Fuzzy Inference System (ANFIS) using MATLAB, where the inputs for training the network were the front, right and left flame sensors and the output is a pump (pump off, pump on). In this study, the performance of the solar panels is first tested using MATLAB and then experimentally under various weather conditions. The performance of the pump is also experimentally tested.
The robot is also tested to detect and extinguish fires. The results showed the effect of temperature change on the solar panel, as when it increases, the panel’s ability to produce decreases, as well as the effect of the reduction in solar radiation resulting from clouds and others, and the extent of its impact on the efficiency of solar panel performance, and monitoring the pump performance in terms of flow rate and height. From this, it can be observed that the designed robot can effectively detect fire sources and extinguish them with minimal errors. Thus, it can be applied in industrial settings to prevent fire damage and extinguish it when it occurs.