Design and Implementation of Multifunction Quadcopter Intelligent System Using Machine Learning Techniques

Abstract

Natural disasters, Corona virus disease (COVID-19) and landmines all represent serious challenging to a large number of countries all over the world, and the classical systems used to face them have many limitations like higher cost, inefficient in many cases and others. Today quadcopters are the trend technologies used for these threatening to help in fight them, but stabilizing quadcopters under complex environment including system uncertainties, unknown noise and/or disturbance is so challenging and require a lot of experience with remote control to do tasks perfectly. This project aimed at designing a quadcopter system, stabilize it as good as possible using fuzzy and adaptive fuzzy control as examples of intelligent and adaptive control methods as well as to make the flight autonomous. Also, it aimed at implementation of a quadcopter vision system by using computer vision as new and future trends in quadcopters technology along with using mobile phone for live video streams, all these to make a system to face and deals with these threating effectively and with the lowest cost possible, and to illustrate examples of detecting and classifying fires/smokes, classify people as wearing face masks or not, detecting and recognition of landmines, and detecting objects such as people and animals in floods are chosen. The overall model is designed and implemented and the simulation is carried where results recorded and discussed, also it is observed from results that the system is achieving very good performance for both stabilization and vision.