Experimental Validation of Deep Learning-Based Adaptive Control for a quadcopter

• Grant number: 23/18487-5
• Support Opportunities: Scholarships abroad - Research Internship - Scientific Initiation
• Start date: May 01, 2024
• End date: August 31, 2024
• Field of knowledge: Engineering - Electrical Engineering - Industrial Electronics, Electronic Systems and Controls
• Principal Investigator: Roberto Santos Inoue
• Grantee: Gabriel Andreazi Bertho
• Supervisor: Mariusz Wzorek
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil
• Institution abroad: Linköping University (LiU), Sweden
• Associated to the scholarship: 23/05069-0 - Intelligent Adaptive Control of a Quadcopter, BP.IC

Abstract

This project targets the critical phase of experimental validation for an advancedneural network-based adaptive control system, specifically designed for UnmannedAerial Vehicles (UAVs) operating under adverse environmental conditions. This re-search topic is being pursued in the candidate's Scientific Initiation, where the controlsystem, developed using Python and TensorFlow, employs deep learning techniques toadaptively respond to external disturbances such as wind gusts and parametric un-certainties. The primary objective is to enhance the UAV's operational stability andefficiency across diverse environmental scenarios. Hence, this project aims to validatethe developed controller through the use of advanced simulation and the Vicon real-time tracking system, testing the DJI Matrice 100 quadcopter's response to adverseconditions.