Modélisation, commande et planification de trajectoire d'un drone quadrirotor

Abstract

décrivant le mouvement du système afin de pouvoir établir un modèle dynamique en utilisant le formalisme de Newton-Euler, ensuite nous présentons une architecture de contrôle basée sur la régulation PID pour le contrôle des différents mouvements. Finalement, nous avons dressé un état de l’art des principales méthodes de planification d’une trajectoire en expliquant brièvement leur principe ainsi que leurs avantages et leurs inconvénients. Une étude comparative sur les algorithmes de planification d’une trajectoire selon plusieurs critères, nous a permis de choisir le meilleur algorithme pour la modélisation, la conception et réalisation d’un Drone Quadricoptère.***

Drones, also called UAVs (Unmanned Aerial Vehicles), are unmanned aircraft. They are adopted by military and civilian organizations to perform difficult tasks in very hostile environments, without any risk to humans. In our study we are interested in the quadricopter drone. Advances in control, detection and calculation technologies have enabled these vehicles to perform missions independent of direct operator control. The trajectory can be planned in advance of the mission and the drone can then execute it automatically. Planning this trajectory in an efficient manner involves determining solutions to achieve a certain objective with precision, such as avoiding obstacles or finding the shortest path in terms of travel time, or even minimization the cost of the mission. This thesis deals with the modeling, the control and the planning of the trajectory of the quadcopter drone, it is in this sense we are interested in two key stages allowing to realize an autonomous navigation : the planning of the trajectories, and the development of the laws of command which will monitor the trajectory in real time. We started with the modeling which consists in gathering all the methods and techniques allowing to have a mathematical presentation describing the movement of the system in order to be able to establish a dynamic model using the Newton- Euler formalism, then we present a control architecture based on PID controller for the control of the different movements. Finally, a state of the art of the main trajectory planning methods was presented, briefly explaining their principle and their advantages and disadvantages. A comparative study of trajectory planning algorithms according to several criteria allowed us to choose the best algorithm for the modeling, design and realization of a Quadricopter Drone.