كنترل كوادروتور بدون سرنشين به كمك روش فرمان پذيري بصري به منظور بهبود پايداري و عملكرد آن

چكيده انگليسي :

Visual Servoing Control of the Unmanned Quadrotor in Order to Improve its Stability and Performance Shiva Bagherzadeh shiva bagherzadeh@me iut ac ir January 20th 2013 Department of Mechanical Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree M Sc Language Farsi Supervisor Saeed Behbahani behbahani@cc iut ac ir Abstract Since the ancient times the dream of flight and making flying objects has attracted human s special attention His attempts to realize this dream have led to great progresses in the field of aeronautics Among different aerial vehicles that have been designed by humans unmanned aerial vehicles that have appeared in the early 20th century have been increasingly considered these days due to their capability of performing different missions without exposing human s life in danger Quadrotor helicopter is one of the unmanned aerial vehicles that have found an especial place in human s life duo to its high maneuverability to perform military and civilian missions Quadrotor is an X form platform with two pairs of counter rotating rotors locating at the corners of the platform All of the maneuver movements of the quadrotor such as movements and rotations along and around the axes are caused by changing the speed of the rotors Quadrotor is an under actuated aerial vehicle that has six degrees of freedom and four control inputs Quadrotor is an unstable vehicle that requires an efficient control system in order to become stable and to be able to perform assigned missions In order to design such a control system the information of the quadrotor position and the amounts of the quadrotor linear and angular velocities must be accessible As well duo to providing the possibility of flying in the unknown environments the information of the environment must be available for the control system One means of acquiring this information is by using cameras to take images of the environment and process these images to obtain the required information This method is called visual servoing method Image processing and sending the extracted information to the control system may cause some problems such as time delays and information losses Thus the designed control system must be able to compensate these problems and ensure the expected performance of the system The main goal of the present work is to design a control system that is able to remove the side effects of time delays happened during the visual servoing method besides stabilizing and controlling the flight of the quadrotor The model predictive control method is an optimal control method that can realize the aims of this work due to its particular characteristics In this work the equations of motions of the quadrotor are obtained based on Newton Euler method The equations show that the translational movements of the quadrotor are depended on the rotational movements while the opposite is not true Therefore the control system is consisted of a translational and a rotational controller which are responsible for controlling the translational and rotational movements of the quadrotor respectively With considering undetermined computational and visual servoing time delays and by using model predictive control method the translational and rotational controllers are designed Finally the quadrotor plant and its control system are simulated using Matlab and Simulink and the performance of the whole system is evaluated The results demonstrate that the proposed control system can efficiently control the flight of the quadrotor in the presence of undetermined time delays and physical constraints of the system Keywords Unmanned aerial vehicles Quadrotor Time delays and information losses Model predictive Control Physical constraints Successive linearization and discretization

باقرزاده، شيوا

كنترل كوادروتور بدون سرنشين به كمك روش فرمان پذيري بصري به منظور بهبود پايداري و عملكرد آن

وسايل هوايي بدون سرنشين , تاخير زماني و از بين رفتن اطلاعات , كنترل كننده ي پيش بيني كننده ي مبتني بر مدل , قيود فيزيكي حاكم بر مسئله , خطي سازي و گسسته سازي گام به گام