Visual Servoing Control Of Serial Aerial Manipulator

UAVs have a wider active area,faster maneuverability and agile response than ground robots.UAVs will be more valuable for applications requiring less time and faster task response.In recent years,the multi-rotor UAV has been widely used in aerial photography due to its simple control and flexible motion.Its main duty is to assist people in monitoring and surveying hard-to-reach areas.However,simply by installing a camera on a multi-rotor UAV for aerial photography,the lack of active handling of the environment greatly limits the value of its application and does not provide the advantages and features of a multi-rotor UAV.In order to improve the active interaction between the multi-rotor aircraft and the environment,this thesis proposes a method of carrying a multi-degree-of-freedom serial robot below the aircraft,using the relationship between the motion of the robot arm and the environment.First of all,this thesis designs six degrees of freedom serial robot and kinematics modeling.Positive and inverse kinematics and differential kinematics models are obtained.The model provides the theoretical basis for the subsequent flight manipulator control.Secondly,in order to improve the accuracy and reliability of the manipulator for the operation of the environment,this thesis presents a visual servo-based method of motion control of the flight manipulator.By comparing the advantages and disadvantages of position-based and image-based two basic visual servoing schemes,this thesis uses a two-dimensional augmented position-based hybrid visual servoing scheme that can be used in position-based visual servoing control scheme Dimensional image error provides additional adjustment to ensure that the visual servoing target is always within the field of view,which overcomes the defect that the target of simply using the position-based visual servoing scheme is easily lost.Furthermore,due to the coupling between the movement of the robot arm and the movement of the aircraft during the flight,in order to weigh the movement between the aircraft body and the robot arm,a motion model of the flying robot arm is proposed based on the kinematics model The movement distribution law makes the movement of the flying craft play a leading role when it is far away from the target.When approaching the target,the movement of the manipulator plays a leading role.Finally,considering the possible singularities in the motion of the manipulator,a damped least square method based on SVD-dependent variable damping factor is proposed to reduce the degenerative joint motion and to make the manipulator move smoothly.In order to verify the effectiveness of the proposed method,this thesis builds a hardware experimental platform based on robot operating system(ROS),and in this platform,we carried out the experiment of grasping and manipulating the flight robotic system in this thesis,Analysis.It can be seen from the experimental results that the visual servoing method proposed in this thesis can well overcome the impact of the arm movement on the aircraft body and thus ensure the success rate of grasping.