Research On Grasping Control Of Rotor Aerial Manipulator

Unmanned Aerial Vehicles(UAVs)are used in more and more fields widely with the rapid development of the aerial robot technology.As one of the most commonly used UAVs,multirotor drones are widely applied in reconnaissance,cruise,aerial photography and other fields for their small size,flexible movement,and fixed-point hovering.But the traditional quadrotor cannot actively apply operations to the environment,this limits the application range of quadrotors.To solve this problem,researchers install an active operation device on the quadrotor and design the rotor aerial manipulator to improve the active operation capability of the quadrotor.This research has important engineering value and scientific significance.In this paper,for the grasping application of the rotor aerial manipulator,the robust attitude control methods are studied aimed at the problem that the movement of the manipulator would cause disturbance to the system.And extensive simulations are designed to demonstrate the control performance.Firstly,in this paper,the state of art at home and abroad is analyzed.The kinematics models of the quadrotor and the two-degree-of-freedom manipulator are established,respectively.After analyzing the advantages and disadvantages of overall modeling and independent modeling,the independent modeling method is selected to establish the dynamic model of the system.The Newton-Euler equation is used to establish the dynamic model of the quadrotor,and the Lagrange equation is used to establish the dynamic model of the two-degree-offreedom manipulator.Secondly,the system controller is designed based on the established system model.A separate control strategy is proposed,and the controllers of the quadrotor and the manipulator are designed,respectively.To deal with the problem that the movement of the manipulator can seriously disturb the attitude of the quadrotor during grasping objects,a fuzzy sliding mode control(FSMC)attitude control method based on the disturbance observer is proposed.The extended state observer(ESO)is used to estimate and compensate the disturbance that caused by the motion of the manipulator.The fuzzy control is utilized to suppress the chattering effect of the traditional sliding mode control.At the same time,a feedforward all-coefficient adaptive control(FACAC)based on characteristic modeling and ESO is proposed,which effectively improves the response speed and robustness of traditional algorithms.Thirdly,for the grasping task of the rotor aerial manipulator,a reasonable segmented grasping strategy is designed based on the eagle predatory action in nature.The whole process is divided into two stages: the movement stage of the manipulator,the grasping and leaving stage.The movement stage of the manipulator is mainly to adjust the posture of the manipulator in advance to make it approach the target in an optimal posture.After approaching the target,the manipulator quickly grasps the target while flying away from the operating point,and hovering to the end point of the prescribed and the task is over.Based on the grasping strategy,the minimized Snap algorithm is applied to plan the flight trajectory of the system to ensure that the system can smoothly and reliably pass through the three key points of the starting point,the grasping point and the ending point.At the same time,the trajectory tracking controller is designed based on the PID control theory.Finally,the robustness,advancement and effectiveness of the proposed algorithm are evaluated through MATLAB simulation.The simulation results show that the designed attitude controller has a significant improvement in the response time of the system,the suppression effect on disturbance and so on.And it can achieve the task of grasping target objects in the air.