Research On AGV Path Following Control Strategy Based On Disturbance Observer

With the development of science and technology and the continuous increase of labor costs,the development and application of AGV(Automated Guided Vehicle),which can realize functions such as automated assembly,transfer,and sorting,has attracted more and more attention from the industry.Path following control technology is an important part of AGV technology.Its control accuracy directly affects the execution of the overall task.At the same time,as the work tasks become more and more complex,the requirements for the path following control accuracy of AGV are getting higher and higher.However,the AGV is susceptible to interference factors such as load,external interference,system parameter perturbation,etc.during the path following task of the AGV,resulting in a decrease in the path following accuracy of the AGV.In order to improve the path following accuracy of AGV under interference environment,this dissertation takes wheeled AGV as the research object,and studies the path following cascade control strategy of AGV based on backstepping,disturbance observation technology and sliding mode control technology to reduce the impact of interference on its path following accuracy.The main research contents are as follows(1)Analyze the various wheel train structures of the AGV,and select the two-wheel differential drive AGV as the research object,consider the load of the AGV and the influence of external interference on the AGV,and use the Lagrange mechanical analysis method to establish the AGV's kinematics and dynamics models.(2)The horizontal curve path following problem of AGV is analyzed.Based on the kinematics model of AGV,the path following algorithm of local convergence and global convergence is studied.Aiming at the problem that the local convergence algorithm cannot achieve global tracking in any direction,a global path following control strategy that can converge along the direction of the vector field is proposed,and the backstepping method is used to prove the stability of the proposed algorithm.The simulation results show that the proposed global path following control strategy can realize that the AGV can follow the given reference path according to the direction of the designed vector field under any initial pose.(3)Aiming at the problem that the AGV is affected by load and external interference in the actual operation of the path following accuracy,a control strategy that is robust to compound disturbances is studied.Based on the dynamic model of AGV,a super-twisting disturbance observer with finite time convergence is designed to estimate the compound interference online,and feedback the interference estimation value to the controller,so as to suppress the influence of the compound interference on the path following accuracy;the non-singular terminal sliding mode control algorithm is used to design a torque controller with finite time convergence to improve the robustness of the system;aiming at the control and chattering problems of the approaching process in sliding mode control,nonlinear functions are introduced into the design of the approaching law to increase the approaching speed and suppress the system chattering phenomenon.The simulation results show that the designed control strategy can improve the dynamic characteristics and robustness of the system,and has a strong inhibitory effect on the load and different types of external interference.