Research On Transport Robot Location By Data Matrix

In recent years,with the development of population aging in China,many factories are facing the problem of labor shortage and rising labor costs.Driven by the concepts "industry 4.0" and "intelligent manufacturing",more and more factories use automation equipment instead of manual labor,which increases productivity,reduces production risks,improves the working environment of workers and improves efficiency.There are some problems that narrow material transportation and accurate docking in SMT production line,magnetic navigation robot do not have enough flexible,and Laser navigation robot has higher requirements for workspace.Pointing at the problems that this paper designed an omni-directional wheeled location by data matrix.The robot has the ability of planning and tracking curve path,and adapt to irregular material conveying path and narrow space factory environment.Based on the data matrix code,it can also achieve accurate positioning,accurate docking with the docking mechanism of production line,and realize automatic material handling.The main work and research results of this paper are as follows:Firstly,this paper analyzed the advantages and disadvantages of differential drive,steering wheel drive and omni-directional drive,combining with the narrow workspace,the omni-directional driving mode based on Mecanum wheel is adopted.According to the motion characteristics and task requirements of Mecanum wheel,the overall mechanical structure design of the robot is completed.The mechanical design of the wheel suspension mechanism is completed for the convex ground in the working scene,which realizes the excellent obstacle-surmounting ability of the robot to obstacles with a height of 10mm or less;and the mechanical structure design of the material-handling platform is completed according to the requirement of the loading task.Secondly,a robot navigation and positioning method based on data matrix is designed.Aiming at the requirement of precise docking of SMT line docking mechanism,data matrix is used as artificial landmark to realize the precise positioning of the robot.The kinematics model of the transport robot are established by using coordinate transformation between the robot coordinate system and the world coordinate system.In order to realize the navigation of the robot in the area without data matrix,the inertial navigation mode using gyroscope and coder is designed;the data fusion of gyroscope and coder is completed by Kalman filter algorithm,and the data periodicity is measured by data matrix.By correcting the inertial navigation positioning error,the global positioning and local positioning of the mobile robot are realized.Thirdly,according to the kinematics model,the tracking error model of the mobile robot is established.And a path tracking controller is designed based on Lyapunov criterion,and the feasibility of the controller is verified by simulation.Because of uncertain time delays in the actual working environment,combined with the kinematics model constraints,the planar motion is extended to three-dimensional space motion,and select the path length as a non-time reference.Used trapezoidal curve to plan the velocity,the time-based path tracking controller is transformed into a non-time reference path tracking controller.A method for calculation the position and pose of reference robot in linear path is presented.The feasibility of the path tracking controller is verified by simulation,and the reference path tracking in the non-time reference frame of the mobile robot is realized.On the basis of linear path tracking,the paper used Bezier curve to fit arbitrary path selected the path length as a non-time reference and Bezier curve parameter as intermediate variable,used trapezoidal curve to plan the velocity,proposed a method for calculating the position and pose of reference robot in Bezier curve,and verifies the tracking of arbitrary curves by transport robot in non-time reference system.Finally,the hardware and software designed of the control system is completed,and designed experiment to verify the feasibility of the above algorithms.And we summarizes the work done,and put forward the prospect of the further research direction and problems.