Research On Trajectory Tracking Algorithm Of Mobile Robot For Loading And Unloading

Loading and unloading mobile robot replaces manual loading and unloading operation to realize its automation and change the situation that the job is mainly completed by manual,which has a broad application prospect.Motion control is one of the core functions of loading and unloading mobile robot.This thesis mainly studies its trajectory tracking algorithm and designs its autonomous navigation system.The main work and conclusions are as follows:According to the robot requirements and the existing constraints of hardware technology,the hardware control scheme based on Ether CAT bus has been designed.UML modeling language is used to model the movable function modules,and an autonomous navigation framework based on 2D lidar is built based on the ros-navigation,and the software system has been developed using unified process.Based on the theory of kinematics and the coordinate system transformation,the kinematics models of velocity and position of four Mecanum wheels are established,and the robot dynamic model is established on the basis of Lagrange theory.DWA+PID trajectory tracking algorithm is proposed to reduce processor resource consumption and achieve high precision control requirements.The dynamic space of DWA algorithm,the cost function of the algorithm for linear speed control and the incremental PID for angular speed control are designed.The online simulation of MATLAB shows that the time of the DWA+PID algorithm is one third of that of the DWA algorithm.Nonlinear MPC,absolute MPC and incremental MPC trajectory tracking algorithms are designed to solve the non-theoretical optimal control problem of DWA algorithm.In order to reduce time-consuming,the incremental PID is used to control the angular velocity and MATLAB is used for offline simulation of straight line and circular trajectory.The results show that the algorithm satisfies the requirement of control.From the comperehensive consideration of control accuracy,feasibility,algorithm time,nonlinear MPC is chosen for the robot control.The time of this algorithm is one half that of the pure MPC algorithm.The autonomous navigation hardware of the robot is built and the supporting software system is designed in this thesis.The simulation and experiment of track tracking are carried out in the actual loading and unloading scene,and its result shows that the longitudinal position meets the accuracy requirement of no more than 45 mm and the system time consumption meets the requirements of less than 0.5s,which proves that the two trajectory tracking algorithms can satisfy the high precision and low time consumption control requirements of practical operation under kinematic constraints.