Research And Design Of Robotic Control System For Facility Agricultural Picking

China is a large agricultural country.Although the agricultural population is large,with the continuous acceleration of the industrialization process,the problem of insufficient agricultural labor force has become increasingly prominent.For the labor-intensive work such as the selection and picking of vegetables and fruits,there is an urgent need for intelligent robots to replace manpower,so as to improve production efficiency.However,due to the complex agricultural production environment,the traditional industrial robot technology is difficult to meet the actual requirements.According to the needs of the picking robot in facility agriculture,this paper studies from the aspects of navigation positioning and operation planning,so as to realize the autonomous navigation of the picking robot and improve the picking efficiency.The specific research contents are as follows:(1)Overall scheme design of picking robot.Through investigation and analysis,according to the actual needs of the picking robot,the overall scheme of the picking robot is designed.Through comparative analysis,the hardware system of the picking robot is designed and selected.The scheme of the combination of the crawler mobile platform and the six axis manipulator is adopted.The industrial computer is used as the brain of the picking robot,the can board is used to complete the information interaction between the industrial computer and the mobile platform,and the camera is used for visual ribbon navigation.The software system of picking robot is divided into two modules: mobile platform navigation system and picking mechanism control system.(2)Design of motion control system for mobile platform of picking robot.The method of visual ribbon navigation is used as the navigation scheme of unmanned harvesting robot.The image information obtained by the camera is preprocessed by means of color space transformation,filtering and binarization,and then Canny operator is used for edge detection and polynomial least square method is used to fit the center line of the color band.Analyze the trajectory tracking algorithm,use the most stable PID algorithm for trajectory tracking,and realize unmanned driving.Finally,experiments are designed to verify the stability and practicability of the mobile platform navigation system.(3)Motion modeling of picking manipulator.The homogeneous coordinate transformation and DH parameter method required for the modeling of the picking manipulator are analyzed.Using the standard DH method and the dimension parameters of the picking manipulator,the model is built and the DH parameter table is filled in.At the same time,the forward kinematics of the manipulator is analyzed.Finally,MATLAB software is used to verify the working space of the picking manipulator by using the numerical method to verify that the selected manipulator meets the working needs of the picking robot.(4)Motion planning of picking manipulator.Firstly,the advantages and disadvantages of several commonly used motion planning algorithms and the scenarios used are analyzed.Then the RRT algorithm is studied.Aiming at the blindness problem,the target bias factor is introduced to improve the convergence speed.The sampling space is optimized at the starting point and the target point to avoid repeated sampling and improve the efficiency of the algorithm.Finally,through the experiments of designing simple map and complex map,the RRT algorithm,RRT star algorithm and improved RRT algorithm are compared to verify the efficiency of the improved RRT algorithm.(5)Experimental verification of picking robot.Firstly,using the three-dimensional model of the manipulator,the URDF model and moveit configuration function package of the manipulator are established step by step to complete the simulation model of the manipulator under ROS.Then the improved RRT algorithm is integrated into ompl.Design experiments to verify the feasibility of the improved algorithm by comparison.Finally,the picking experiment environment is built to complete the picking experiment of the picking robot.