Automatic Sorting Robot With Machine Vision

With the development of robots and machine vision,automatic sorting robots use machine vision to locate targets and use robots to crawl targets.With the development of the logistics industry and the processing industry,the use of automatic sorting robots is getting higher and higher.Automatic sorting robots generally have two structures.One is the combination of computer and machine vision.The camera captures the target image and transmits the information to the computer.The computer processes the image to obtain the target information and controls the robotic arm to capture the target.The other is that the camera captures the image and processes it to obtain the target information,and sends the information to the control cabinet through the communication protocol to complete the control of the robot arm.The latter is the development trend of automatic sorting robots.Based on the above situation,an automatic sorting robot of the micro controller is designed in this paper.The hardware structure is as follows: DSP processor using TMS320C6748 as the core,using the camera with VPIF interface to connect with the hardware of the DSP development board,using the stepper drive Three-axis robotic arm to achieve the target's grasp.The microprocessor integrates the image processing and robotic arm control program.The designed microprocessor has a small space,and the processing function is more powerful than the computer,reducing the communication links between image processing and control systems,and reducing robotic arm controller hardware such as control cabinets.the verification of the overall system meets the automatic sorting function.In this paper,a two-phase hybrid stepper motor is adopted for the stepping control system.The motor operation is divided into acceleration,uniform speed,and deceleration phases.The acceleration and deceleration phase time should be as short as possible,and the time for the motor to move at a constant speed should be as long as possible.In order to avoid step loss of stepping motor,a step control function is proposed.Through the verification of the control function,the rate of change of pulse frequency is getting larger and larger.the initial value of the speed of the stepping motor is small,the acceleration is small,and the lost step is avoided.When the motor rotor adapts to moment of inertia,the motor acceleration increases to complete the acceleration phase as soon as possible.In this paper,the three-axis robotic arm is modeled using SolidWorks 3D drawing software.The robot coordinate system is established for each link of the robot arm.The DH parameter method is used to obtain the relevant parameters of the robot arm.The positive kinematics solution of the robot arm is calculated.Calculate the transformation matrix of the end of the manipulator relative to the base coordinate system and verify it with a special position;the inverse solution of the robot arm is obtained according to the geometric method,and the relationship between the joint angle and the end coordinate of the manipulator is obtained and verified by MATLAB.A Jacobian matrix between the arm tip velocity and the joint velocity is established,and the trajectory of the articulated arm joint is smoothed using a cubic polynomial.This article calibrates the camera,uses the MATLAB calibration toolbox to obtain the camera's internal and external parameters,obtaining information such as the camera's main point coordinates,focal length,the camera coordinate system is related to the robot coordinate system transformation matrix.The optimal threshold method is used to binarize the grayscale image,and the Sobel operator is used to perform edge detection to obtain the target information.The acquired target information is compared with the information of the known control,and the target is identified in multiple spaces.The conveyor belt is transported in a non-uniform motion,the movement of the conveyor belt is analyzed,and the capture position information is acquired through the target position in conjunction with the movement of the conveyor belt,the movement of the robot arm.In this paper,the hardware and software part adopt the translational mechanical claw to cope with the change of the space coordinates when the claw catches the target,avoids the revision of the coordinate error,and calls the OpenCV function library to the CCS project for image processing to improve the image processing capability.By verifying the overall system,the micro controller designed in this paper satisfies the function of automatic sorting,reduces the communication links between the machine vision and the robot arm control system,cancels the hardware of the robot arm control cabinet.