Research And Implementation Of CoreXY Structured Writing Robot Based On Machine Vision

The design of CoreXY structure writing robot based on machine vision is based on the basic functions of conventional writing robot.It uses machine vision and embedded technology to solve the problems of low intelligence level and difficult to operate independently of the current writing robot.Specifically,the machine vision technology is used to recognize the words in the written Chinese pictures,and then the recognition results are sent to the ARM master control equipped with embedded Linux system,and then the CoreXY structure writing robot is controlled to write.First of all,based on the two problems of handwritten Chinese recognition and control CoreXY structure writing robot,this paper carried out investigation and research,and put forward the use of machine vision technology to recognize Chinese characters in handwritten Chinese pictures and use motion control program to control CoreXY structure writing robot writing solutions.After demonstration,determine the system research technology route and implementation scheme.Secondly,this paper expounds the overall design of the system,the mechanical structure design of the writing robot,the hardware design of the system,the software design of the system and the system operation test.Among them,the overall design of the system describes the overall structure of the system,the workflow of the system and the technical route of the system research in detail,showing the idea of system implementation from a macro perspective;CoreXY structure and motion principle are introduced,and how to assemble CoreXY structure writing robot is discussed.The hardware design of the system introduces the hardware modules of each functional unit and the circuit schematic diagram.The software design of the system is described from four aspects: the main program design of the system,the motion control program design of the writing robot,the Chinese handwriting recognition algorithm design and the upper computer control software design of the writing robot.Finally,the system operation test is carried out,including text editing and rendering test and text image recognition rendering test,and the relevant parameters are given and the test result map is drawn step by step.Thirdly,after several experiments and optimization of the system architecture,the results show that the convolutional neural network based on LeNet-5 model is built by using the deep learning framework TensorFlow to conduct training and testing on the Chinese handwriting data set HWDB1.1 with high recognition accuracy.The writing robot can be controlled by using a PC or the ARM controller combined with the window control program developed by Qt software and the Arduino microcontroller running GRBL motion control program,which can complete the writing function.Finally,the CoreXY structure writing robot design based on machine vision can realize handwritten Chinese recognition and control the robot writing through human-computer interaction window,with advantages of high recognition accuracy and smooth writing.In addition,the design can also be used in teaching research,text image recognition,text transcription,engineering drawing and other aspects,has a certain economic and practical value.