Visual And Inertial Integrated Guided Control System Of AGV

With the development of "industry 4.0",automatic guided vehicle as a kind of intelligent mobile robots integrates the external environment perception,intelligent decision and motion control technology.It undertakes the task of transporting materials in intelligent system and promotes the development of intelligent factory,intelligent production and intelligent logistics.At present,the traditional AGV uses the high cost laser guidance or uses the magnetic guided and visual guidance.However,the path planning of magnetic guided and visual guidance are not flexible.In order to solve the above problems,in this paper,QR code is used as the positioning module to design the path of the trackless guidance AGV and the inertial guidance module is used to provide the position and orientation information of AGV between adjacent QR codes.This hybrid guidance makes AGV more accurate positioning,lower cost,higher flexibility in path planning and change,more convenient for front-end coding and terminal decoding.The main research work is as follows:(1)According to the technical index completed the overall design of the project,the design of the AGV vehicle control system and the general control system.The working principle of the AGV control system is introduced.(2)In order to obtain the real-time position and attitude information of the AGV,determined the visual and inertial integrated guidance scheme,the output of the inertial device is filtered,and the velocity calculation,the position calculation and the heading calculation are carried out.Because the inertial guidance has accumulated error,and the precision of the vision guidance is much higher than that of the inertial guidance so the output of the inertial device is corrected in the QR code positioning module.(3)Because the QR image acquired by AGV in the course of motion is fuzzy,it can't be directly identified.Firstly,according to the inertial device's output of the AGV's working speed and the industrial camera exposure time to estimate the degradation function parameters.Secondly,the ROI image of QR code is obtained by preprocessing the image,and the ROI image motion blur restoration is realized by using the L-R algorithm.Finally,we use the affine transformation principle to traverse the corrected QR code to get the standard QR code,which is used to identify the position information.(4)The AGV fuzzy PID motion control system with double loop model is designed,which makes the AGV with high stop precision and high tracking precision.(5)According to the characteristics of AGVS,Complete the main control system of the AGV,including: selecting the communication protocol and data format,the dispatching strategy,task assignment method and path planning.