Research And Implementation Of Intelligent Assistant Maintenance System Based On Augmented Reality Technology

Currently,China is experiencing a period of rapid industrial growth,which has increased the demand for logistics production lines in the field of industrial production,and promoted more professional demand for the maintenance of logistics production line equipment.The maintenance of traditional logistics production line equipment has problems such as complex environmental factors,cumbersome access to technical data in paper or traditional IETM systems,etc.,which ultimately leads to time-consuming maintenance work and low work efficiency of maintenance personnel.In view of the problems that will occur in the maintenance of equipment of traditional logistics production lines,this thesis takes the logistics production line as the background,and YF gearmotor,one of the main equipment of the logistics production line,as a representative research object,studies and realizes a set of intelligent auxiliary maintenance system based on augmented reality technology,which is based on augmented reality technology(AR)and uses object detection and three-dimensional registration and tracking technology to superimpose virtual information such as IETM system Chinese characters and three-dimensional virtual models in real scenes.Improve the experience of maintenance operators when working in an open environment,and then effectively improve the maintenance efficiency of maintenance operators.The main research work of this thesis is as follows:(1)The importance of integrating the IETM system into AR technology for overhaul and maintenance work is described,the S1000 D standard used in the construction of IETM is analyzed and explained,and the advantages of public source database management and data module for equipment technical information management are briefly described.Through the analysis of the maintenance characteristics of logistics production line equipment,a systematic and structured maintenance information model is established.The multimedia guidance information integrated into the IETM system of AR technology is classified,and the multimedia information library is constructed based on XML technology to realize the management of multimedia guidance information.(2)In order to solve the problems of difficulty in collecting datasets and time-consuming labeling datasets in the process of target recognition,the target detection method based on multi-view feature library was studied.When building a multi-view feature library,the 3D virtual model is rendered with materials and colors.the authenticity is restored through random background fusion to realize the self-generation of the target dataset.the model weight parameters obtained by the transfer learning method are applied to the SSD neural network,which can effectively improve the accuracy of target detection and achieve fast and accurate target recognition effect.Experiments show that the accuracy of different target recognition tasks is above 95%,and the real-time frame rate can reach 51.2 fps,which effectively improves the speed and accuracy in complex environments,and provides better position information for subsequent virtual information and real scene overlay tasks.(3)Aiming at the problems of poor real-time and insufficient accuracy of 3D registration tracking,a 3D tracking registration method based on point cloud registration was designed,the principle of 3D registration tracking was analyzed firstly,and the camera calibration method was determined to prepare for the subsequent point cloud registration work.Then,the two-dimensional target region of interest is mapped to the three-dimensional point cloud space and preprocessed,the initial target pose is calculated by using the ICP registration algorithm based on sampling consistency,and the particle position is predicted on this basis,and finally the KLD particle filter algorithm is improved,and the experimental results show that the improved two indicators of distance mean squared error and consumption time have been significantly improved compared with the original algorithm,which can better realize the realtime tracking of equipment to be repaired in complex scenarios.(4)Based on the above research results,an intelligent auxiliary maintenance system based on augmented reality technology is realized.Firstly,taking the YF geared motor of logistics production line equipment as an example,the technical data of YF geared motor is collected and sorted,and the fault tree model of the geared motor is constructed on the basis of decomposing its fault repair work.Finally,the case of augmented reality intelligent auxiliary maintenance was completed with the geared motor as the object,and the effectiveness of the system was verified,and on the basis of the system effectiveness,four indicators were cited to evaluate the system,and the evaluation results verified the advantages of the system in response time,virtual and real fusion effect,intuitiveness and operation difficulty.