Research On Equipment Maintenance And Remote Guidance Based On AR Technology

The combination of Augmented Reality(AR)technology and remote assistance technology in the industrial field brings extraordinary experiences and effectively solves the maintenance inefficiencies caused by the limited skill level of maintenance personnel.The thesis proposes an AR-based equipment maintenance guidance system to address the problems of high bandwidth requirements,low interactivity,lack of privacy,and few guidance methods in the existing video-based remote guidance.The system uses an optical see-through HMD as the display device,and maintenance personnel can obtain virtual guidance content for troubleshooting through the headset.When the intelligent guidance is not enough to assist the maintenance personnel to complete the repair task,a remote expert can be connected through the system and both parties can collaborate to solve the fault problem in a hybrid environment.In the process of the expert guiding the maintenance personnel remotely,the former can observe the maintenance status of th e latter in real time by means of Virtual Reality(VR)and achieve the purpose of mutual communication with the maintenance personnel on site by changing the display information of the maintenance site.Through specific cases,it is shown that the system can effectively improve the efficiency of maintenance personnel in solving equipment faults.The main research work of the thesis is as follows:(1)Research on three-dimensional registration and occlusion for optical see-through HMD.Realize the three-dimensional registration of virtual devices in the real environment.Manage the rendering sequence of virtual parts,solve the problem of occlusion,and ensure the accuracy and clarity of the presentation position of the guidance information.Take a world anchor to the virtual device to achieve its maintenance in the real environment.(2)Dynamic presentation of maintenance information and interactive methods.Definition and description of maintenance knowledge content.Mapping and parsing methods of guideline information in the maintenance end.Three human-computer interaction methods of ray,gesture and voice are proposed in the maintenance client to improve the interaction experience of maintenance personnel.Research on the implementation method of intelligent guidance and establish fault case multiway tree to assist maintenance personnel to obtain coincident and effective fault solutions.(3)Communication method and realization for remote assistance.Establish the system network communication method,design the communication message format and communication content,and define the specific flow of messages.The room management strategy is adopted to increase the capacity of the server to the client to meet the needs of multi-users using the system.Discuss and realize the remote synchronization technology of maintenance scene,respectively realize the remote synchronization of equipment and maintenance personnel’s head and palms,so that remote experts are informed of the real maintenance status.The real-time and accuracy of remote synchronization are verified,and the analysis shows that the real-time and accuracy of remote synchronization of scenes satisfy the requirements of system usage.(4)Development and realization of AR-based equipment maintenance guidelines prototype system.Taking manual transmission as the object of study,verifying and analyzing the effectiveness of the guidelines for transmission oil seepage and transmission noise cases,and determined that the system has good feasibility and practicality,which can effectively guide the maintenance personnel to complete the maintenance work of the equipment.The system proposed in the thesis can effectively overcome the problems of interactivity and privacy that exist in video-based remote assistance,improve the experience and efficiency of both users,and have more obvious advantages in application scenarios with low bandwidth,high privacy,complex equipment,and difficult descript ion.The research results of the thesis provide a feasible idea for AR-based remote assistance maintenance research.