Research On Equipment Detection And Location Method In Robot Inventory System Based On Deep Learning

With the rapid development of computer and Internet technologies,there are more and more application scenarios in the computer room,and the effective management of computer room equipment has become more and more important to enterprises.The inventory of equipment as an indispensable part of the equipment room management,its importance is self-evident.However,traditional manual inventorying has the problems of high cost,low efficiency,error-prone changes,inability of timely feedback,and the inability to locate problems in a timely manner.Afterwards,the method of using handheld inventory devices is not flexible enough and can only be used for the inventory of fixed assets.Therefore,we urgently need an efficient and powerful inventory system to solve these problems.Therefore,we propose a robot inventory system based on deep learning: fixed information for the device is placed in the QR code,and then captured and recognized by the camera;for the variable location information of the device,the binocular vision and deep learning algorithm are combined to detect and locate the device.Through the performance comparison of the target detection algorithm based on deep learning,combined with the actual detection requirements,we chose to use YOLOv2 algorithm as the basis for optimization to implement our inspection tasks.After that,we chose Kinect as the image acquisition device to acquire the depth information to achieve the positioning task.Through the combination of binocular vision,deep learning algorithm and robot,the system can autonomously complete the detection and positioning functions of the equipment.Not only saves labor costs,but also is more efficient,and can adapt to frequent machine room inventory changes.More importantly,due to the use of deep learning algorithms,the accuracy of system detection and positioning will increase with use,which is the advantage of deep learning.Therefore,it is of great significance to apply deep learning to inventory in the equipment room.