The conventional operation and maintenance of 5G computer rooms is characterized by a low level of automation, which can result in untimely replacement of hard disks and potential data loss. In order to achieve efficient and timely replacement of hard disks in 5G computer rooms, this paper proposes a novel methodology for attitude recognition of such disks using an RGB-D depth camera and quintic polynomial interpolation algorithm. This method obtains two-dimensional position information of the hard disk through RGB images, and then combines depth images to obtain the coordinate system of the three-dimensional hard disk. The precise identification of the area to grasp a hard disk is achieved through the design of the grasping process, RGB-D image preprocessing, attitude estimation, and grasping trajectory planning. Additionally, by combining hard disk attitude estimation, the robot arm can be effectively controlled to complete the grasping process of the hard disk. The experiments carried out on the visual recognition method proposed in the article have shown that it achieves high accuracy in recognizing the grasping area of a hard disk. Moreover, the robot arm grasping system based on this method has been used to replace hard disks in 5G computer rooms automatically.
Hard Disk Posture Recognition and Grasping Based on Depth Vision
Lect.Notes Computer
International Conference on Intelligent Robotics and Applications ; 2023 ; Hangzhou, China July 05, 2023 - July 07, 2023
2023-10-21
12 pages
Article/Chapter (Book)
Electronic Resource
English
RGB-D , Hard disk , Attitude estimation , Robot arm grasping trajectory planning Computer Science , Artificial Intelligence , Software Engineering/Programming and Operating Systems , Computer Applications , Computer Communication Networks , Special Purpose and Application-Based Systems , User Interfaces and Human Computer Interaction
Hard Disk Posture Recognition and Grasping Based on Depth Vision
TIBKAT | 2023
|Stereo-Vision-Guided Object Grasping
British Library Conference Proceedings | 1999
|Stereo-vision-guided object grasping
Automotive engineering | 1999
|A 3D Human Posture Approach for Activity Recognition Based on Depth Camera
BASE | 2016
|