Yanan Liu, T. Hu, Hepeng Ni, Chengrui Zhang, Baoying Lin, E. Judd
{"title":"Design of a PC-based Open Industrial Robot Control System Integrated with Real-Time Machine Vision*","authors":"Yanan Liu, T. Hu, Hepeng Ni, Chengrui Zhang, Baoying Lin, E. Judd","doi":"10.1109/WRC-SARA.2018.8584214","DOIUrl":null,"url":null,"abstract":"This paper presents a new industrial robot control system integrated with real-time image processing based on a Windows PC. This enables both machine vision and robot control tasks to be carried out on a single general-purpose computer rather than using two separate systems to perform different tasks. To accomplish this, a three-layer software structure is developed along with a series of standard peripheral devices in the hardware layout. Real-time performance of the system is tested and we demonstrate the system on a DELTA parallel manipulator. Experimental results show that the manipulator can pick up 120 objects on a moving conveyor per minute. This work demonstrates the feasibility of integrating real-time machine vision tasks with robot control tasks without dedicated hardware.","PeriodicalId":185881,"journal":{"name":"2018 WRC Symposium on Advanced Robotics and Automation (WRC SARA)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 WRC Symposium on Advanced Robotics and Automation (WRC SARA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/WRC-SARA.2018.8584214","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
This paper presents a new industrial robot control system integrated with real-time image processing based on a Windows PC. This enables both machine vision and robot control tasks to be carried out on a single general-purpose computer rather than using two separate systems to perform different tasks. To accomplish this, a three-layer software structure is developed along with a series of standard peripheral devices in the hardware layout. Real-time performance of the system is tested and we demonstrate the system on a DELTA parallel manipulator. Experimental results show that the manipulator can pick up 120 objects on a moving conveyor per minute. This work demonstrates the feasibility of integrating real-time machine vision tasks with robot control tasks without dedicated hardware.