A LOCATION SYSTEM COMBINING IMAGE PROCESSING OF MULTIPLE CAMERAS WITH 3D MODEL OF A WAREHOUSE FOR AUTOMATED GUIDED VEHICLE
In many warehouses, Automated Guided Vehicles (AGVs) are used for increasing the efficiency. In general, AGVs depend on electromagnetic-or magnetic-type of guiding elements for navigation. Hence, AGVs that do not require any guiding elements for navigation are desired. Therefore, in this paper, we develop a location system combining an image processing of multiple cameras with 3D model of a warehouse for AGV. In this system, four cameras are installed to AGV, and input images of these cameras are compared to forecasting images which are made by 3D model of the warehouse. Then, the viewpoints of four cameras in the 3D model which are able to input images corresponding images of cameras are estimated. Using the estimated viewpoints in the 3D model, the self-position of AGV in the warehouse is estimated. Furthermore, a mobile robot system which is adopted the developed location system is composed, and the effectiveness of the developed location system is demonstrated by the experiments in an environment simulated according to an actual warehouse.
image processing 3D model self-position estimation automated guided vehicle
Mitsuru Jindai Masahiro Kameda Tomio Watanabe
Faculty of Computer Science and System Engineering, Okayama Prefectural, University, 111 Kuboki, Soj Kougakukiso Co., Ltd., 2-1-1 Toshiharu Building 2F, 2-3-8, Heiwadoori, Matsuyama, Ehime,790-0807,Jap
国际会议
The Tneth International Conference on Industrial Management(第十届工业管理国际会议 ICIM 2010)
北京
英文
723-728
2010-09-16（万方平台首次上网日期，不代表论文的发表时间）