首页    期刊浏览 2024年12月03日 星期二
登录注册

文章基本信息

  • 标题:視界情報による潜水艇の制御情報取得に関する考察 (第3報)
  • 本地全文:下载
  • 作者:石谷 久 ; 馬場 康子
  • 期刊名称:日本造船学会論文集
  • 印刷版ISSN:0514-8499
  • 电子版ISSN:1884-2070
  • 出版年度:1990
  • 卷号:1990
  • 期号:168
  • 页码:267-274
  • DOI:10.2534/jjasnaoe1968.1990.168_267
  • 出版社:The Japan Society of Naval Architects and Ocean Engineers
  • 摘要:

    In the previous studies, basic methods to estimate motion or attitude of the vehicle (camera), i. e. ego motion, from video image data obtained continuously by a VTR are proposed, and their performances and applicability to the actual images are investigated. By applying an iterative estimating algorithm based on the Extended Kalman Filter to continuous pictures of the ground surface, it is assured that the motion can be estimated with enough stability. In this study, continuous underwater images are obtained by a ROV, and its motion is estimated by the same algorithm to investigate applicability of the methods for actual use in the water. The accuracy of the estimation is evaluated for various conditions, and that of 0.25 edgree (maximum error) in yaw angle is assured for ordinary images with appropriate landmarks. The necessity of calibration for the edge distortion is pointed out in case very few landmarks exist in the pictures.

国家哲学社会科学文献中心版权所有