Automatic Calibration of Camera to World Mapping in omnidirectional vision system using error descent algorithm

摘要：

In RoboCup Middle Size League, Most teams use omnidirectional vision system which provides a 360 degree view of the robot’s surroundings. In such vision systems, a calibration function that maps the image distance to real distance on the floor is essential. With this function, the robot is able to locate itself and determine the distance between objects. However, the function can change drastically after reassembly of the robots. Unfortunately, the method we used to obtain the calibration function is painstaking and error prone. In this paper we describe an error descent algorithm to calculate the function automatically which only requires the detection of the points on field lines.

关键词： RoboCup omnidirectional vision system error descent calibration method RPROP algorithm

作者: Zongjie Xiang Wanmi Chen Haiting Fan

作者单位: School of Mechanical&Electronic Engineering and Automation Shanghai University,Shanghai,China

会议类型: 国际会议

会议名称: 2008高等智能国际会议(2008 International Conference on Advanced Intelligence)

会议地点: 北京

会议语种:英文

在线出版日期: 2008-10-18（万方平台首次上网日期，不代表论文的发表时间）