Hybrid image plane/stereo (HIPS) manipulation for robotic space applications

被引:4
作者
Robinson, Matthew L.
Baumgartner, Eric T.
Nickels, Kevin M.
Litwin, Todd E.
机构
[1] CALTECH, Jet Prop Lab, Pasadena, CA 91109 USA
[2] Ohio No Univ, Ada, OH 45810 USA
[3] Trinity Univ, San Antonio, TX 78212 USA
基金
美国国家航空航天局;
关键词
vision-based manipulation; stereo triangulation; stereo vision; space robotics;
D O I
10.1007/s10514-007-9032-0
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Manipulation systems for planetary exploration operate under severe limitations due to power and weight restrictions and extreme environmental conditions. Typically such systems employ carefully calibrated stereo cameras and carefully calibrated manipulators to achieve precision on the order of ten millimeters with respect to instrument placement activities. The environmental and functional restrictions under which these systems are used limit the operational accuracy of these approaches. This paper presents a novel approach to stereo-based manipulation designed to robustly achieve high precision levels despite the aforementioned limitations. The basic principle of the approach, known as Hybrid Image Plane/Stereo (HIPS) Manipulation, is the generation of camera models through direct visual sensing of the manipulator's end-effector. The HIPS method estimates and subsequently uses these models to position the manipulator at a target location specified in the image-planes of a stereo camera pair using stereo correlation and triangulation. In-situ estimation and adaptation of the manipulator/camera models in this method accounts for changes in the system configuration, thus ensuring consistent precision for the life of the mission. The end result is a increase in positioning precision by a factor of approximately two for a limited version of HIPS, and an order of magnitude increase in positioning precision for the full on-line version of HIPS.
引用
收藏
页码:83 / 96
页数:14
相关论文
empty
未找到相关数据