The Effect of Contact Surface Curvature on The Accuracy of Fingernail Imaging for Tactile Force Measurement

被引:2
|
作者
Fallahinia, Navid [1 ,2 ]
Mascaro, Stephen A. [1 ,2 ]
机构
[1] Univ Utah, Robot Ctr, Salt Lake City, UT 84112 USA
[2] Univ Utah, Dept Mech Engn, Salt Lake City, UT 84112 USA
基金
美国国家科学基金会;
关键词
HAND;
D O I
10.1109/HAPTICS45997.2020.ras.HAP20.14.cecd307b
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Fingernail imaging has been shown to be effective in estimating three-dimensional tactile forces when the finger pad is pressed against a flat surface. However, the effectiveness of this method when touching curved surfaces has not been comprehensively established. The objective of this paper is to independently investigate the possible changes in the calibration model and force estimation error due to the variation in contact surface curvature. In this study, experiments are conducted using 18 different surfaces by changing their radii of curvature in two axes. The experimental results show a maximum RMS validation error of 0.61N and 0.59N for calibration with a spherical and a flat surface, respectively in all three force dimensions (10% of the full range of forces) for radii of curvature greater than 7mm, with curvature having no significant effect on force estimation error in this range. As the surface becomes sharply curved (r<7mm), the force estimation error significantly degrades. Therefore this paper establishes a clear range of surface curvatures for which the fingernail imaging method is robust, enabling this method to be used in tactile experiments with curved surfaces.
引用
收藏
页码:760 / 766
页数:7
相关论文
共 50 条