A Robust Position and Posture Measurement System Using Visual Markers and an Inertia Measurement Unit

A Robust Position and Posture Measurement System Using Visual Markers and an Inertia Measurement Unit
复制标题

使用视觉标记和惯性测量单元的稳健位置和姿势测量系统

DOI:
10.1109/iros40897.2019.8967887
复制
发表时间:
2019
期刊:
2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
影响因子:
--
通讯作者:
Y. Matsumoto
Y. Matsumoto
中科院分区:
--
文献类型:
--
作者:
Kunihiro Ogata;Hideyuki Tanaka;Y. Matsumoto

文献摘要

参考文献

被引文献

相似文献

移动机器人和机械臂的自动控制需要高精度和鲁棒性的物体位置和方向估计技术。虽然已经开发了使用标记或机器学习的方法,但尚未实现通用和高度准确的估计。我们的团队开发了一种高精度的视觉标记“LentiMark”,用于高精度的位置和姿态估计,但当相机无法获得标记图像时,数据会丢失。因此,我们开发了标记-惯性测量单元系统,用于将视觉标记与惯性测量单元(IMU)集成。当摄像机无法获取视觉标记的图像时,从IMU数据中恢复任何丢失的数据。然而,当根据加速度传感器值计算位置时,传感器误差增加了估计误差。因此,我们开发了一种利用缺失数据前后的测量值进行误差校正的方法。评价实验证明,该方法可以有效地估计缺失数据。
Automatic control of mobile robots and robot arms requires techniques for estimating the position and orientation of objects with high accuracy and robustness. Although methods using markers or machine learning have been developed, general-purpose and highly accurate estimations have not been realized. Our team developed a high-accuracy visual marker “LentiMark” for high-accuracy estimations of position and posture, but data are lost when the camera cannot obtain marker images. We therefore developed the Marker-IMU system for integrating visual markers with an inertia measurement unit (IMU). When cameras cannot acquire the image of a visual marker, any missing data are restored from IMU data. However, when calculating positions from acceleration sensor values, sensor error increases estimation error. Therefore, we developed a method for error correction using measurements from before and after the missing data. Evaluation experiments confirm that missing data can be estimated using the proposed method.
二维伊辛模型温度估计的深度学习
DOI: --
发表时间: 2018
期刊:
影响因子: --
作者:
青木健一;藤田達大;神宮翼;小林玉青
通讯作者: 小林玉青