Autonomous Grasping Using Novel Distance Estimator

Q3 Engineering

International Journal of Mechanical Engineering and Robotics Research Pub Date : 2023-01-01 DOI:10.18178/ijmerr.12.2.64-77

M. Skaldebø, B. A. Haugaløkken, I. Schjølberg

{"title":"Autonomous Grasping Using Novel Distance Estimator","authors":"M. Skaldebø, B. A. Haugaløkken, I. Schjølberg","doi":"10.18178/ijmerr.12.2.64-77","DOIUrl":null,"url":null,"abstract":"—This paper introduces a novel distance estimator using monocular vision for autonomous underwater grasping. The presented method is also applicable to topside grasping operations. The estimator is developed for robot manipulators with a monocular camera placed near the gripper. The fact that the camera is attached near the gripper makes it possible to design a method for capturing images from different positions, as the relative position change can be measured. The presented system can estimate relative distance to an object of unknown size with good precision. The manipulator applied in the presented work is the SeaArm-2, a fully electric underwater small modular manipulator. The manipulator is unique in its integrated monocular camera in the end-effector module, and its design facilitates the use of different end-effector tools. The camera is used for supervision, object detection, and tracking. The distance estimator was validated in a laboratory setting through autonomous grasping experiments. The manipulator was able to search for and find, estimate the relative distance of, grasp, and retrieve the relevant object in 12 out of 12 trials.","PeriodicalId":37784,"journal":{"name":"International Journal of Mechanical Engineering and Robotics Research","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Mechanical Engineering and Robotics Research","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.18178/ijmerr.12.2.64-77","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"Engineering","Score":null,"Total":0}

引用次数: 0

Abstract

—This paper introduces a novel distance estimator using monocular vision for autonomous underwater grasping. The presented method is also applicable to topside grasping operations. The estimator is developed for robot manipulators with a monocular camera placed near the gripper. The fact that the camera is attached near the gripper makes it possible to design a method for capturing images from different positions, as the relative position change can be measured. The presented system can estimate relative distance to an object of unknown size with good precision. The manipulator applied in the presented work is the SeaArm-2, a fully electric underwater small modular manipulator. The manipulator is unique in its integrated monocular camera in the end-effector module, and its design facilitates the use of different end-effector tools. The camera is used for supervision, object detection, and tracking. The distance estimator was validated in a laboratory setting through autonomous grasping experiments. The manipulator was able to search for and find, estimate the relative distance of, grasp, and retrieve the relevant object in 12 out of 12 trials.

查看原文本刊更多论文

基于新型距离估计器的自主抓取

本文介绍了一种基于单目视觉的水下自主抓取距离估计器。该方法同样适用于上层甲板抓取作业。该估计器是针对在抓取器附近放置单目摄像机的机器人机械手而开发的。事实上，相机是附着在抓手附近，使得它有可能设计一种方法来捕捉图像从不同的位置，因为相对位置的变化可以测量。该系统能够以较好的精度估计未知大小物体的相对距离。本文所应用的机械臂是一种全电动水下小型模块化机械臂SeaArm-2。该机械手的独特之处在于其在末端执行器模块中集成了单目摄像头，其设计方便了不同末端执行器工具的使用。摄像机用于监控、目标检测和跟踪。在实验室环境下，通过自主抓取实验对距离估计器进行了验证。在12次试验中，该机械手能够在12次试验中搜索和发现、估计相对距离、抓取和检索相关物体。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

International Journal of Mechanical Engineering and Robotics Research Engineering-Mechanical Engineering

CiteScore

2.80

自引率

0.00%

发文量

期刊介绍： International Journal of Mechanical Engineering and Robotics Research. IJMERR is a scholarly peer-reviewed international scientific journal published bimonthly, focusing on theories, systems, methods, algorithms and applications in mechanical engineering and robotics. It provides a high profile, leading edge forum for academic researchers, industrial professionals, engineers, consultants, managers, educators and policy makers working in the field to contribute and disseminate innovative new work on Mechanical Engineering and Robotics Research.