A practical tactile exploration with enhanced sampling for fast shape estimation

IF 5.2 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement Pub Date : 2025-03-26 DOI:10.1016/j.measurement.2025.117350

Feng Luan , Chengjin Wang , Zhipeng Wang , Jiguang Yue , Yanmin Zhou , Bin He

{"title":"A practical tactile exploration with enhanced sampling for fast shape estimation","authors":"Feng Luan , Chengjin Wang , Zhipeng Wang , Jiguang Yue , Yanmin Zhou , Bin He","doi":"10.1016/j.measurement.2025.117350","DOIUrl":null,"url":null,"abstract":"<div><div>Automated measurement and three-dimensional reconstruction help robots to understand complex environments. In situations of darkness or smoke, tactile perception is pivotal. Here, we propose a tactile exploration method to estimate the shape of objects with Gaussian process implicit surfaces (GPIS), especially considering the sparsity and unevenness nature of tactile data. A data enhancement method is employed to improve the data integrity of tactile exploration for the implicit surface construction, which combines the external and internal position information of the object. In addition, a practical sampling strategy is constructed to improve the reconstruction efficiency limited by the nonuniform distribution of tactile data. This strategy considers the local data sparsity, shape uncertainty, local shape curvature, and exploration path cost for systematic and comprehensive data collection. Simulation and experiment results show that our method improves the model accuracy by 35.03% in the Hausdorff distance with the same sampling steps compared to other methods.</div><div>© 2017 Published by Elsevier.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"252 ","pages":"Article 117350"},"PeriodicalIF":5.2000,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Measurement","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0263224125007092","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, MULTIDISCIPLINARY","Score":null,"Total":0}

引用次数: 0

Abstract

Automated measurement and three-dimensional reconstruction help robots to understand complex environments. In situations of darkness or smoke, tactile perception is pivotal. Here, we propose a tactile exploration method to estimate the shape of objects with Gaussian process implicit surfaces (GPIS), especially considering the sparsity and unevenness nature of tactile data. A data enhancement method is employed to improve the data integrity of tactile exploration for the implicit surface construction, which combines the external and internal position information of the object. In addition, a practical sampling strategy is constructed to improve the reconstruction efficiency limited by the nonuniform distribution of tactile data. This strategy considers the local data sparsity, shape uncertainty, local shape curvature, and exploration path cost for systematic and comprehensive data collection. Simulation and experiment results show that our method improves the model accuracy by 35.03% in the Hausdorff distance with the same sampling steps compared to other methods.

查看原文本刊更多论文

求助全文

约1分钟内获得全文求助全文

来源期刊

Measurement 工程技术-工程：综合

CiteScore

10.20

自引率

12.50%

发文量

1589

审稿时长

12.1 months

期刊介绍： Contributions are invited on novel achievements in all fields of measurement and instrumentation science and technology. Authors are encouraged to submit novel material, whose ultimate goal is an advancement in the state of the art of: measurement and metrology fundamentals, sensors, measurement instruments, measurement and estimation techniques, measurement data processing and fusion algorithms, evaluation procedures and methodologies for plants and industrial processes, performance analysis of systems, processes and algorithms, mathematical models for measurement-oriented purposes, distributed measurement systems in a connected world.