Measurement最新文献_第9页

FBG-based 3D micro-force sensor integrated into ophthalmic microsurgical robotic forceps 基于fbg的三维微力传感器集成到眼科显微外科机器人钳中

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-04-28 Epub Date: 2026-02-24 DOI: 10.1016/j.measurement.2026.120909

Ning Wang , Xiaodong Zhang , Mingyang Li , Agostino Stilli , Sophia Bano , Danail Stoyanov

Posterior segment ophthalmic surgery requires precise manipulation of handheld instruments while avoiding injury to delicate fundus tissues. In procedures such as internal limiting membrane (ILM) peeling, even experienced clinicians using dedicated ophthalmic surgical robots struggle to maintain appropriate contact force without inducing tissue damage. This difficulty arises primarily from the absence of precise three-dimensional (3D) intraoperative force measurements. In this paper, we design a rhombus-shaped Force-sensitive Mechanism (FSM) that amplifies the FBG strain under axial loading. Guided by an intraoperative force analysis of surgical forceps, we further propose a parameter optimization method that improves sensor resolution and accuracy. Moreover, by evenly distributing four FBGs around the outer tube of the forceps, we achieve accurate measurement of intraoperative lateral forces and implement a robust and computationally efficient temperature compensation strategy. We also construct a strain-transfer model comprising the “FBG–adhesive layer–deformable body”, and theoretically analyze the strain-transfer rate from the deformable bodies to the FBGs along with its governing factors. Experiments show that optimization increases the strain at FBG5 by 94.32% relative to the unoptimized design. The sensor attains a sensitivity of 0.45 pm/mN and a resolution of 0.05 mN over a 5–25 mN range, with hysteresis and repeatability errors of 3.4% and 3.3%, respectively. In ILM-simulant peeling experiments, the sensor accurately captures both lateral and axial contact forces of the forceps, providing actionable force information for robotic-assisted fundus surgery.

眼后段手术需要精确操作手持式器械，同时避免损伤脆弱的眼底组织。在诸如内限制膜（ILM）剥离的手术中，即使是使用专用眼科手术机器人的经验丰富的临床医生也很难在不引起组织损伤的情况下保持适当的接触力。这种困难主要是由于缺乏精确的三维（3D）术中力测量。在本文中，我们设计了一个菱形的力敏感机构（FSM）来放大光纤光栅在轴向载荷下的应变。在手术钳术中受力分析的指导下，我们进一步提出了一种提高传感器分辨率和精度的参数优化方法。此外，通过在钳外管周围均匀分布四个fbg，我们实现了术中侧力的精确测量，并实现了鲁棒性和计算效率高的温度补偿策略。建立了由“fbg -粘接层-变形体”组成的应变传递模型，从理论上分析了变形体向fbg的应变传递速率及其影响因素。实验表明，优化后的FBG5处应变比未优化时提高了94.32%。该传感器在5-25 mN范围内的灵敏度为0.45 pm/mN，分辨率为0.05 mN，滞后和重复性误差分别为3.4%和3.3%。在ilm模拟剥离实验中，传感器准确捕获钳的横向和轴向接触力，为机器人辅助眼底手术提供可操作的力信息。

{"title":"FBG-based 3D micro-force sensor integrated into ophthalmic microsurgical robotic forceps","authors":"Ning Wang , Xiaodong Zhang , Mingyang Li , Agostino Stilli , Sophia Bano , Danail Stoyanov","doi":"10.1016/j.measurement.2026.120909","DOIUrl":"10.1016/j.measurement.2026.120909","url":null,"abstract":"<div><div>Posterior segment ophthalmic surgery requires precise manipulation of handheld instruments while avoiding injury to delicate fundus tissues. In procedures such as internal limiting membrane (ILM) peeling, even experienced clinicians using dedicated ophthalmic surgical robots struggle to maintain appropriate contact force without inducing tissue damage. This difficulty arises primarily from the absence of precise three-dimensional (3D) intraoperative force measurements. In this paper, we design a rhombus-shaped Force-sensitive Mechanism (FSM) that amplifies the FBG strain under axial loading. Guided by an intraoperative force analysis of surgical forceps, we further propose a parameter optimization method that improves sensor resolution and accuracy. Moreover, by evenly distributing four FBGs around the outer tube of the forceps, we achieve accurate measurement of intraoperative lateral forces and implement a robust and computationally efficient temperature compensation strategy. We also construct a strain-transfer model comprising the “FBG–adhesive layer–deformable body”, and theoretically analyze the strain-transfer rate from the deformable bodies to the FBGs along with its governing factors. Experiments show that optimization increases the strain at FBG5 by 94.32% relative to the unoptimized design. The sensor attains a sensitivity of 0.45 pm/mN and a resolution of 0.05 mN over a 5–25 mN range, with hysteresis and repeatability errors of 3.4% and 3.3%, respectively. In ILM-simulant peeling experiments, the sensor accurately captures both lateral and axial contact forces of the forceps, providing actionable force information for robotic-assisted fundus surgery.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"271 ","pages":"Article 120909"},"PeriodicalIF":5.6,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147387327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Electromagnetic radiation (EMR) energy evolution law of loaded coal-rock and its fracture precursor identification strategy 载煤岩电磁辐射能量演化规律及破裂前兆识别策略

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-04-28 Epub Date: 2026-02-24 DOI: 10.1016/j.measurement.2026.120957

Hao Li , Yuanpeng Zhao , Xin Li , Zhen Yang , Qiaochu Zhang , Xuejiao Wang , Feng Feng

The EMR energy of loaded coal-rock changes with stress, which provides a way for non-contact early warning of coal-rock fracture. Based on the analysis of the energy conversion at the fracture, this paper uses experiment at different rates of 0.3 – 0.5 mm/min to study the evolution law of coal-rock’s EMR energy. Subsequently, a coal-rock fracture precursor identification strategy based on EMR energy characteristic factors is proposed. Finally, its effect and universality are verified. The results show that: the energy conversion due to friction at the coal-rock fracture is the source of the EMR energy, which belongs to the dissipated energy. When coal-rock is loaded, the superposition of EMR energy density is similar to the dissipated energy, both of which show a trend of “slow increase − rapid increase”, which reflects the characteristics of EMR energy more stably and effectively. The variation of the EMR energy of loaded coal-rock has three phases. When approaching fracture, the precursor of EMR energy is 16.6 s earlier than dissipated energy, which fundamentally ensures the early warning effect. The constructed EMR energy characteristic factor EF is related to the damage development rate, whose adjacent maxima can be used to identify fracture precursors of coal-rock. The effect of the proposed two-level identification and early warning strategy for fracture precursors is not affected by the loading rate or the type of coal-rock, and has the “deviation correction” performance, whose average effective early warning duration can reach 33.095 s. These results are beneficial to improving the ability to prevent coal and rock dynamic disasters.

加载煤岩的EMR能量随应力变化，为煤岩断裂的非接触预警提供了一种方法。在分析裂隙处能量转换的基础上，采用0.3 ~ 0.5 mm/min不同速率的实验，研究了煤岩EMR能量的演化规律。随后，提出了基于EMR能量特征因子的煤岩断裂前兆识别策略。最后，验证了该方法的有效性和普遍性。结果表明：煤岩裂隙处摩擦产生的能量转换是EMR能量的来源，属于耗散能；煤岩加载时，EMR能量密度叠加与耗散能量相似，均呈现“缓慢增加-快速增加”的趋势，更稳定有效地反映了EMR能量的特征。煤岩加载后EMR能量的变化有三个阶段。接近裂缝时，EMR能量的前兆比耗散能量早16.6 s，从根本上保证了预警效果。构造的EMR能量特征因子EF与损伤发育速率有关，其相邻最大值可用于识别煤岩断裂前兆。所提出的两级裂缝前兆识别预警策略的效果不受加载速率和煤岩类型的影响，具有“纠偏”性能，平均有效预警持续时间可达33.095 s。这些结果有利于提高煤岩动力灾害的防治能力。

{"title":"Electromagnetic radiation (EMR) energy evolution law of loaded coal-rock and its fracture precursor identification strategy","authors":"Hao Li , Yuanpeng Zhao , Xin Li , Zhen Yang , Qiaochu Zhang , Xuejiao Wang , Feng Feng","doi":"10.1016/j.measurement.2026.120957","DOIUrl":"10.1016/j.measurement.2026.120957","url":null,"abstract":"<div><div>The EMR energy of loaded coal-rock changes with stress, which provides a way for non-contact early warning of coal-rock fracture. Based on the analysis of the energy conversion at the fracture, this paper uses experiment at different rates of 0.3 – 0.5 mm/min to study the evolution law of coal-rock’s EMR energy. Subsequently, a coal-rock fracture precursor identification strategy based on EMR energy characteristic factors is proposed. Finally, its effect and universality are verified. The results show that: the energy conversion due to friction at the coal-rock fracture is the source of the EMR energy, which belongs to the dissipated energy. When coal-rock is loaded, the superposition of EMR energy density is similar to the dissipated energy, both of which show a trend of “slow increase − rapid increase”, which reflects the characteristics of EMR energy more stably and effectively. The variation of the EMR energy of loaded coal-rock has three phases. When approaching fracture, the precursor of EMR energy is 16.6 s earlier than dissipated energy, which fundamentally ensures the early warning effect. The constructed EMR energy characteristic factor <em>EF</em> is related to the damage development rate, whose adjacent maxima can be used to identify fracture precursors of coal-rock. The effect of the proposed two-level identification and early warning strategy for fracture precursors is not affected by the loading rate or the type of coal-rock, and has the “deviation correction” performance, whose average effective early warning duration can reach 33.095 s. These results are beneficial to improving the ability to prevent coal and rock dynamic disasters.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"271 ","pages":"Article 120957"},"PeriodicalIF":5.6,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147387334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

3DASE a novel three dimensional strain measurement technique based on particle tracking 3DASE是一种新型的基于粒子跟踪的三维应变测量技术

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-04-28 Epub Date: 2026-02-24 DOI: 10.1016/j.measurement.2026.120864

Thomas Pritchard, Megan McNamee, Raoul van Loon, Hari Arora

This study introduces a novel methodology for reconstructing full-field three dimensional strain and displacement fields in highly deformable soft solids, leveraging advanced particle tracking techniques. The approach utilises three dimensional displacement data acquired through volumetric particle tracking, specifically the Shake the Box (STB) technique, originally developed for fluid flow analysis, to enable accurate measurement over larger volumes (up to 50 ml) during indenter loading. Displacement vectors are binned to a voxel grid using a Gaussian-weighted scheme, forming the nodes of a hexahedral finite element mesh from which strain fields are computed. Synthetic particle datasets were employed to verify accuracy and quantify error and bias, achieving root mean square errors as low as 1.2% under ideal conditions. Application to experimental datasets demonstrated the method’s ability to capture detailed strain and displacement fields. Comparison with finite element models derived via inverse parameter identification showed strong agreement, confirming the technique’s validity. This integrated method shows significant promise for characterising internal mechanics of soft materials and tissue phantoms where conventional techniques fall short.

本研究介绍了一种利用先进的颗粒跟踪技术重建高度可变形软固体中全场三维应变和位移场的新方法。该方法利用通过体积颗粒跟踪获得的三维位移数据，特别是最初为流体流动分析而开发的Shake The Box （STB）技术，可以在压头加载期间对更大体积（高达50毫升）进行精确测量。位移向量使用高斯加权格式被分割成体素网格，形成六面体有限元网格的节点，从而计算应变场。采用合成粒子数据集验证精度，量化误差和偏差，理想条件下均方根误差低至1.2%。对实验数据集的应用表明，该方法能够捕获详细的应变和位移场。与通过参数反辨识得到的有限元模型比较，结果吻合较好，证实了该方法的有效性。这种综合方法在表征软材料和组织幻影的内部力学方面显示出巨大的希望，而传统技术在这方面做得不够。

{"title":"3DASE a novel three dimensional strain measurement technique based on particle tracking","authors":"Thomas Pritchard, Megan McNamee, Raoul van Loon, Hari Arora","doi":"10.1016/j.measurement.2026.120864","DOIUrl":"10.1016/j.measurement.2026.120864","url":null,"abstract":"<div><div>This study introduces a novel methodology for reconstructing full-field three dimensional strain and displacement fields in highly deformable soft solids, leveraging advanced particle tracking techniques. The approach utilises three dimensional displacement data acquired through volumetric particle tracking, specifically the Shake the Box (STB) technique, originally developed for fluid flow analysis, to enable accurate measurement over larger volumes (up to 50 ml) during indenter loading. Displacement vectors are binned to a voxel grid using a Gaussian-weighted scheme, forming the nodes of a hexahedral finite element mesh from which strain fields are computed. Synthetic particle datasets were employed to verify accuracy and quantify error and bias, achieving root mean square errors as low as 1.2% under ideal conditions. Application to experimental datasets demonstrated the method’s ability to capture detailed strain and displacement fields. Comparison with finite element models derived via inverse parameter identification showed strong agreement, confirming the technique’s validity. This integrated method shows significant promise for characterising internal mechanics of soft materials and tissue phantoms where conventional techniques fall short.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"271 ","pages":"Article 120864"},"PeriodicalIF":5.6,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147387439","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Bi-LSTM neural network for optimizing the consistency of satellite clock bias data of beidou satellite navigation system 基于Bi-LSTM神经网络的北斗卫星导航系统卫星时钟偏差数据一致性优化

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-04-28 Epub Date: 2026-02-23 DOI: 10.1016/j.measurement.2026.120928

Wei Wang , Mingquan Lu , Yupu Wang , Chengpan Tang

The BeiDou-3 Navigation Satellite System (BDS-3) innovatively utilizes Two-Way Time Transfer (TWTT) and Inter-Satellite Link (ISL) observation data to determine satellite clock bias, significantly enhancing the accuracy of clock bias prediction compared to BDS-2. However, nanosecond-level inconsistency exists in TWTT and ISL data, and this may jeopardize the accuracy of clock bias predictions. Aiming to acquire better consistency and smoothness in clock bias data for prediction accuracy improvement, an error detection and reduction method using Bidirectional Long Short-Term Memory (Bi-LSTM) neural network is proposed. After processing with the Bi-LSTM neural network, fluctuations and discrete jumps are significantly mitigated, resulting in approximately 70% improvement in clock bias data consistency. Compared with other methods, the clock data processed by the Bi-LSTM neural network can achieve better frequency stability and fitting residuals. Finally, the processed clock data are used to generate clock parameters, indicating 60% accuracy improvement compared to the currently used clock prediction approach based on the non-processed clock data.

北斗三号卫星导航系统（BDS-3）创新地利用双向时间传输（TWTT）和星间链路（ISL）观测数据来确定卫星时钟偏差，与BDS-2相比，显著提高了时钟偏差预测的精度。然而，在TWTT和ISL数据中存在纳秒级的不一致性，这可能会危及时钟偏差预测的准确性。为了提高时钟偏差数据的一致性和平滑性，提高预测精度，提出了一种基于双向长短期记忆（Bi-LSTM）神经网络的误差检测与降准方法。经过Bi-LSTM神经网络处理后，波动和离散跳变得到显著缓解，时钟偏差数据一致性提高了约70%。与其他方法相比，经Bi-LSTM神经网络处理的时钟数据具有更好的频率稳定性和拟合残差。最后，使用处理后的时钟数据生成时钟参数，与目前使用的基于未处理时钟数据的时钟预测方法相比，精度提高了60%。

{"title":"Bi-LSTM neural network for optimizing the consistency of satellite clock bias data of beidou satellite navigation system","authors":"Wei Wang , Mingquan Lu , Yupu Wang , Chengpan Tang","doi":"10.1016/j.measurement.2026.120928","DOIUrl":"10.1016/j.measurement.2026.120928","url":null,"abstract":"<div><div>The BeiDou-3 Navigation Satellite System (BDS-3) innovatively utilizes Two-Way Time Transfer (TWTT) and Inter-Satellite Link (ISL) observation data to determine satellite clock bias, significantly enhancing the accuracy of clock bias prediction compared to BDS-2. However, nanosecond-level inconsistency exists in TWTT and ISL data, and this may jeopardize the accuracy of clock bias predictions. Aiming to acquire better consistency and smoothness in clock bias data for prediction accuracy improvement, an error detection and reduction method using Bidirectional Long Short-Term Memory (Bi-LSTM) neural network is proposed. After processing with the Bi-LSTM neural network, fluctuations and discrete jumps are significantly mitigated, resulting in approximately 70% improvement in clock bias data consistency. Compared with other methods, the clock data processed by the Bi-LSTM neural network can achieve better frequency stability and fitting residuals. Finally, the processed clock data are used to generate clock parameters, indicating 60% accuracy improvement compared to the currently used clock prediction approach based on the non-processed clock data.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"271 ","pages":"Article 120928"},"PeriodicalIF":5.6,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147387526","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Damped sinusoid fitting algorithm for parameter estimation of real-valued noisy damped sinusoids 阻尼正弦波拟合算法用于实值噪声阻尼正弦波的参数估计

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-04-28 Epub Date: 2026-02-28 DOI: 10.1016/j.measurement.2026.121007

Daniel Belega , Dario Petri

The paper proposes an algorithm for the estimation of the frequency, decay rate, amplitude, phase, and offset of real-valued noisy damped sinusoids. Initial values for the frequency and the decay rate are determined by means of the two-point interpolated discrete Fourier transform algorithm based on the Hann window. Then the linear least-squares approach is applied iteratively. The performances of the proposed algorithm are analysed and compared with those of two state-of-the-art algorithms by means of both simulation and experimental results.

本文提出了一种估计实值含噪阻尼正弦波的频率、衰减率、幅度、相位和偏移量的算法。频率和衰减率的初始值由基于Hann窗的两点插值离散傅里叶变换算法确定。然后迭代地应用线性最小二乘方法。通过仿真和实验结果，对所提算法的性能进行了分析和比较。

引用次数: 0

A novel visible and infrared image registration method for pig-body multi-feature evaluation 一种新的猪体多特征评价的可见光和红外图像配准方法

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-04-28 Epub Date: 2026-02-28 DOI: 10.1016/j.measurement.2026.121005

Zhen Zhong

The pig-body health evaluation has been a key concern to prevent the spread of swine fever. However, traditional animal health evaluation methods mostly rely on artificial, which can easily lead to cause zoonotic diseases. It is crucial for preventing the spread of swine fever by rapid and non-contact evaluation. To evaluate pig-body health by extracting their shape and temperature features, visible and infrared imaging are valuable for obtaining multi-source pig-body images from different modalities. However, due to excessive interference feature points in visible pig-body images and different resolutions of multi-source images, only considering the main orientation information of feature points in visible and infrared pig-body images can easily lead to spatial and structural information misalignment. To overcome this issue, a novel structural feature representation method of visible and infrared pig-body image is proposed by adjacent self-similarity and named Gabor-based Adjacent Self-Similarity (GASS). Moreover, a novel multi-source image registration method is represented for pig-body multi-feature evaluation. Firstly, the GASS features are obtained by Gabor filters with variable scales and optimized offset mean filtering. Then, the minimum self-similarity maps (SSM) are extracted based on GASS features, which have robustness to resolution differences. Finally, the accurate matches are achieved by nearest-neighbor distance ratio (NNDR) matching and the fast sample consensus (FSC) algorithm. Experimental results show that the proposed visible and infrared pig-body image registration method has better illumination invariance, and is capable of achieving higher the number of accurate matches than current registration algorithms.

猪体健康评估一直是预防猪瘟传播的关键问题。然而，传统的动物健康评价方法大多依靠人工，容易导致人畜共患疾病。通过快速和非接触评估对预防猪瘟的传播至关重要。为了通过提取猪体的形状和温度特征来评估猪体的健康状况，可见光和红外成像对于获得不同模态的多源猪体图像是有价值的。然而，由于可见猪体图像中干扰特征点过多，多源图像分辨率不同，仅考虑可见和红外猪体图像中特征点的主要方位信息，容易导致空间和结构信息错位。为了克服这一问题，提出了一种基于相邻自相似度的猪体可见光和红外图像结构特征表示方法，并命名为基于gabor的相邻自相似度（GASS）。此外，提出了一种新的猪体多特征评价的多源图像配准方法。首先，利用变尺度Gabor滤波器和优化偏移平均滤波获得GASS特征；然后，基于GASS特征提取对分辨率差异具有鲁棒性的最小自相似图（SSM）；最后，采用最近邻距离比（NNDR）匹配和快速样本一致性（FSC）算法实现精确匹配。实验结果表明，所提出的可见和红外猪体图像配准方法具有较好的光照不变性，且能实现比现有配准算法更高的精确匹配次数。

{"title":"A novel visible and infrared image registration method for pig-body multi-feature evaluation","authors":"Zhen Zhong","doi":"10.1016/j.measurement.2026.121005","DOIUrl":"10.1016/j.measurement.2026.121005","url":null,"abstract":"<div><div>The pig-body health evaluation has been a key concern to prevent the spread of swine fever. However, traditional animal health evaluation methods mostly rely on artificial, which can easily lead to cause zoonotic diseases. It is crucial for preventing the spread of swine fever by rapid and non-contact evaluation. To evaluate pig-body health by extracting their shape and temperature features, visible and infrared imaging are valuable for obtaining multi-source pig-body images from different modalities. However, due to excessive interference feature points in visible pig-body images and different resolutions of multi-source images, only considering the main orientation information of feature points in visible and infrared pig-body images can easily lead to spatial and structural information misalignment. To overcome this issue, a novel structural feature representation method of visible and infrared pig-body image is proposed by adjacent self-similarity and named Gabor-based Adjacent Self-Similarity (<em>GASS</em>). Moreover, a novel multi-source image registration method is represented for pig-body multi-feature evaluation. Firstly, the <em>GASS</em> features are obtained by <em>Gabor</em> filters with variable scales and optimized offset mean filtering. Then, the minimum self-similarity maps (<em>SSM</em>) are extracted based on <em>GASS</em> features, which have robustness to resolution differences. Finally, the accurate matches are achieved by nearest-neighbor distance ratio (<em>NNDR</em>) matching and the fast sample consensus (<em>FSC</em>) algorithm. Experimental results show that the proposed visible and infrared pig-body image registration method has better illumination invariance, and is capable of achieving higher the number of accurate matches than current registration algorithms.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"271 ","pages":"Article 121005"},"PeriodicalIF":5.6,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147387959","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A robust framework for PDR/magnetic fingerprint/GNSS integrated navigation with boundary constraints 基于边界约束的PDR/磁指纹/GNSS组合导航鲁棒框架

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-04-28 Epub Date: 2026-02-24 DOI: 10.1016/j.measurement.2026.120947

Wei Liao, Shang Junna, Zhengtao Wu, Lu Chundong, Su Mingkun

The lack of robustness in multi-source integrated navigation systems under complex environments has become a widely discussed issue in recent years, especially in schemes involving magnetic fingerprinting as a key data source. To address this challenge, this paper proposes a robust optimization framework for integrated pedestrian dead reckoning (PDR), magnetic fingerprint vectors, and GNSS navigation. The framework introduces boundary constraints based on the inherent noise characteristics of sensors, combined with the hyperbolic tangent activation function, to ensure linear factor responses within constrained regions and avoid local extrema during iterative optimization. A robust loss function is applied to magnetometer and GNSS factors, effectively mitigating the impact of outliers. Additionally, a sliding window strategy with momentum-based optimization is employed to enhance performance in dynamic and complex scenarios. Simulation and field experiments demonstrate that the proposed framework achieves both high accuracy and strong robustness, with reductions in positioning errors of up to 46% compared to conventional factor graph approaches, meeting real-time positioning requirements even under satellite occlusion and electromagnetic interference.

近年来，复杂环境下多源组合导航系统鲁棒性不足的问题已成为人们广泛讨论的问题，特别是在以磁指纹为关键数据源的方案中。为了解决这一挑战，本文提出了一种集成行人航位推算（PDR）、磁指纹矢量和GNSS导航的鲁棒优化框架。该框架基于传感器固有噪声特性引入边界约束，结合双曲正切激活函数，保证约束区域内因子响应线性化，避免迭代优化过程中出现局部极值。对磁强计和GNSS因子采用鲁棒损失函数，有效地减轻了异常值的影响。此外，采用基于动量优化的滑动窗口策略来提高动态和复杂场景下的性能。仿真和现场实验表明，该框架具有较高的精度和较强的鲁棒性，与传统的因子图方法相比，定位误差降低了46%，即使在卫星遮挡和电磁干扰下也能满足实时定位要求。

{"title":"A robust framework for PDR/magnetic fingerprint/GNSS integrated navigation with boundary constraints","authors":"Wei Liao, Shang Junna, Zhengtao Wu, Lu Chundong, Su Mingkun","doi":"10.1016/j.measurement.2026.120947","DOIUrl":"10.1016/j.measurement.2026.120947","url":null,"abstract":"<div><div>The lack of robustness in multi-source integrated navigation systems under complex environments has become a widely discussed issue in recent years, especially in schemes involving magnetic fingerprinting as a key data source. To address this challenge, this paper proposes a robust optimization framework for integrated pedestrian dead reckoning (PDR), magnetic fingerprint vectors, and GNSS navigation. The framework introduces boundary constraints based on the inherent noise characteristics of sensors, combined with the hyperbolic tangent activation function, to ensure linear factor responses within constrained regions and avoid local extrema during iterative optimization. A robust loss function is applied to magnetometer and GNSS factors, effectively mitigating the impact of outliers. Additionally, a sliding window strategy with momentum-based optimization is employed to enhance performance in dynamic and complex scenarios. Simulation and field experiments demonstrate that the proposed framework achieves both high accuracy and strong robustness, with reductions in positioning errors of up to 46% compared to conventional factor graph approaches, meeting real-time positioning requirements even under satellite occlusion and electromagnetic interference.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"271 ","pages":"Article 120947"},"PeriodicalIF":5.6,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147387973","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Development of a GOSRM model for real-time surface roughness monitoring in face milling of Inconel 718 superalloy 基于GOSRM模型的Inconel 718高温合金面铣削表面粗糙度实时监测研究

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-04-28 Epub Date: 2026-02-21 DOI: 10.1016/j.measurement.2026.120904

Sarat Babu Mulpur

This study presents a Grey-based Online Surface Roughness Monitoring (GOSRM) model for real-time prediction of machined surface roughness during face milling of Inconel 718 superalloy. Unlike conventional post-process inspection approaches, the proposed framework integrates optimized machining parameters, surface texture imaging, and grey system theory to enable continuous monitoring without interrupting the machining process. Machining parameters were first optimized using response surface methodology to minimize surface roughness, followed by surface texture analysis under optimal conditions. Texture features were extracted using Gray-Level Co-occurrence Matrix (GLCM) and Gabor Wavelet Transform techniques and employed as inputs to an GOSRM model. The developed GOSRM model demonstrated strong predictive capability by achieving an RMSE of 0.2004 and a prediction error of 7.87%, confirming its suitability for real-time surface roughness estimation. In comparison, Multiple Linear Regression (MLR) exhibited substantially higher prediction errors with an RMSE of 0.5442 and a prediction error of 21.61%. The Artificial Neural Network (ANN) also showed inferior performance yielding a higher RMSE of 0.8415 and a prediction error of 33.41%, indicating reduced accuracy and robustness under limited data machining conditions. The proposed approach offers a computationally efficient and data-light solution for online surface quality monitoring and can be extended to other materials and machining processes with appropriate retraining and calibration.

提出了一种基于灰色的在线表面粗糙度监测（GOSRM）模型，用于实时预测Inconel 718高温合金端面铣削过程中的加工表面粗糙度。与传统的加工后检测方法不同，该框架集成了优化的加工参数、表面纹理成像和灰色系统理论，可以在不中断加工过程的情况下进行连续监测。首先利用响应面法优化加工参数，使表面粗糙度最小，然后在最优条件下进行表面织构分析。利用灰度共生矩阵（GLCM）和Gabor小波变换技术提取纹理特征，并将其作为GOSRM模型的输入。建立的GOSRM模型具有较强的预测能力，RMSE为0.2004，预测误差为7.87%，可用于实时的表面粗糙度估计。而多元线性回归（Multiple Linear Regression， MLR）的预测误差更大，RMSE为0.5442，预测误差为21.61%。人工神经网络（ANN）也表现出较差的性能，RMSE为0.8415，预测误差为33.41%，表明在有限的数据加工条件下精度和鲁棒性降低。该方法为在线表面质量监测提供了一种计算效率高、数据量少的解决方案，并且可以通过适当的再培训和校准扩展到其他材料和加工过程中。

{"title":"Development of a GOSRM model for real-time surface roughness monitoring in face milling of Inconel 718 superalloy","authors":"Sarat Babu Mulpur","doi":"10.1016/j.measurement.2026.120904","DOIUrl":"10.1016/j.measurement.2026.120904","url":null,"abstract":"<div><div>This study presents a Grey-based Online Surface Roughness Monitoring (GOSRM) model for real-time prediction of machined surface roughness during face milling of Inconel 718 superalloy. Unlike conventional post-process inspection approaches, the proposed framework integrates optimized machining parameters, surface texture imaging, and grey system theory to enable continuous monitoring without interrupting the machining process. Machining parameters were first optimized using response surface methodology to minimize surface roughness, followed by surface texture analysis under optimal conditions. Texture features were extracted using Gray-Level Co-occurrence Matrix (GLCM) and Gabor Wavelet Transform techniques and employed as inputs to an GOSRM model. The developed GOSRM model demonstrated strong predictive capability by achieving an RMSE of 0.2004 and a prediction error of 7.87%, confirming its suitability for real-time surface roughness estimation. In comparison, Multiple Linear Regression (MLR) exhibited substantially higher prediction errors with an RMSE of 0.5442 and a prediction error of 21.61%. The Artificial Neural Network (ANN) also showed inferior performance yielding a higher RMSE of 0.8415 and a prediction error of 33.41%, indicating reduced accuracy and robustness under limited data machining conditions. The proposed approach offers a computationally efficient and data-light solution for online surface quality monitoring and can be extended to other materials and machining processes with appropriate retraining and calibration.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"271 ","pages":"Article 120904"},"PeriodicalIF":5.6,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147387615","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Importance of temporal encoding in data-driven estimation of downhole variables 时间编码在井下变量数据驱动估计中的重要性

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-04-28 Epub Date: 2026-02-21 DOI: 10.1016/j.measurement.2026.120915

Prince Okoli , Shengnan Chen , Roman J. Shor , Henry Leung

Estimating downhole torque on bit (TOB) from surface measurements has been explored in literature as a promising way to potentially provide real-time insight into torsional vibration without costly wired telemetry. However, many data-driven estimators in literature ignore temporal dependencies, even though drilling dynamics suggests that nonlinear bit-rock interaction makes TOB intrinsically history-dependent, so estimators that encode temporal context should better capture the underlying physics. We investigate temporal encoding via feature-based and model-based approaches for soft-sensing TOB, using 1 Hz surface data (282,601 records) paired with 50 Hz downhole TOB (14,130,001 records). We benchmark non-temporal baselines (MLP, SVM, RF, GBDT), feature-based temporal encoding (lagged features), and model-based temporal encoding (LSTM stacks, encoder-decoder LSTM with/without cross-attention, and Transformers). Model-based temporal encoding consistently outperforms alternatives: A unidirectional LSTM stack achieves the best median mean absolute error (MAE) of 475.8 lbf·ft for model-based temporal encoding, versus 550.7 lbf·ft for feature-based (MLP with lagged features) and 643.8 lbf·ft without temporal encoding (MLP); R² shows matching gains. We further present a connection scheme for the encoder-decoder used in exogenous autoregression that lowers median MAE from 584.0 to 520.9. We show that the most informative input history is short (≈3–12 s), which is in line with history dependence expected for regenerative effects of bit-rock interaction and the transmission latency due to speed of mechanical energy.

根据地面测量数据估算井下扭矩（TOB）是一种很有前景的方法，可以在不使用昂贵的有线遥测技术的情况下实时了解扭转振动。然而，文献中许多数据驱动的估计方法忽略了时间依赖性，即使钻井动力学表明非线性钻头-岩石相互作用使TOB本质上依赖于历史，因此编码时间背景的估计方法应该更好地捕捉底层物理。我们使用1 Hz地面数据（282,601条记录）和50 Hz井下TOB（14,130,001条记录），通过基于特征和基于模型的方法研究了软测量TOB的时间编码。我们对非时间基线（MLP、SVM、RF、GBDT）、基于特征的时间编码（滞后特征）和基于模型的时间编码（LSTM堆栈、有/没有交叉注意的编码器-解码器LSTM和变压器）进行了基准测试。基于模型的时间编码始终优于其他方法：基于模型的时间编码，单向LSTM堆栈的最佳中位数平均绝对误差（MAE）为475.8 lbf·ft，而基于特征（具有滞后特征的MLP）的中位数平均绝对误差为550.7 lbf·ft，而没有时间编码（MLP）的中位数平均绝对误差为643.8 lbf·ft；R2显示匹配增益。我们进一步提出了用于外生自回归的编码器-解码器的连接方案，该方案将中位数MAE从584.0降低到520.9。我们发现，信息量最大的输入历史很短（≈3-12 s），这符合比特-岩石相互作用再生效应的历史依赖性和机械能速度导致的传输延迟。

{"title":"Importance of temporal encoding in data-driven estimation of downhole variables","authors":"Prince Okoli , Shengnan Chen , Roman J. Shor , Henry Leung","doi":"10.1016/j.measurement.2026.120915","DOIUrl":"10.1016/j.measurement.2026.120915","url":null,"abstract":"<div><div>Estimating downhole torque on bit (TOB) from surface measurements has been explored in literature as a promising way to potentially provide real-time insight into torsional vibration without costly wired telemetry. However, many data-driven estimators in literature ignore temporal dependencies, even though drilling dynamics suggests that nonlinear bit-rock interaction makes TOB intrinsically history-dependent, so estimators that encode temporal context should better capture the underlying physics. We investigate temporal encoding via feature-based and model-based approaches for soft-sensing TOB, using 1 Hz surface data (282,601 records) paired with 50 Hz downhole TOB (14,130,001 records). We benchmark non-temporal baselines (MLP, SVM, RF, GBDT), feature-based temporal encoding (lagged features), and model-based temporal encoding (LSTM stacks, encoder-decoder LSTM with/without cross-attention, and Transformers). Model-based temporal encoding consistently outperforms alternatives: A unidirectional LSTM stack achieves the best median mean absolute error (MAE) of 475.8 lbf·ft for model-based temporal encoding, versus 550.7 lbf·ft for feature-based (MLP with lagged features) and 643.8 lbf·ft without temporal encoding (MLP); R<sup>2</sup> shows matching gains. We further present a connection scheme for the encoder-decoder used in exogenous autoregression that lowers median MAE from 584.0 to 520.9. We show that the most informative input history is short (≈3–12 s), which is in line with history dependence expected for regenerative effects of bit-rock interaction and the transmission latency due to speed of mechanical energy.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"271 ","pages":"Article 120915"},"PeriodicalIF":5.6,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147387666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Uncertainty analysis of encoder-based method for the calibration design of on-board vehicle speed meters 基于编码器的车载速度计标定设计方法的不确定度分析

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-04-28 Epub Date: 2026-02-18 DOI: 10.1016/j.measurement.2026.120810

Federico Carere , Luca Podestà , Gianfranco Miele , Silvia Sangiovanni , Marco Laracca

Vehicle speed meters have an important role in various application fields such as monitoring and controlling speed limit for safety purposes, evaluating driver and vehicle performance, managing the traffic in real-time, etc. Compliance with ISO/IEC 17,025 and the ILAC (International Laboratory Accreditation Cooperation) policy on measurement traceability is required in calibration procedures to guarantee the quality of traceability of the results obtained by speed meters. The scientific literature, as well as the activity of some calibration centers, reports a variety of methodologies for calibration. Designing the most suitable calibration procedure for a specific category of instrument is not an easy task due to the high number of parameters affecting the calibration uncertainty.

In this context, the paper proposes a design tool for encoder-based methods for the calibration of on-board vehicle speed meters. After the definition of all the measurement uncertainty contributions involving the calibration process, a sensitivity analysis was carried out to assess the applicability and limitations of the selected calibration method for speeds up to 300 km/h. Afterwards, a design tool is proposed to enable the optimal design of the calibration method aiming to find the best trade-off between technical requirements (calibration uncertainty, speed range) and economic aspects.

车速表在安全监控限速、评价驾驶员和车辆性能、实时交通管理等应用领域具有重要作用。校准程序要求符合ISO/IEC 17025和ILAC（国际实验室认可合作组织）测量可追溯性政策，以保证速度计获得的结果可追溯性的质量。科学文献，以及一些校准中心的活动，报告了各种校准方法。由于影响校准不确定度的参数很多，为特定类别的仪器设计最合适的校准程序不是一件容易的事情。在此背景下，本文提出了一种基于编码器的车载速度计标定方法的设计工具。在定义了涉及校准过程的所有测量不确定度贡献后，进行了灵敏度分析，以评估所选校准方法在300 km/h以下速度下的适用性和局限性。然后，提出了一种设计工具来实现校准方法的优化设计，旨在找到技术要求（校准不确定度，速度范围）和经济方面的最佳权衡。

{"title":"Uncertainty analysis of encoder-based method for the calibration design of on-board vehicle speed meters","authors":"Federico Carere , Luca Podestà , Gianfranco Miele , Silvia Sangiovanni , Marco Laracca","doi":"10.1016/j.measurement.2026.120810","DOIUrl":"10.1016/j.measurement.2026.120810","url":null,"abstract":"<div><div>Vehicle speed meters have an important role in various application fields such as monitoring and controlling speed limit for safety purposes, evaluating driver and vehicle performance, managing the traffic in real-time, etc. Compliance with ISO/IEC 17,025 and the ILAC (International Laboratory Accreditation Cooperation) policy on measurement traceability is required in calibration procedures to guarantee the quality of traceability of the results obtained by speed meters. The scientific literature, as well as the activity of some calibration centers, reports a variety of methodologies for calibration. Designing the most suitable calibration procedure for a specific category of instrument is not an easy task due to the high number of parameters affecting the calibration uncertainty.</div><div>In this context, the paper proposes a design tool for encoder-based methods for the calibration of on-board vehicle speed meters. After the definition of all the measurement uncertainty contributions involving the calibration process, a sensitivity analysis was carried out to assess the applicability and limitations of the selected calibration method for speeds up to 300 km/h. Afterwards, a design tool is proposed to enable the optimal design of the calibration method aiming to find the best trade-off between technical requirements (calibration uncertainty, speed range) and economic aspects.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"271 ","pages":"Article 120810"},"PeriodicalIF":5.6,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147387882","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0