最新刊期
卷 33 ， 期 23 ， 2025
- “在光谱仪器领域，衍射光栅研究取得新进展，专家总结了宽波段光栅的四类特点，为后续研究提供依据。”摘要：Diffraction gratings serve as the core optical components in spectroscopic instrumentation. Broadband gratings， in contrast to conventional gratings， fulfill spectral analysis requirements across extended wavelength ranges. This paper reviews four principal types of broadband gratings： complex groove-shaped gratings， zoned gratings， intermediate-step gratings， and multilayer dielectric gratings. Their respective structures and operational principles are introduced， accompanied by a summary of recent research progress. Key characteristics are delineated as follows： complex groove-shaped gratings exhibit high spectral continuity and stable diffraction efficiency， yet their performance is highly sensitive to fabrication precision； zoned gratings offer considerable design flexibility and benefit from mature processing techniques， but suffer from phase mismatch， diffraction field disparities between sub-zones， and susceptibility to stray light； intermediate-step gratings enable high-resolution response over broad wavelength ranges at high diffraction orders， though their accuracy is constrained by manufacturing equipment and process complexity； multilayer dielectric gratings provide adaptable structural design and strong environmental adaptability， but impose stringent demands on material selection and multilayer film thickness control. Finally， future development directions for broadband diffraction gratings are outlined to inform subsequent research.关键词：Spectrometer;broadband grating;diffraction grating;blaze grating;echelle grating64|18|0更新时间：2025-12-19
- “在动态目标高精度空间定位领域，专家提出了基于分步光束法平差的多目视觉测量系统重建算法，优化系统参数，并通过FPGA平台实现快速运行。实验结果表明，该算法的平均空间重投影误差小于68 μm，优于直接线性变换重建算法与传统光束法平差重建算法的精度。该FPGA硬件架构方案可同时对4路高清图像数据实现约40 frame/s的处理速度，满足动态目标空间位置的实时测量需求。”摘要：Due to its high measurement accuracy， rapid acquisition speed， large measurable range， and non-contact characteristics， the multi-camera vision measurement system has been widely applied to high-precision spatial positioning of dynamic targets in aerospace， automotive， and related fields. However， existing multi-camera vision measurement systems based on bundle adjustment cannot simultaneously satisfy the stringent requirements of high accuracy and high speed in three-dimensional reconstruction of dynamic target positions. In this study， a reconstruction algorithm for multi-camera vision measurement systems based on stepwise bundle adjustment is proposed. System parameters with different orders of magnitude are optimized sequentially， and the rapid execution of the reconstruction algorithm is implemented on an field-programmable gate array （FPGA） platform. Experimental results demonstrate that the proposed stepwise bundle adjustment reconstruction algorithm achieves an average spatial reprojection error of less than 68 μm， outperforming both the direct linear transformation reconstruction algorithm and the conventional bundle adjustment reconstruction algorithm. The designed FPGA hardware architecture attains a processing speed of approximately 40 frame/s for four channels of high-definition image data （2 048×2 048×8 bit） simultaneously， thereby meeting the real-time， high-precision measurement requirements for the spatial positioning of dynamic targets.关键词：multi-camera vision measurement;stepwise bundle adjustment;spatial reprojection error;Field-Programmable Gate Array （FPGA）30|9|0更新时间：2025-12-19
-
Absolute interferometric calibration of liquid crystal spatial light modulators using dual gratings AI导读
“在光场调控领域，专家利用衍射光学元件和U-Net神经网络，成功实现液晶空间光调制器高精度相位标定，为精确应用提供可靠方案。”摘要：Liquid-crystal spatial light modulators （LC-SLMs） hold great promise for applications such as light-field modulation. However， the phase of uncalibrated SLM outputs is typically irregular， and the accuracy of phase modulation is difficult to control due to experimental and environmental perturbations. Accurate testing and calibration of LC-SLMs is therefore essential to ensure stable and precise performance in practical applications. In this study， a diffractive optical element （DOE） combining a grating and a diffractive lens is designed based on Fermat's principle and the Huygens–Fresnel principle. A pair of gratings is employed to generate three pairs of focal points； interference fringes are produced by the overlapping central foci， while the remaining beams form an absolute spatial coordinate system that determines the reference point for fringe motion. A U-Net neural network is applied to preprocess the interference fringe images， after which the key centroid positions are extracted using contour detection and a center-of-mass algorithm. The phase delay is then obtained by calculating the fringe displacement and period. A phase calibration system comprising a high-coherence light source， beam splitter， LC-SLM， beam-splitting grating， and CCD camera is constructed to perform the experiments. During the loading of different grayscale modulation sequences， the interference fringes exhibit clearly observable displacements as the grayscale values vary， enabling the successful extraction of the phase-grayscale response curve. The fitted results show high consistency with theoretical predictions， with a coefficient of determination R² of 0.985 1 and a root mean square error （RMSE） of 0.241 1 rad. Under fan-induced perturbations， the method still maintains R²>0.95， and the RMSE increases by only 0.14 rad， demonstrating strong robustness and resistance to interference.A high-precision phase calibration method for LC-SLMs is thus proposed， in which a double-grating structure is used to construct an absolute spatial coordinate system. This approach overcomes the limitations of traditional SLM calibration methods， including stringent stability requirements， susceptibility to environmental disturbances， and limited accuracy， and provides a reliable scheme for accurate phase calibration of LC-SLMs.关键词：liquid-crystal spatial light modulator;absolute interferometric calibration;grating diffraction;vibration resistance;U-Net neural network58|24|0更新时间：2025-12-19 - “在液晶显示模组力学研究领域，专家建立了三维有限元模型，深入揭示了模组在静压及挤压弯曲载荷下的全场位移分布特征与局部响应规律，为结构优化与可靠性设计提供理论依据与技术支持。”摘要：To address the simulation errors arising from the omission of actual structural details in mechanical studies of LCD modules， as well as the limitations of conventional universal testing machines that provide only global load information and cannot directly validate complex displacement field distributions， a three-dimensional finite element model incorporating complete structural features is established. The model is developed through precise geometric modeling that preserves key structural characteristics， an optimized mesh partitioning strategy that balances computational efficiency and accuracy， and material parameter definitions derived from experimental measurements and reliable data sources. Based on this model， the full-field displacement distribution and local response characteristics of LCD modules under static indentation and compression–bending loads are systematically investigated. Furthermore， a comprehensive full-field displacement verification framework is constructed using three-dimensional digital image correlation （DIC） technology. Experimental results indicate that， in single-point static indentation tests， the relative error between the simulated displacement field and DIC measurements in the compressed region is less than 10%； additional comparison of static indentation behavior at seven characteristic points further reduces the discrepancy to within 5%. Detailed analysis reveals that boundary constraint conditions—particularly the treatment of lateral degrees of freedom-and local structural stiffness effects， such as edge stiffening， are key factors governing the predictive accuracy of the model. These factors are closely associated with the mechanical origins of optical defects， including light leakage and surface indentations， commonly observed in LCD modules during service. The results demonstrate that the developed finite element model provides high accuracy in predicting the behavior of core compressed regions and in-plane responses. The modeling and validation methodology presented herein offers a reliable theoretical and technical basis for subsequent structural optimization and reliability design of LCD modules and establishes an important foundation for understanding and mitigating optical display defects from a mechanical perspective.关键词：LCD module;digital image correlation;full-field displacement;finite element simulation17|7|0更新时间：2025-12-19
- “在光频域反射测量技术领域，研究者设计了一种宽调谐范围、无跳模、高环境适应性的可调谐半导体激光源，实现了从1499.9998nm至1649.9996nm的无跳模可调谐输出，满足了高精度光传链路感知光频域反射测量仪的应用需求。”摘要：To meet the application requirements of tunable laser sources for optical frequency-domain reflectometry （OFDR）， a wide-tuning-range， mode-hopping-free semiconductor tunable laser source with high environmental adaptability is designed. Based on the theory of synchronous variation of feedback light and cavity modes， the pivot point is calculated and analyzed to achieve optimal mode matching between the oscillating light in the internal cavity and the external cavity during the tuning process. Subsequently， the output characteristics of the tunable laser source， including wavelength stability and power fluctuation under various conditions such as high and low temperatures and mechanical vibration， are experimentally investigated. The results demonstrate that the tunable laser source operates stably over long durations within an ambient temperature range of 0 °C to 40 °C and under conditions of drop and mechanical shock. At a driving current of 400 mA， continuous， mode-hopping-free tuning from 1 499.999 8 nm to 1 649.999 6 nm is achieved； the power fluctuation over the entire tuning range is better than 0.033 1%， the wavelength stability is better than 0.33 pm over 1 h， and the absolute wavelength accuracy is better than ±0.85 pm. These performance metrics essentially meet the application requirements of high-precision optical link sensing OFDR systems.关键词：fiber optics;tunable laser source;pattern matching;mode-hopping-free;tuning range;optical frequency domain reflection15|5|0更新时间：2025-12-19
- “在高精度测量领域，专家建立了基于相位测量偏折术的条纹反射法测量系统，实现了复杂曲面的高精度检测，为提升装备性能提供保障。”摘要：To enable high-precision measurement of complex surfaces， a fringe reflection measurement system based on phase measuring deflectometry was developed. The study addressed the system's structural design， calibration procedures， and metrology algorithms for a configuration comprising a camera， a display screen， and the tested object. High-performance display screens and cameras were selected according to the object's size and surface geometry. Instrument-specific calibrations were performed for the chosen display and camera， and the spatial relationships among the three components were established using a high-precision articulated-arm measuring device. System calibration was completed by combining computer-aided optimization with CODE V simulation analysis. Finally， an integral reconstruction algorithm was implemented to obtain high-precision surface measurements. Experimental results indicate that the proposed system achieves measurement performance comparable to LUPHOScan and ZYGO&CGH， with wavefront RMS accuracy exceeding λ/50. The system features a simple architecture， large dynamic range， and strong immunity to interference， thereby enhancing measurement efficiency and accuracy for complex surfaces and satisfying industrial processing and assembly requirements while improving equipment performance.关键词：optical metrology;complex surface;fringe reflection;phase measuring deflectometry;integral reconstruction15|4|0更新时间：2025-12-19
Modern Applied Optics
- “在超精密器件领域，专家综述了微纳米三坐标测量机的研究进展，提出了基于阿贝与布莱恩原则的空间误差建模与补偿创新方法，为实现纳米级精度测量提供解决方案。”摘要：In the field of advanced manufacturing， ultra-precision components are indispensable for national key equipment and high-end instruments， including aircraft engine fuel nozzles， precision gears， optical elements， and semiconductor devices. The characteristic dimensions of these components range from hundreds of micrometers to hundreds of millimeters， while the required dimensional accuracy can reach tens to hundreds of nanometers. To guarantee the quality and functional performance of such components， three-dimensional ultra-precision measurement is essential. This paper presents a comprehensive review of the development of micro- and nano-coordinate measuring machines （micro/nano CMMs） over the past two decades. The structural designs and measurement principles of micro/nano CMMs are systematically summarized， and key technologies—including error compensation methods， nano-probe design， measurement environment control， and system testing and application—are analyzed in depth. Particular emphasis is placed on the research progress and latest advances in four generations of micro/nano CMM prototypes developed at Hefei University of Technology. Recent work is highlighted， including an innovative volumetric error modeling and compensation method for micro/nano CMMs based on the Abbe and Bryan principles. This method enables nano-precision measurement of internal dimensions with characteristic sizes below 100 μm and large depth-to-width ratios. Finally， future research directions and potential application prospects of micro/nano CMMs in ultra-precision manufacturing are discussed.关键词：precision measurement;Micro-Electro-Mechanical System（MEMS）;Coordinate Measuring Machine（CMM）;high aspect ratio structures;nano probe;Abbe error;volumetric error compensation36|5|0更新时间：2025-12-19
- “在高精度控制领域，Stewart平台正运动学求解难题获突破。专家提出基于粒子滤波的求解方法，有效处理奇异区域估计问题，显著提升求解稳定性与精度，工程应用价值高。”摘要：Pose estimation of a Stewart platform is critical for achieving high-precision motion control， yet its forward kinematics problem remains challenging due to the strong coupling between pose variables and actuator lengths， the pronounced nonlinearity of the governing equations， and the presence of singular configurations. Conventional numerical algorithms often exhibit poor convergence and unstable estimation accuracy， particularly in proximity to singular regions. To enhance the stability and accuracy of forward kinematics computation， a novel solution based on a particle filter （PF） is proposed. A state-space model that incorporates both pose variables and actuator lengths is formulated， and probabilistic state inference is performed within the PF framework， thereby mitigating estimation difficulties associated with singular configurations. MATLAB simulations demonstrate that， relative to traditional methods， the proposed algorithm reduces mean absolute error （MAE） by 85.35%-99.43% and root mean square error （RMSE） by 86.98%-99.79%. These results confirm that the proposed approach achieves accurate forward kinematics estimation under complex operating conditions， offering strong adaptability， high robustness， and superior convergence behavior， and indicating substantial potential for practical engineering applications.关键词：parellel mechanism;Stewart platform;forward kinematics;particle filter;strongly coupled system;nonlinear system14|5|0更新时间：2025-12-19
- “在人形机器人领域，研究者提出了一种集成解耦算法，有效提高六维力传感器测量精度，为人形机器人运动控制提供高可靠性力感知保障。”摘要：To address the degradation of force measurement performance caused by interdimensional coupling in piezoelectric six-dimensional force/torque sensors， an integrated decoupling algorithm （BO-CNN-BiLSTM） combining Bayesian optimization （BO）， convolutional neural networks （CNN）， and bidirectional long short-term memory networks （BiLSTM） is proposed. In this algorithm， CNN is first employed to enhance the extraction of spatial coupling features from six-dimensional force signals. BiLSTM is then utilized to exploit bidirectional temporal modeling capabilities and dynamically capture cross-dimensional time-domain dependencies of the loads. Subsequently， BO is introduced to achieve adaptive global optimization of hyperparameters. In this way， the limitations of traditional decoupling methods in terms of real-time performance， generalization ability， and physical consistency are effectively overcome. The proposed BO-CNN-BiLSTM algorithm eliminates the empirical dependence on manually tuned parameters in conventional approaches and enables adaptive modeling of the nonlinear characteristics of sensors. Experimental results demonstrate that the maximum nonlinear error and cross-coupling error of the six-dimensional force/torque sensor outputs are 0.87% and 0.52%， respectively. The BO-CNN-BiLSTM decoupling algorithm effectively reduces both intra-dimensional and interdimensional coupling in six-dimensional force sensors， significantly improving measurement accuracy and providing important support for anthropomorphic motion control and environmental interaction in humanoid robots.关键词：six-dimensional force sensors;static decoupling;Bayesian optimization;CNN-BiLSTM13|3|0更新时间：2025-12-19
- “在微力测量领域，研究人员提出了基于模态局部化技术的弱耦合谐振式微力传感器，实现了超高灵敏度微力测量。”摘要：To overcome the limitations of traditional micro-force sensors in complex measurement scenarios requiring high sensitivity and high resolution， a weakly coupled resonant micro-force sensor based on mode localization technology is proposed and developed. A hybrid architecture is adopted， in which a beryllium bronze coupled resonator is integrated with a flexible lever mechanism， enabling micro-force measurement through monitoring of the amplitude ratio. By employing a flexure hinge structure， the applied micro-force is efficiently transformed into an axial stiffness variation of the perturbed resonator， thereby achieving ultra-high measurement sensitivity. The validity of the design model is verified through theoretical analysis and numerical simulation. An experimental system is constructed for mechanical amplitude measurement and micro-force calibration， allowing comprehensive evaluation of sensor performance. Experimental results indicate that the developed micro-force sensor provides a measurement range of 0-120 mN， a relative sensitivity exceeding 135×10-6/μN-more than 4，600 times higher than that of conventional strain-type micro-force sensors-and a resolution better than 30 μN. These results confirm the feasibility of the weakly coupled resonant structural scheme and demonstrate a novel approach for ultra-high-sensitivity micro-force measurement.关键词：micro-force sensor;weakly coupled resonators;mode localization;ultra-high sensitivity;amplitude ratio output8|3|0更新时间：2025-12-19
- “在微操作与微装配技术领域，专家设计了跨尺度微夹持器，实现了厘米级宏行程、纳米级分辨率且双臂平动，为解决大行程、高精度、快速反应和平动夹持需求提供解决方案。”摘要：To meet the requirements of large stroke， high precision， high speed， and translational clamping in micro-operation and micro-assembly， a symmetric stick–slip actuation mechanism is realized by employing a single macro fiber composite （MFC） to drive a compliant mechanism. This configuration enables the design of a cross-scale microgripper that combines centimeter-level travel with nanometer-scale resolution and dual-arm translational motion. An elongated compliant driving foot is used to store and release elastic potential energy， and its elastic recovery is exploited to counteract reverse friction， thereby suppressing stick-slip backlash. At the same time， the large single-step displacement of the MFC allows the microgripper to achieve high-speed motion at low driving frequencies， effectively reducing frictional wear. A static mechanical model of the compliant driving unit is established using the finite element method， and its output displacement and natural frequency are numerically analyzed. An experimental setup is then constructed to evaluate the performance of the proposed system. Experimental results indicate that the output single-step displacements of the stick–slip microgripper are 52.31 μm， 82.86 μm， and 124.68 μm under driving voltages of 400 V， 600 V， and 800 V， respectively， with a motion resolution of 9.6 nm. Under continuous operation with a driving voltage range of -400-800 V， the single-step displacement reaches up to 163.35 μm， with a step resolution of 7.8 nm and a clamping stroke of 14.4 mm. In addition， the designed microgripper achieves a motion speed of 1.25 mm/s at a low driving frequency of 5 Hz. The clamping force ranges from 0 to 167 mN under an 800 V trapezoidal wave voltage. These experimental results confirm the effectiveness and high-performance output characteristics of the proposed stick–slip microgripper.关键词：micro-operation system;piezoelectric actuation;stick-slip motion;compliant mechanism;microgripper8|3|0更新时间：2025-12-19
Micro/Nano Technology and Fine Mechanics
-
Contrastive learning combined with DenseNet for open-set classification of hyperspectral images AI导读
“在高光谱图像分类领域，专家提出了结合对比学习与DenseNet的新方法，有效提升了未知类别识别能力。”摘要：Hyperspectral image classification typically assumes that the training and test data share identical categories and that no unknown classes appear in the test set. However， this assumption is rarely satisfied in practical applications. In addition， the subtle inter-class differences inherent in hyperspectral data often lead to overlapping feature distributions and consequent decision boundary ambiguity. To address these issues， an open-set classification method for hyperspectral images is proposed that integrates contrastive learning with DenseNet. First， a spectral feature extraction module is employed to obtain the original spectral features， and multi-level feature interaction is realized through DenseNet. A transition module is further applied to compress spectral channels， thereby yielding clearer class boundary distributions. Second， the extracted spectral features are mapped to a spatial feature extraction module to obtain spatial-domain representations， where ResNet is adopted to capture local spatial structural information and enhance spatial perception. Subsequently， contrastive learning is introduced to reinforce intra-class compactness and inter-class separability， and is combined with a hard-sample mining mechanism to optimize ambiguous boundary features and improve the model’s discriminative capability for boundary-region samples. Experiments conducted on the Houston 2013， Pavia University， and WHU Hi-LongKou datasets demonstrate that the proposed method achieves superior ground-cover classification performance on unknown categories， with accuracies of 68.81%， 69.24%， and 59.26%， respectively. Meanwhile， overall accuracies of 89.49%， 95.06%， and 95.03% are obtained， indicating that the recognition of unknown categories is effectively enhanced while maintaining high classification accuracy for known categories.关键词：hyperspectral image;open-set;DenseNet;contrastive Learning;hard samples mining20|3|0更新时间：2025-12-19 - “在多磁源定位领域，专家提出了飞蛾扑火优化和密度聚类融合的定位方法，实现了高精度多磁源定位。”摘要：Multi-source localization is often affected by geomagnetic noise and ambiguous source boundaries， which significantly constrain its practical application. To achieve accurate and simultaneous localization of multiple magnetic targets， a multi-magnetic-source localization method based on Moth-Flame Optimization （MFO） nonlinear optimization and density clustering fusion is proposed. First， magnetic anomaly detection （MAD） is performed on a two-dimensional planar grid using a high-precision magnetic sensor to determine the point of maximum MAD value for each magnetic target. To address geomagnetic interference， a two-point magnetic anomaly localization scheme is designed in the region of maximum magnetic anomaly. The obtained localization data are then used as the initial input of the MFO algorithm， and the target positions are further refined through nonlinear optimization to improve localization accuracy. Subsequently， repeated optimization of each magnetic target is conducted， generating dense point clusters at the true target locations. Simultaneous localization of multiple magnetic targets is then achieved using the density-based DBSCAN clustering algorithm， which effectively suppresses the influence of noise and optimization outliers on localization accuracy. Simulation results for nine magnetic targets indicate that， after MFO-based initial localization， the average root mean square error （RMSE） is 0.134 6 m. Following clustering， the interference of noise is mitigated and the average RMSE is reduced to 0.076 4 m， representing a 57.1% improvement in accuracy. In experimental tests， the average RMSE for localizing three typical magnetic targets within a 1 m×1 m×1 m test area is less than 0.081 2 m， confirming that the proposed method enables high-precision multi-magnetic-source localization.关键词：geomagnetic detection;two-point localization;moth-flame optimization（MFO）;density clustering;multiple magnetic sources;nonlinear optimization;geomagnetic inteference;spatial clustering algorithm8|3|0更新时间：2025-12-19
- “在智能监测领域，研究者提出了基于改进GRU神经网络的刀具磨损状态预测方法，有效减小预测误差，取得良好效果。”摘要：To address the challenges of cumbersome signal feature extraction in traditional intelligent monitoring methods and the adverse effects of tool wear on workpiece quality and production efficiency during milling， a tool wear state prediction method based on an improved GRU neural network （BiGRU-1DCNN-CBAM） is proposed. Using statistical methods， time‑domain analysis， frequency‑domain analysis， and wavelet transform， 24 feature parameters of the tool signals-such as mean， kurtosis， and power spectral density-are extracted， and the resulting multimodal time‑series data are converted into time‑series images of tool features. A convolutional neural network （CNN） is then introduced to mine deep features of the signal data， and a convolutional block attention module （CBAM） is integrated to enhance the capability of the model to capture feature maps of vibration and cutting force signals. After flattening and concatenating the feature layers， the fused features are fed into a bidirectional GRU （BiGRU） to capture long-term dependencies， and the tool wear amount is predicted through a fully connected layer， thereby enabling remaining useful life prediction of the tool wear state during CNC machining. Experimental results on the PHM2010 dataset show that the RMSE and MAE of the proposed model are 2.17 μm and 1.29 μm， respectively. Compared with Bayesian-MCMC-Prognostics， SBiLSTM， RIME-CNN-SVM， MobileNetV3， TDConvLSTM， ISABO-IBiLSTM， IWOA-IECA-BiLSTM， and LSTM-CNN-CBAM models， the prediction accuracy in terms of RMSE and MAE is improved by more than 40.5% and 52.1%， respectively， while the time consumption is reduced by at least 2.8% relative to similar models. These results demonstrate that the proposed model can effectively characterize tool wear， reduce prediction errors， and achieve superior prediction performance.关键词：tool wear;deep learning;Convolutional Neural Network （CNN）;Gate Recurrent Unit （GRU）;attention module;state prediction18|5|0更新时间：2025-12-19
Information Sciences
- Address：No.3888 Dong Nanhu Road, Changchun, Jilin, China Postal code：130033
- Tel：0431-86176855 Email：gxjmgc@ciomp.ac.cn
- Technical support is provided by Beijing Founder electronics co., LTD 吉ICP备11002662号-17
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0