最新刊期
卷 34 ， 期 13 ， 2026
- 摘要：To meet the requirements of airborne applications and resolve the design trade-offs among wide field of view， large exit pupil diameter， and lightweight structure in high-performance helmet-mounted displays， an opposite-side oblique projection optical system based on freeform surfaces is proposed， featuring a streamlined configuration. A reverse ray-tracing method combined with an incremental optimization strategy was employed to refine the surface profiles of the relay lens group and the freeform combiner， effectively correcting astigmatism， distortion， and coma. To reduce structural and assembly complexity， the number of relay lens elements was limited to five and arranged coaxially. The resulting system achieves a diagonal field of view of 50°， an exit pupil diameter of 11 mm， and an eye relief of 40 mm， with the modulation transfer function exceeding 0.46 at 30 lp/mm across the full field， while maintaining a total weight of 84.5 g. Tolerance analysis indicates low sensitivity of both MTF and distortion to standard manufacturing errors， confirming the engineering feasibility of the design. Prototype validation demonstrates clear imaging in both transmission and projection paths， providing a robust solution for compact and easily aligned airborne helmet-mounted display systems.关键词：optical design;freeform surface;helmet-mounted display optical system;catadioptric system;large diameter of exit pupil53|85|0更新时间：2026-07-22
- 摘要：Parallel plates， as typical transmission optical components， are widely employed in optical detection， semiconductor manufacturing， and defense applications. Their optical homogeneity has a significant impact on overall system performance. However， existing high-precision measurement methods face challenges， including difficulty in decoupling interference aliasing， time-consuming multi-frame acquisition， and limited capability in suppressing environmental noise. To address these limitations， a high-precision， deep learning-based single-frame interferometric decoupling method for measuring the optical homogeneity of parallel plates is proposed. First， a mapping model between aliased and single-sided interferograms is constructed to decouple the single-frame aliased interferogram， enabling the retrieval of single-sided interferograms for the front and back surfaces and effective separation of interference fringes. Subsequently， a virtual phase-shifting reconstruction is performed to generate a sequence with equal phase-shift intervals from the single-frame interferogram. In combination with the conventional five-step phase-shifting method， phase extraction and surface profile reconstruction are achieved， enabling high-precision evaluation of optical homogeneity. A two-stage convolutional neural network is developed， in which the first stage performs the mapping from aliased fringes to single-sided fringes， and the second stage generates the five-step phase-shifting sequence and reconstructs the surface profiles of the front and back surfaces. In addition， a deep learning-based single-frame interferometric decoupling experimental system is established for optical homogeneity measurement. Experiments conducted on Φ75 mm and Φ50 mm parallel plate samples demonstrate that the proposed method yields results in good agreement with those obtained using a ZYGO interferometer， with absolute deviations on the order of 10⁻⁷. The proposed approach enables high-precision and rapid measurement of optical homogeneity using only a single-frame aliased interferogram， providing an effective solution for high-throughput and in situ inspection of optical components.关键词：parallel plates;optical homogeneity;deep learning;interferometric decoupling;high-precision43|58|0更新时间：2026-07-22
- 摘要：In order to improve the resolution of the laser self-mixing interference displacement measurement based on the fringe counting method， a micro-displacement measurement system using the dual-wavelength single-longitudinal-mode （DWSLM） ytterbium-doped fiber laser self-mixing interference has been designed， which can output two self-mixing interference signals with opposite phases. An optimal threshold method is used for subdividing the signals to extract the countable feature points with a resolution of λ/8， and the measured displacement is then reconstructed by employing the cubic spline interpolation method. The self-mixing interference characteristics of the DWSLM ytterbium-doped fiber laser and the micro-displacement measurement errors have been investigated experimentally. The experimental results have indicated that when the target mirror vibrates periodically with a frequency range from 3 Hz to 9 Hz and an amplitude range from 2 μm to 11 μm， the micro-displacement measurement error varies within a range from 96 nm to 105 nm， corresponding to an average measurement error of 100.2 nm， and the average relative error is 1.98%. Such a micro-displacement measurement system using the DWSLM ytterbium-doped fiber laser self-mixing interference is simple and feasible， which has a broad application prospect.关键词：fiber laser;laser self-mixing interference;dual-wavelength single-longitudinal-mode;fringe subdivision;Micro-displacement measurement9|6|0更新时间：2026-07-22
- 摘要：Compared with conventional Gaussian speckle fields， speckles generated by a perfect optical vortex （POV） exhibit distinctive characteristics in light-field propagation， speckle size， and the optical memory effect. However， the annular intensity distribution produces a donut-shaped structure， leading to degraded imaging quality， reduced measurement accuracy， and increased difficulty in data analysis. To address this limitation， an optical system was designed to enhance the illumination uniformity of POV speckle fields. Using the POV speckle field as the light source， a highly uniform Köhler illumination system was designed in ZEMAX. Key parameters， including the radius of curvature and lens spacing， were systematically optimized by minimizing the merit function. The system consisted of a three-element collecting mirror and a cemented doublet condenser. The collecting mirror was comparatively optimized using crown glass H-ZK9A and flint glass ZF2， whereas the condenser was constructed with commercial Thorlabs lenses. Under single-wavelength conditions in sequential mode， the illumination uniformity of both systems exceeded 96%， with the ZF2-based design reaching 98.9%. In non-sequential mode， in which physical scattering and phase modulation were introduced， the system uniformity remained above 90%. Under multi-wavelength conditions， both systems maintained stable illumination uniformity across the visible spectrum in sequential and non-sequential modes， demonstrating their suitability for broadband illumination. In terms of microscopic statistical properties， the speckle contrast at the image plane approached 1， and the simulated characteristic size of the spatial autocorrelation showed strong agreement with the theoretical value. Further experiments demonstrated that a dynamic POV speckle field could be stably generated. The measured illumination uniformity obtained from the horizontal and vertical profiles， as well as from the central rectangular region， exceeded 86%， confirming the effectiveness and feasibility of the design. The experimental values were slightly lower than the simulation results， mainly owing to physical factors such as alignment errors， SLM zero-order light， and environmental noise. Uniform POV speckle illumination was successfully achieved in this study. The macroscopic annular intensity distribution was eliminated while the microscopic statistical properties were well preserved. This approach provides an effective strategy for optimizing specially structured light systems and offers potential applications in precision imaging and information processing.关键词：optical design;Köhler illumination;perfect optical vortex speckle;ZEMAX50|75|0更新时间：2026-07-22
Modern Applied Optics
- 摘要：In the monitoring of cutting forces for large-sized work pieces， the enlargement of the working area in large-area dynamometers leads to the technical challenges of increased mass and decreased natural frequency. To address this， this study introduces a lightweight design philosophy and proposes a large-area piezoelectric multi-component dynamometer. A systematic theoretical analysis and experimental verification are conducted. Through structural-functional analysis， the design constraints and optimization direction for lightweighting are clarified. Topology optimization is employed to obtain the initial structure and identify the key dimensional parameters. Design of experiments combined with the response surface methodology is adopted to construct surrogate models relating the parameters to the optimization objectives. A multi-objective optimization algorithm is then applied for parameter optimization， determining the final design solution and completing the fabrication of the dynamometer prototype. The performance of the lightweight dynamometer is validated through static and dynamic calibration tests as well as practical milling force experiments. The experimental results show that the overall mass of the dynamometer is reduced by 8.3%， while its lowest first-order natural frequency is increased by 33.2%. The linearity and repeatability errors are both below 0.5%， the cross-axis interference errors are less than 3% in all directions， and the natural frequency of each axis exceeds 1.9 kHz. The milling force experiments confirm its measurement stability across the working area and clearly reveal its sensitive response to different working conditions. This verifies that the lightweight dynamometer retains excellent performance. The study provides theoretical guidance for the subsequent structural design and natural frequency enhancement of dynamometers and holds clear value for engineering application.关键词：test area;cutting force measurement;response surface analysis;lightweight12|6|0更新时间：2026-07-22
- 摘要：To achieve stable separation and reduce cutting-induced damage， a double-pulse picosecond lasers and mechanical cleaving combined cutting process for SiC has been proposed.High-quality and stable cutting of the material was achieved through a process route that integrates the machining of fracture-guiding grooves with stress-induced fracture. Meanwhile， the effects of double-pulse delay， scanning interval， laser power， and scanning speed on the ablation behavior， cleaving stability， and cross-sectional quality were systematically investigated.The double-pulse sequence markedly enhanced pulse synergy. The highest material removal was achieved at a 1 ps delay， where the single-pass groove depth increased by 81.34% relative to that of a single pulse. An interlayer scanning interval of 30 μm provided the best overall balance among cutting quality， efficiency， and cleaving controllability. The optimal laser power and scanning speed were 18.6 W and 15 mm/s， respectively. Elemental analysis showed increased C and O contents but decreased Si content in the sidewall with increasing ablation， indicating Si volatilization， C enrichment， and oxidation induced by laser irradiation.The proposed technique enables the cutting and separation of silicon carbide. Under optimized conditions， it generates more stable cleaving-guiding grooves and internal cracks， which is beneficial for achieving low-damage and highly controllable SiC cutting. The research findings provide a technical foundation for high-quality laser cutting of hard and brittle single-crystal materials such as SiC.关键词：laser fabrication;4H-SiC;picosecond laser;double-pulse;laser cutting20|6|0更新时间：2026-07-22
- 摘要：To address the thermal cracking and fracture that readily occur in the deep-cavity/shallow-groove transition region during the precision molding of stepped optical glass microfluidic chips， the mechanism of microcrack formation was investigated and a crack-suppression method was proposed. The cooling and annealing process three-dimensional thermo-mechanical coupled finite element model was established based on the high-temperature viscoelastic constitutive behavior of D-ZK2N glass. The Deborah number was introduced to characterize the competition between structural relaxation and stress freezing， and a gradient cooling strategy was developed and verified through molding experiments. The results show that non-uniform thermal inertia and volumetric shrinkage mismatch are the main causes of the “compressive outside–tensile inside” stress concentration in the transition region. Under linear cooling， stress in this region continuously accumulated and eventually formed a high risk of cracking. After adopting the gradient cooling strategy， structural relaxation in the transition region was enhanced， and simulation results showed the residual stress in the transition region decreased by 22.4%. No cracks were observed in the molded part， and the molded depth of the deep-cavity structure reached 4.522 mm. These results indicate that the Deborah-number-based gradient cooling strategy can effectively suppress thermal cracking in stepped transition regions and is suitable for the precisionmolding of glass microfluidic chips with large aspect-ratio structures.关键词：microfluidic chip;precision molding;gradient cooling;Deborah number;thermal cracking7|5|0更新时间：2026-07-22
- 摘要：To improve the applicability of robots in high-precision automation and enhance robot positioning accuracy， this study constructs a robotic drilling system based on a passive scale tracker and conducts systematic research on system error source analysis， coordinate transformation calibration， and error compensation strategies. Firstly， the measurement mechanism and internal error sources of the passive scale tracker are analyzed， and error compensation is achieved by combining direct measurement and nonlinear fitting of spatial circular planes. Secondly， the overall architecture of the drilling system is elaborated， and the coordinate transformation and calibration methods of the system are defined. On this basis， a cyclic compensation strategy for robot pose errors is proposed， establishing the compensation idea of "attitude first， position later". To verify the effectiveness of the proposed method， comparative drilling experiments are carried out on polymethyl methacrylate （PMMA） workpieces. The experimental results show that the positioning accuracy of the machined holes is significantly improved， with the positioning error reduced from 1.23 mm to 0.15 mm， which can meet the stringent requirements of precision machining and automated assembly in terms of accuracy， stability， and reliability.关键词：robots;passive scale tracker;error compensation;drilling accuracy7|5|0更新时间：2026-07-22
Micro\/Nano Technology and Fine Mechanics
- 摘要：Space–ground integrated information networks constitute a key infrastructure for sixth-generation （6G） communications， offering wide coverage， high capacity， and flexible global access. However， practical link management remains constrained by two coupled challenges： limited onboard payload resources at satellite feeder terminals and pronounced spatial-temporal heterogeneity in terrestrial traffic. These factors lead to access congestion， redundant inter-satellite links， and inefficient energy utilization.To address these issues， a hierarchical collaborative dynamic link management strategy is proposed， in which the global joint optimization problem is decomposed into two interrelated subtasks： satellite–ground access scheduling and inter-satellite topology reconstruction. At the access layer， a Multi-Agent Proximal Policy Optimization algorithm is developed， incorporating satellite ephemeris， link visibility， link distance， connection history， and population weights to allocate limited ground-satellite link resources preferentially to high-demand regions. At the coupling layer， a traffic heat estimation model is constructed to map terrestrial demand onto satellite nodes via established access links， followed by diffusion across the inter-satellite topology with hop-based attenuation. At the topology layer， an Advantage Actor-Critic agent is introduced， leveraging traffic heat， historical link states， backbone masks， and node-degree constraints to selectively deactivate low-load redundant laser inter-satellite links while preserving critical paths and maintaining fundamental network connectivity.A Walker Delta constellation comprising 484 low-Earth-orbit satellites is simulated， with 100 major cities serving as traffic sources. The results demonstrate a blocking rate of 1.33%， a quality-of-service-constrained network capacity of 310.44 Gb/s， and a 32% reduction in energy consumption compared with a full-mesh baseline. The normalized resource utilization efficiency increases to 1.81 times that of the baseline， while the average link handover ratio remains as low as 1.22% per time slot. These findings indicate that the proposed strategy significantly enhances network capacity and energy efficiency， while exhibiting strong robustness in large-scale dynamic satellite network scenarios.关键词：space-ground integrated information networks;hierarchical collaborative architecture;dynamic link management;multi-agent reinforcement learning79|63|0更新时间：2026-07-22
- 摘要：To enhance the measurement stability and interference immunity of the resistance measurement system in infrared radiation reference satellites， we propose a noise analysis and digital averaging parameters optimization method. The method addresses frequency aliasing at low data acquisition rates by recovering noise frequencies through multiple under-sampled data sets， while providing recovery correction for sampling rate deviations. We perform equivalent frequency response modeling of the measurement process and optimize the digital averaging parameters by constructing a cost function that balances noise suppression effectiveness and data acquisition time. The experiments show that the method achieves frequency identification in the 50 Hz to 8765 Hz range at data acquisition rates below 31 Hz； the optimized filter parameters provide suppression of the environmental interference exceeding 47 dB. Compared with conventional parameters， the peak noise of the voltage measurement is reduced from 7.1517 mV to 0.0765 mV， and the stability of the resistance measurement is improved from 0.09727 Ω to 0.00138 Ω. The analysis-optimization workflow allows accurate noise identification at low data acquisition rates， realizes multi-frequency discrimination and aliasing component analysis， and guides the optimization of filter parameters， thus providing a reliable solution for the precision resistance measurement system operating in complex electromagnetic environments of infrared radiation reference satellites.关键词：precision resistance measurement;noise analysis;multi-rate sampling;digital averaging filter34|30|0更新时间：2026-07-22
- 摘要：In multi-robot visual Simultaneous Localization and Mapping （SLAM）， loop closure detection consists of intra-robot and inter-robot loop closures， serving local trajectory optimization and inter-robot relative pose estimation， respectively. Nevertheless， it remains challenged by illumination changes and viewpoint differences， resulting in false matches and missed detections. Loop closure detection in multi-robot visual SLAM is challenged by illumination changes and viewpoint differences. These conditions lead to false matches and missed detections. This paper proposes a robust loop closure detection method with three contributions. First， a sliding-window candidate frame screening mechanism is designed. It adapts the detection threshold using the mean and variance of image similarity scores. This strategy effectively prevents missed loop closures. Second， a sliding-window temporal consistency test is introduced. It exploits the temporal continuity of intra-robot loop closures. This test eliminates spurious false matches caused by similar scenes. Third， a covisibility-enhanced geometric verification method is developed. It constructs a local submap by leveraging covisible neighbor frames. This expands the observation field and improves verification quality. Experiments are conducted on multiple datasets from outdoor， tunnel， and underground scenes. Compared with DBOW2， the proposed method increases the proportion of high-quality loop closures by more than 6%. The absolute trajectory error is reduced by 10.0% to 63.7%.关键词：multi-robot visual simultaneous localization and mapping;intra-robot loop closure;inter-robot loop closure;sliding window;co-visibility relationship22|27|0更新时间：2026-07-22
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