arXiv subjects

Chuang Sun

Publications and source records attributed to Chuang Sun.

18 recordsLinked to original sources

arXiv2026

PRISM: Precision and contact-rich Real-world Industrial Skill dataset with Multimodal sensing

Recent progress in robotic learning has been fueled by large-scale datasets collected in everyday environments. However, most existing datasets emphasize short-horizon, low-contact tasks such as pick-and-place, and therefore do not capture the precision control, force/torque or tactile regulation, and multimodal feedback required for industrial assembly. To address this gap, we introduce PRISM, a large-scale multimodal dataset for contact-rich industrial operations. The dataset spans more than 25 manipulation tasks (e.g., electronic components plug/unplug, conveyor-based sorting) and covers diverse mechanical constraints. PRISM includes more than 5,000 trajectories totaling 45 hours of teleoperated demonstrations, recorded using synchronized multi-view RGB-D, force/torque, tactile, and robot-state measurements. In contrast to datasets collected in household or laboratory settings, PRISM provides a realistic benchmark for multimodal perception and control under high-precision industrial constraints, and serves as a foundation for contact-rich, generalizable manipulation in real-world manufacturing environments. The dataset is open-sourced at: https://tengbo-yu.github.io/PRISM/

cs.RO↗

arXiv2025

Atomic Norm Soft Thresholding for Sparse Time-frequency Representation

Time-frequency (TF) representation of non-stationary signals typically requires the effective concentration of energy distribution along the instantaneous frequency (IF) ridge, which exhibits intrinsic sparsity. Inspired by the sparse optimization over continuum via atomic norm, a novel atomic norm soft thresholding for sparse TF representation (AST-STF) method is proposed, which ensures accurate TF localization under the strong duality. Numerical experiments demonstrate that the performance of the proposed method surpasses that of conventional methods.

eess.SP↗

arXiv2024

Near-infrared metalens empowered dual-mode high resolution and large FOV microscope

The spiral phase contrast microscope can clearly distinguish the morphological information of the low contrast objects (i.e., biological samples) because of the isotropic edge-enhancement effect, while the bright field microscope can image the overall morphology of amplitude objects. However, the imaging resolution, magnification, and field of view of conventional spiral phase contrast microscopes based on 4f filtering configuration are limited by the system's complexity. Here, we reported compact dual-mode microscopes working at near-infrared using the engineered metalens which can be tuned between the spiral phase contrast imaging and bright field imaging by polarization control. The metalens combines the high-resolution objective lens and polarization-controlled phase filter into a single-layer nanofins array. We demonstrated two infinity-corrected microscope systems to achieve subwavelength resolution (0.7 times of wavelength), large magnification (58X), and large field of view (600um times 800um). Unstained onion epidermal is imaged by the microscope to show the dual-mode imaging ability for the biological sample. Finally, a singlet dual-mode microscope system is demonstrated to show the edge-detection application for industrial standards. Our results could open new opportunities in applications of biological imaging, industrial machine vision, and semiconductor inspection.

physics.optics↗

arXiv2024

Longitudinal tri-foci Metalens empowered multiple-magnification and diffraction-limited microscope

Dielectric metalens has emerged as an attractive device for advanced imaging system because of its powerful manipulation ability of light beam, small volume, and light weight. However, the applications of silicon nitride (Si3N4) metalens are limited by the low refraction index of Si3N4, and multi-foci metalens has not been realized based on a Si3N4 metalens. Here, we deeply explore the working mechanism of a truncated waveguide meta-atom and obtain a Si3N4 metalens with longitudinal three diffraction-limited focal points. By utilizing the metalens sample as a condenser lens, a commercial microscope can obtain three magnifications based on a single objective lens. Finally, an infinity-corrected microscope with three high magnifications (9.5X, 10X, and 29X) and diffraction-limited resolution is integrated into centimetre-dimension for the first time by using the tri-foci metalens sample as an objective lens. This research would boost the scaling up of metalens microscope as well as the multifunctional application of Si3N4 metalens.

physics.optics↗

arXiv2024

Tunable on-chip optical traps for levitating particles based on single-layer metasurface

Optically levitated multiple nanoparticles has emerged as a platform for studying complex fundamental physics such as non-equilibrium phenomena, quantum entanglement, and light-matter interaction, which could be applied for sensing weak forces and torques with high sensitivity and accuracy. An optical trapping landscape of increased complexity is needed to engineer the interaction between levitated particles beyond the single harmonic trap. However, existing platforms based on spatial light modulators for studying interactions between levitated particles suffered from low efficiency, instability at focal points, the complexity of optical systems, and the scalability for sensing applications. Here, we experimentally demonstrated that a metasurface which forms two diffraction-limited focal points with a high numerical aperture (0.9) and high efficiency (31%) can generate tunable optical potential wells without any intensity fluctuations. A bistable potential and double potential wells were observed in the experiment by varying the focal points distance, and two nanoparticles were levitated in double potential wells for hours, which could be used for investigating the levitated particles nonlinear dynamics, thermal dynamics, and optical binding. This would pave the way for scaling the number of levitated optomechanical devices or realizing paralleled levitated sensors.

physics.optics↗

arXiv2023

A Novel Unsupervised Graph Wavelet Autoencoder for Mechanical System Fault Detection

Reliable fault detection is an essential requirement for safe and efficient operation of complex mechanical systems in various industrial applications. Despite the abundance of existing approaches and the maturity of the fault detection research field, the interdependencies between condition monitoring data have often been overlooked. Recently, graph neural networks have been proposed as a solution for learning the interdependencies among data, and the graph autoencoder (GAE) architecture, similar to standard autoencoders, has gained widespread use in fault detection. However, both the GAE and the graph variational autoencoder (GVAE) have fixed receptive fields, limiting their ability to extract multiscale features and model performance. To overcome these limitations, we propose two graph neural network models: the graph wavelet autoencoder (GWAE), and the graph wavelet variational autoencoder (GWVAE). GWAE consists mainly of the spectral graph wavelet convolutional (SGWConv) encoder and a feature decoder, while GWVAE is the variational form of GWAE. The developed SGWConv is built upon the spectral graph wavelet transform which can realize multiscale feature extraction by decomposing the graph signal into one scaling function coefficient and several spectral graph wavelet coefficients. To achieve unsupervised mechanical system fault detection, we transform the collected system signals into PathGraph by considering the neighboring relationships of each data sample. Fault detection is then achieved by evaluating the reconstruction errors of normal and abnormal samples. We carried out experiments on two condition monitoring datasets collected from fuel control systems and one acoustic monitoring dataset from a valve. The results show that the proposed methods improve the performance by around 3%~4% compared to the comparison methods.

eess.SP↗

arXiv2023

Industrial Anomaly Detection with Domain Shift: A Real-world Dataset and Masked Multi-scale Reconstruction

Industrial anomaly detection (IAD) is crucial for automating industrial quality inspection. The diversity of the datasets is the foundation for developing comprehensive IAD algorithms. Existing IAD datasets focus on the diversity of data categories, overlooking the diversity of domains within the same data category. In this paper, to bridge this gap, we propose the Aero-engine Blade Anomaly Detection (AeBAD) dataset, consisting of two sub-datasets: the single-blade dataset and the video anomaly detection dataset of blades. Compared to existing datasets, AeBAD has the following two characteristics: 1.) The target samples are not aligned and at different scales. 2.) There is a domain shift between the distribution of normal samples in the test set and the training set, where the domain shifts are mainly caused by the changes in illumination and view. Based on this dataset, we observe that current state-of-the-art (SOTA) IAD methods exhibit limitations when the domain of normal samples in the test set undergoes a shift. To address this issue, we propose a novel method called masked multi-scale reconstruction (MMR), which enhances the model's capacity to deduce causality among patches in normal samples by a masked reconstruction task. MMR achieves superior performance compared to SOTA methods on the AeBAD dataset. Furthermore, MMR achieves competitive performance with SOTA methods to detect the anomalies of different types on the MVTec AD dataset. Code and dataset are available at https://github.com/zhangzilongc/MMR.

cs.CV↗

arXiv2023

Filter-informed Spectral Graph Wavelet Networks for Multiscale Feature Extraction and Intelligent Fault Diagnosis

Intelligent fault diagnosis has been increasingly improved with the evolution of deep learning (DL) approaches. Recently, the emerging graph neural networks (GNNs) have also been introduced in the field of fault diagnosis with the goal to make better use of the inductive bias of the interdependencies between the different sensor measurements. However, there are some limitations with these GNN-based fault diagnosis methods. First, they lack the ability to realize multiscale feature extraction due to the fixed receptive field of GNNs. Secondly, they eventually encounter the over-smoothing problem with increase of model depth. Lastly, the extracted features of these GNNs are hard to understand owing to the black-box nature of GNNs. To address these issues, a filter-informed spectral graph wavelet network (SGWN) is proposed in this paper. In SGWN, the spectral graph wavelet convolutional (SGWConv) layer is established upon the spectral graph wavelet transform, which can decompose a graph signal into scaling function coefficients and spectral graph wavelet coefficients. With the help of SGWConv, SGWN is able to prevent the over-smoothing problem caused by long-range low-pass filtering, by simultaneously extracting low-pass and band-pass features. Furthermore, to speed up the computation of SGWN, the scaling kernel function and graph wavelet kernel function in SGWConv are approximated by the Chebyshev polynomials. The effectiveness of the proposed SGWN is evaluated on the collected solenoid valve dataset and aero-engine intershaft bearing dataset. The experimental results show that SGWN can outperform the comparative methods in both diagnostic accuracy and the ability to prevent over-smoothing. Moreover, its extracted features are also interpretable with domain knowledge.

eess.SP↗

arXiv2022

Superpixel perception graph neural network for intelligent defect detection of aero-engine blade

Aero-engine is the core component of aircraft and other spacecraft. The high-speed rotating blades provide power by sucking in air and fully combusting, and various defects will inevitably occur, threatening the operation safety of aero-engine. Therefore, regular inspections are essential for such a complex system. However, existing traditional technology which is borescope inspection is labor-intensive, time-consuming, and experience-dependent. To endow this technology with intelligence, a novel superpixel perception graph neural network (SPGNN) is proposed by utilizing a multi-stage graph convolutional network (MSGCN) for feature extraction and superpixel perception region proposal network (SPRPN) for region proposal. First, to capture complex and irregular textures, the images are transformed into a series of patches, to obtain their graph representations. Then, MSGCN composed of several GCN blocks extracts graph structure features and performs graph information processing at graph level. Last but not least, the SPRPN is proposed to generate perceptual bounding boxes by fusing graph representation features and superpixel perception features. Therefore, the proposed SPGNN always implements feature extraction and information transmission at the graph level in the whole SPGNN pipeline, to alleviate the reduction of receptive field and information loss. To verify the effectiveness of SPGNN, we construct a simulated blade dataset with 3000 images. A public aluminum dataset is also used to validate the performances of different methods. The experimental results demonstrate that the proposed SPGNN has superior performance compared with the state-of-the-art methods.

cs.CV↗

arXiv2020

Deep Learning Algorithms for Rotating Machinery Intelligent Diagnosis: An Open Source Benchmark Study

With the development of deep learning (DL) techniques, rotating machinery intelligent diagnosis has gone through tremendous progress with verified success and the classification accuracies of many DL-based intelligent diagnosis algorithms are tending to 100\%. However, different datasets, configurations, and hyper-parameters are often recommended to be used in performance verification for different types of models, and few open source codes are made public for evaluation and comparisons. Therefore, unfair comparisons and ineffective improvement may exist in rotating machinery intelligent diagnosis, which limits the advancement of this field. To address these issues, we perform an extensive evaluation of four kinds of models, including multi-layer perception (MLP), auto-encoder (AE), convolutional neural network (CNN), and recurrent neural network (RNN), with various datasets to provide a benchmark study within the same framework. We first gather most of the publicly available datasets and give the complete benchmark study of DL-based intelligent algorithms under two data split strategies, five input formats, three normalization methods, and four augmentation methods. Second, we integrate the whole evaluation codes into a code library and release this code library to the public for better development of this field. Third, we use specific-designed cases to point out the existing issues, including class imbalance, generalization ability, interpretability, few-shot learning, and model selection. By these works, we release a unified code framework for comparing and testing models fairly and quickly, emphasize the importance of open source codes, provide the baseline accuracy (a lower bound) to avoid useless improvement, and discuss potential future directions in this field. The code library is available at https://github.com/ZhaoZhibin/DL-based-Intelligent-Diagnosis-Benchmark.

eess.SP↗

arXiv2019

Applications of Unsupervised Deep Transfer Learning to Intelligent Fault Diagnosis: A Survey and Comparative Study

Recent progress on intelligent fault diagnosis (IFD) has greatly depended on deep representation learning and plenty of labeled data. However, machines often operate with various working conditions or the target task has different distributions with the collected data used for training (the domain shift problem). Besides, the newly collected test data in the target domain are usually unlabeled, leading to unsupervised deep transfer learning based (UDTL-based) IFD problem. Although it has achieved huge development, a standard and open source code framework as well as a comparative study for UDTL-based IFD are not yet established. In this paper, we construct a new taxonomy and perform a comprehensive review of UDTL-based IFD according to different tasks. Comparative analysis of some typical methods and datasets reveals some open and essential issues in UDTL-based IFD which are rarely studied, including transferability of features, influence of backbones, negative transfer, physical priors, etc. To emphasize the importance and reproducibility of UDTL-based IFD, the whole test framework will be released to the research community to facilitate future research. In summary, the released framework and comparative study can serve as an extended interface and basic results to carry out new studies on UDTL-based IFD. The code framework is available at \url{https://github.com/ZhaoZhibin/UDTL}.

eess.SP↗

arXiv2019

WaveletKernelNet: An Interpretable Deep Neural Network for Industrial Intelligent Diagnosis

Convolutional neural network (CNN), with ability of feature learning and nonlinear mapping, has demonstrated its effectiveness in prognostics and health management (PHM). However, explanation on the physical meaning of a CNN architecture has rarely been studied. In this paper, a novel wavelet driven deep neural network termed as WaveletKernelNet (WKN) is presented, where a continuous wavelet convolutional (CWConv) layer is designed to replace the first convolutional layer of the standard CNN. This enables the first CWConv layer to discover more meaningful filters. Furthermore, only the scale parameter and translation parameter are directly learned from raw data at this CWConv layer. This provides a very effective way to obtain a customized filter bank, specifically tuned for extracting defect-related impact component embedded in the vibration signal. In addition, three experimental verification using data from laboratory environment are carried out to verify effectiveness of the proposed method for mechanical fault diagnosis. The results show the importance of the designed CWConv layer and the output of CWConv layer is interpretable. Besides, it is found that WKN has fewer parameters, higher fault classification accuracy and faster convergence speed than standard CNN.

cs.CV↗

arXiv2017

Discrete Symmetries of Calabi-Yau Hypersurfaces in Toric Four-Folds

We analyze freely-acting discrete symmetries of Calabi-Yau three-folds defined as hypersurfaces in ambient toric four-folds. An algorithm which allows the systematic classification of such symmetries which are linearly realised on the toric ambient space is devised. This algorithm is applied to all Calabi-Yau manifolds with $h^{1,1}(X)\leq 3$ obtained by triangulation from the Kreuzer-Skarke list, a list of some $350$ manifolds. All previously known freely-acting symmetries on these manifolds are correctly reproduced and we find five manifolds with freely-acting symmetries. These include a single new example, a manifold with a $\mathbb{Z}_2\times\mathbb{Z}_2$ symmetry where only one of the $\mathbb{Z}_2$ factors was previously known. In addition, a new freely-acting $\mathbb{Z}_2$ symmetry is constructed for a manifold with $h^{1,1}(X)=6$. While our results show that there are more freely-acting symmetries within the Kreuzer-Skarke set than previously known, it appears that such symmetries are relatively rare.

hep-th↗

arXiv2017

A Comprehensive Survey of Brane Tilings

An infinite class of $4d$ $\mathcal{N}=1$ gauge theories can be engineered on the worldvolume of D3-branes probing toric Calabi-Yau 3-folds. This kind of setup has multiple applications, ranging from the gauge/gravity correspondence to local model building in string phenomenology. Brane tilings fully encode the gauge theories on the D3-branes and have substantially simplified their connection to the probed geometries. The purpose of this paper is to push the boundaries of computation and to produce as comprehensive a database of brane tilings as possible. We develop efficient implementations of brane tiling tools particularly suited for this search. We present the first complete classification of toric Calabi-Yau 3-folds with toric diagrams up to area 8 and the corresponding brane tilings. This classification is of interest to both physicists and mathematicians alike.

hep-th↗

arXiv2015

Statistics of Stationary Points of Random Finite Polynomial Potentials

The stationary points (SPs) of the potential energy landscapes (PELs) of multivariate random potentials (RPs) have found many applications in many areas of Physics, Chemistry and Mathematical Biology. However, there are few reliable methods available which can find all the SPs accurately. Hence, one has to rely on indirect methods such as Random Matrix theory. With a combination of the numerical polynomial homotopy continuation method and a certification method, we obtain all the certified SPs of the most general polynomial RP for each sample chosen from the Gaussian distribution with mean 0 and variance 1. While obtaining many novel results for the finite size case of the RP, we also discuss the implications of our results on mathematics of random systems and string theory landscapes.

cond-mat.stat-mech↗

arXiv2014

The Scattering Variety

The so-called Scattering Equations which govern the kinematics of the scattering of massless particles in arbitrary dimensions have recently been cast into a system of homogeneous polynomials. We study these as affine and projective geometries which we call Scattering Varieties by analyzing such properties as Hilbert series, Euler characteristic and singularities. Interestingly, we find structures such as affine Calabi-Yau threefolds as well as singular K3 and Fano varieties.

hep-th↗

arXiv2014

Heterotic Model Building: 16 Special Manifolds

We study heterotic model building on 16 specific Calabi-Yau manifolds constructed as hypersurfaces in toric four-folds. These 16 manifolds are the only ones among the more than half a billion manifolds in the Kreuzer-Skarke list with a non-trivial first fundamental group. We classify the line bundle models on these manifolds, both for SU(5) and SO(10) GUTs, which lead to consistent supersymmetric string vacua and have three chiral families. A total of about 29000 models is found, most of them corresponding to SO(10) GUTs. These models constitute a starting point for detailed heterotic model building on Calabi-Yau manifolds in the Kreuzer-Skarke list. The data for these models can be downloaded here: http://www-thphys.physics.ox.ac.uk/projects/CalabiYau/toricdata/index.html

hep-th↗

arXiv2013

Superconformal Block Quivers, Duality Trees and Diophantine Equations

We generalize previous results on N=1, (3+1)-dimensional superconformal block quiver gauge theories. It is known that the necessary conditions for a theory to be superconformal, i.e. that the beta and gamma functions vanish in addition to anomaly cancellation, translate to a Diophantine equation in terms of the quiver data. We re-derive results for low block numbers revealing an new intriguing algebraic structure underlying a class of possible superconformal fixed points of such theories. After explicitly computing the five block case Diophantine equation, we use this structure to reorganize the result in a form that can be applied to arbitrary block numbers. We argue that these theories can be thought of as vectors in the root system of the corresponding quiver and superconformality conditions are shown to associate them to certain subsets of imaginary roots. These methods also allow for an interpretation of Seiberg duality as the action of the affine Weyl group on the root lattice.

hep-th↗