arXiv subjects

Weipeng Li

Publications and source records attributed to Weipeng Li.

12 recordsLinked to original sources

arXiv2026

HarvestPoint-ACT: Explicit Target Selection and Harvest-Point Conditioning for Robotic Fruit Harvesting under Occlusion

End-to-end imitation learning avoids hand-made robot motion for approaching and grasping, but the policy must still decide which fruit to pick and where to close the gripper. Occlusion can make the policy lose the selected fruit during harvesting, and the correct closing point is difficult to infer from pixels alone. This paper presents HarvestPoint-ACT, which makes both decisions explicit in perception and provides them to the policy. An instance segmentation front end with a keypoint branch predicts a mask and a harvest point for each visible fruit, where the harvest point specifies the location to close the gripper. A scheduler ranks detected candidates by occlusion and travel distance and selects one target. After each attempt, it redetects and reranks the candidates because the canopy may have changed. The selected fruit is encoded for an action chunking transformer as an eight-dimensional state, containing the absolute harvest point, the vector from the gripper to that point, a validity flag, and a confidence score. When the selected fruit is temporarily undetected, the system retains the last harvest point estimate in the robot base frame and marks it as stale, and aborts the attempt if the loss persists. On a canopy mock-up, HarvestPoint-ACT achieves a success rate of 88%, and of 75% under heavy occlusion.

cs.RO↗

arXiv2026

Long-horizon prediction of three-dimensional wall-bounded turbulence with CTA-Swin-UNet and resolvent analysis

Long-horizon prediction of three-dimensional (3D) wall-bounded turbulence with machine-learning methods remains a challenging task, due to the rapid accumulation of autoregressive errors and the substantially computational cost. To address these challenges, we present a hybrid machine-learning framework, in which a channel-time-attention Swin-UNet (CTA-Swin-UNet) and a multi-time-scale fusion correction (MTFC) strategy are developed to predict the turbulent flow fields in a wall-parallel plane, with affordable computational cost. Then, 3D flow fields are reconstructed via a resolvent-based spectral linear stochastic estimation (SLSE), rooting from the predicted planar flow. Results show that the CTA-Swin-UNet outperforms the baseline models (LSTM, FNO and traditional Swin-UNet) in both single-step prediction and autoregressive rollouts, indicating the effectiveness of introducing the CTA module into the Swin-UNet architecture. At the same temporal interval, the CTA-Swin-UNet remains stable for approximately 150 rollout steps, while the baseline models fail within 20 to 50 rollout steps. After introducing the MTFC strategy, a longer horizon upto 300 steps is achieved. Using the resolvent-based SLSE reconstruction further recovers the 3D flow structures and energy spectral distributions from the predicted planar inputs, which demonstrates that the proposed framework provides an effective and computationally efficient approach for long-horizon autoregressive prediction of 3D wall-bounded turbulence.

physics.flu-dyn↗

arXiv2026

Causal structures of turbulent skin-friction drag in wall-bounded turbulent flows

Understanding the mechanism of turbulent skin-friction drag (TSD) generation is of fundamental and practical importance for designing effective drag reduction strategies. However, many previous studies adopted correlation analysis to reveal the causal map between turbulent motions and TSD generation, an approach that is potentially risky as correlation does not necessarily imply causation. In this study, a novel causal inference method called Liang-Kleeman information flow (LKIF) is utilized for the first time to identify the velocity-induced causal structures related to TSD generation in a turbulent channel flow. The statistical properties of the causal structures are comprehensively investigated. The positive and negative causal structures, defined by their signs and respectively associated with an increase and decrease in TSD information entropy, promote and suppress the generation of extreme TSD. Particularly, we find that the underlying physics of causal structures is essentially associated with the processes of streamwise streaks and rolls approaching or receding from the extreme events. Results indicate that the physics-informed LKIF framework can reveal a more explicit and interpretable causal relationship than correlation analysis.

physics.flu-dyn↗

arXiv2024

From Tissue Plane to Organ World: A Benchmark Dataset for Multimodal Biomedical Image Registration using Deep Co-Attention Networks

Correlating neuropathology with neuroimaging findings provides a multiscale view of pathologic changes in the human organ spanning the meso- to micro-scales, and is an emerging methodology expected to shed light on numerous disease states. To gain the most information from this multimodal, multiscale approach, it is desirable to identify precisely where a histologic tissue section was taken from within the organ in order to correlate with the tissue features in exactly the same organ region. Histology-to-organ registration poses an extra challenge, as any given histologic section can capture only a small portion of a human organ. Making use of the capabilities of state-of-the-art deep learning models, we unlock the potential to address and solve such intricate challenges. Therefore, we create the ATOM benchmark dataset, sourced from diverse institutions, with the primary objective of transforming this challenge into a machine learning problem and delivering outstanding outcomes that enlighten the biomedical community. The performance of our RegisMCAN model demonstrates the potential of deep learning to accurately predict where a subregion extracted from an organ image was obtained from within the overall 3D volume. The code and dataset can be found at: https://github.com/haizailache999/Image-Registration/tree/main

cs.LG↗

arXiv2024

Prediction of turbulent energy based on low-rank resolvent modes and machine learning

A modelling framework based on the resolvent analysis and machine learning is proposed to predict the turbulent energy in incompressible channel flows. In the framework, the optimal resolvent response modes are selected as the basis functions modelling the low-rank behaviour of high-dimensional nonlinear turbulent flow-fields, and the corresponding weight functions are determined by data-driven neural networks. Turbulent-energy distribution in space and scales, at the friction Reynolds number 1000, is predicted and compared to the data of direct numerical simulation. Close agreement is observed, suggesting the feasibility and reliability of the proposed framework for turbulence prediction.

physics.flu-dyn↗

arXiv2023

A hybrid Decoder-DeepONet operator regression framework for unaligned observation data

Deep neural operators (DNOs) have been utilized to approximate nonlinear mappings between function spaces. However, DNOs face the challenge of increased dimensionality and computational cost associated with unaligned observation data. In this study, we propose a hybrid Decoder-DeepONet operator regression framework to handle unaligned data effectively. Additionally, we introduce a Multi-Decoder-DeepONet, which utilizes an average field of training data as input augmentation. The consistencies of the frameworks with the operator approximation theory are provided, on the basis of the universal approximation theorem. Two numerical experiments, Darcy problem and flow-field around an airfoil, are conducted to validate the efficiency and accuracy of the proposed methods. Results illustrate the advantages of Decoder-DeepONet and Multi-Decoder-DeepONet in handling unaligned observation data and showcase their potentials in improving prediction accuracy.

cs.LG↗

arXiv2023

PT-symmetric feedback induced linewidth narrowing

Narrow linewidth is a long-pursuing goal in precision measurement and sensing. We propose a parity-time (PT )-symmetric feedback method to narrow the linewidths of resonance systems. By using a quadrature measurement-feedback loop, we transform a dissipative resonance system into a PT-symmetric system. Unlike the conventional PT-symmetric systems which typically require two or more modes, here the PT-symmetric feedback system contains only a single resonance mode, which greatly extends the scope of applications. The method enables remarkable linewidth narrowing and enhancement of measurement sensitivity. We illustrate the concept in a thermal ensemble of atoms, achieving a 48-fold narrowing of the magnetic resonance linewidth. By applying the method in magnetometry, we realize a 22-times improvement of the measurement sensitivity. This work opens the avenue for studying non-Hermitian physics and high-precision measurements in resonance systems with feedback.

physics.optics↗

arXiv2022

Constraints on Spin-Spin-Velocity-Dependent Interaction

The existence of exotic spin-dependent forces may shine light on new physics beyond the Standard Model. We utilize two iron shielded SmCo$_5$ electron-spin sources and two optically pumped magnetometers to search for exotic long-range spin-spin-velocity-dependent force. The orientations of spin sources and magnetometers are optimized such that the exotic force is enhanced and common-mode noise is effectively subtracted. We set direct limit on proton-electron interaction in the force range from 1\,cm to 1\,km. Our experiment represents more than ten orders of magnitude improvement than previous works.

physics.atom-ph↗

arXiv2021

Using AntiPatterns to avoid MLOps Mistakes

We describe lessons learned from developing and deploying machine learning models at scale across the enterprise in a range of financial analytics applications. These lessons are presented in the form of antipatterns. Just as design patterns codify best software engineering practices, antipatterns provide a vocabulary to describe defective practices and methodologies. Here we catalog and document numerous antipatterns in financial ML operations (MLOps). Some antipatterns are due to technical errors, while others are due to not having sufficient knowledge of the surrounding context in which ML results are used. By providing a common vocabulary to discuss these situations, our intent is that antipatterns will support better documentation of issues, rapid communication between stakeholders, and faster resolution of problems. In addition to cataloging antipatterns, we describe solutions, best practices, and future directions toward MLOps maturity.

cs.LG↗

arXiv2021

Decomposition of the mean friction drag on a NACA4412 airfoil under uniform blowing/suction

The application of drag-control strategies on canonical wall-bounded turbulence, such as periodic channel and zero- or adverse-pressure-gradient boundary layers, raises the question of how to describe control effects consistently for different reference cases. We employ the RD identity (Renard & Deck, J. Fluid Mech., 790, 2016, pp. 339-367) to decompose the mean friction drag and investigate the control effects of uniform blowing and suction applied to a NACA4412 airfoil at chord Reynolds numbers Re_c=200,000 and 400,000. The connection of the drag reduction/increase by using blowing/suction with the turbulence statistics (including viscous dissipation, turbulence-kinetic-energy production, and spatial growth of the flow) across the boundary layer, subjected to adverse or favorable pressure gradients, are examined. We found that the peaks of the statistics associated with the friction-drag generation exhibit good scaling in either inner or outer units throughout the boundary layer. They are also independent of the Reynolds number, control scheme, and intensity of the blowing/suction. The small- and large-scale structures are separated with an adaptive scale-decomposition method, i.e. empirical mode decomposition (EMD), aiming to analyze the scale-specific contribution of turbulent motions to friction-drag generation. Results unveil that blowing on the suction side of the airfoil is able to enhance the contribution of large-scale motions and to suppress that of small-scales; on the other hand, suction behaves contrarily. The contributions related to cross-scale interactions remain almost unchanged with different control strategies.

physics.flu-dyn↗

arXiv2021

Training Sparse Neural Network by Constraining Synaptic Weight on Unit Lp Sphere

Sparse deep neural networks have shown their advantages over dense models with fewer parameters and higher computational efficiency. Here we demonstrate constraining the synaptic weights on unit Lp-sphere enables the flexibly control of the sparsity with p and improves the generalization ability of neural networks. Firstly, to optimize the synaptic weights constrained on unit Lp-sphere, the parameter optimization algorithm, Lp-spherical gradient descent (LpSGD) is derived from the augmented Empirical Risk Minimization condition, which is theoretically proved to be convergent. To understand the mechanism of how p affects Hoyer's sparsity, the expectation of Hoyer's sparsity under the hypothesis of gamma distribution is given and the predictions are verified at various p under different conditions. In addition, the "semi-pruning" and threshold adaptation are designed for topology evolution to effectively screen out important connections and lead the neural networks converge from the initial sparsity to the expected sparsity. Our approach is validated by experiments on benchmark datasets covering a wide range of domains. And the theoretical analysis pave the way to future works on training sparse neural networks with constrained optimization.

cs.LG↗

arXiv2020

Adaptive Trajectory Estimation with Power Limited Steering Model under Perturbation Compensation

Trajectory estimation of maneuvering objects is applied in numerous tasks like navigation, path planning and visual tracking. Many previous works get impressive results in the strictly controlled condition with accurate prior statistics and dedicated dynamic model for certain object. But in challenging conditions without dedicated dynamic model and precise prior statistics, the performance of these methods significantly declines. To solve the problem, a dynamic model called the power-limited steering model (PLS) is proposed to describe the motion of non-cooperative object. It is a natural combination of instantaneous power and instantaneous angular velocity, which relies on the nonlinearity instead of the state switching probability to achieve switching of states. And the renormalization group is introduced to compensate the nonlinear effect of perturbation in PLS model. For robust and efficient trajectory estimation, an adaptive trajectory estimation (AdaTE) algorithm is proposed. By updating the statistics and truncation time online, it corrects the estimation error caused by biased prior statistics and observation drift, while reducing the computational complexity lower than O(n). The experiment of trajectory estimation demonstrates the convergence of AdaTE, and the better robust to the biased prior statistics and the observation drift compared with EKF, UKF and sparse MAP. Other experiments demonstrate through slight modification, AdaTE can also be applied to local navigation in random obstacle environment, and trajectory optimization in visual tracking.

cs.IT↗