arXiv subjects

Zhuonan Hao

Publications and source records attributed to Zhuonan Hao.

14 recordsLinked to original sources

arXiv2026

Organizational Control Layer: Governance Infrastructure at the Execution Boundary of LLM Agent Systems

LLM-based agents are increasingly deployed in workflows where generated outputs may trigger state-changing actions, such as price offers, refunds, payments, or tool calls. This creates an execution-boundary problem: a platform must decide whether an agent's proposed action is authorized before the action is executed. We introduce the Organizational Control Layer (OCL), a model-agnostic governance layer that separates proposal generation from environment-facing execution. OCL intercepts generated actions, checks them against role, policy, and economic constraints, and either approves, revises, blocks, or escalates them without modifying the underlying LLM generator. We evaluate OCL on adversarial buyer--seller negotiation environments adapted from AgenticPay. Across multiple frontier LLM backends, OCL reduces observed unsafe executions from 88% to 0% while increasing valid success from 12% to 96%. Ablations show that this gain comes from combining pre-execution enforcement with structured recovery, rather than from prompting or blocking alone. These results suggest that deployment-grade LLM agent systems require explicit governance at the boundary between language generation and executable action.

cs.MA↗

arXiv2026

Sheet As Token: A Graph-Enhanced Representation for Multi-Sheet Spreadsheet Understanding

Workbook-scale spreadsheet understanding is increasingly important for language-model-based data analysis agents, but remains challenging because relevant information is often distributed across multiple sheets with heterogeneous schemas, layouts, and implicit relationships. Existing retrieval-augmented approaches typically decompose spreadsheets into rows, columns, or blocks to improve scalability; however, such chunk-centric representations can fragment worksheets into isolated text spans and weaken global sheet-level semantics. We propose Sheet As Token (SAT), a graph-enhanced framework that treats each worksheet as a unified semantic unit for multi-sheet spreadsheet retrieval. SAT serializes sparse schema-aware features, including sheet name, shape, and column headers, and encodes each worksheet into a compact dense token. Given a query, SAT retrieves candidates with a BGE-initialized Sheet Encoder and refines them with a gated relational GNN. In strict full-corpus evaluation, SAT reaches 0.9173 NDCG@5 on IndustryTab-614 and 0.9222 on IndustryTab-1K, relative improvements of 44.6% and 46.7% over zero-shot BGE RAG, respectively. SAT therefore improves both retrieval accuracy and serving efficiency: on IndustryTab-1K, it exceeds a Qwen3.5-9B RAG reranker by 12.5% while reducing online latency from 2.61 s to 9.24 ms, approximately 283X faster. These results show that SAT provides accurate and latency-efficient retrieval in the evaluated fixed-corpus setting. Code and data are available at https://github.com/SHITIANYU-hue/SheetasToken .

cs.AI↗

arXiv2025

A tutorial on simulating nonlinear behaviors of flexible structures with the discrete differential geometry (DDG) method

Flexible elastic structures, such as beams, rods, ribbons, plates, and shells, exhibit complex nonlinear dynamical behaviors that are central to a wide range of engineering and scientific applications, including soft robotics, deployable structures, and biomedical devices. While various numerical methods have been developed to simulate these behaviors, many conventional approaches struggle to simultaneously capture geometric and material nonlinearities, as well as nonlinear external interactions, particularly in highly deformable and dynamically evolving systems. The Discrete Differential Geometry (DDG) method has emerged as a robust and efficient numerical framework that intrinsically preserves geometric properties, accommodates material nonlinearity, and accurately models interactions with external environments and fields. By directly discretizing geometric and mechanical quantities, DDG provides an accurate, stable, and efficient approach to modeling flexible structures, addressing key limitations of traditional numerical methods. This tutorial provides a systematic introduction to the DDG method for simulating nonlinear behaviors in flexible structures. It covers DDG theory, simulation frameworks, and MATLAB implementation, with examples spanning dynamic systems, geometric and material nonlinearities, and external interactions like magnetics and fluids, culminating in practical insights and future directions. By offering a comprehensive and practical guide, together with open-source MATLAB code, this tutorial aims to facilitate the broader adoption of DDG-based numerical tools among researchers and engineers in computational mechanics, applied mathematics, and structural design.

cond-mat.soft↗

arXiv2025

Real-time simulation enabled navigation control of magnetic soft continuum robots in confined lumens

Magnetic soft continuum robots (MSCRs) have emerged as a promising technology for minimally invasive interventions, offering enhanced dexterity and remote-controlled navigation in confined lumens. Unlike conventional guidewires with pre-shaped tips, MSCRs feature a magnetic tip that actively bends under applied magnetic fields. Despite extensive studies in modeling and simulation, achieving real-time navigation control of MSCRs in confined lumens remains a significant challenge. The primary reasons are due to robot-lumen contact interactions and computational limitations in modeling MSCR nonlinear behavior under magnetic actuation. Existing approaches, such as Finite Element Method (FEM) simulations and energy-minimization techniques, suffer from high computational costs and oversimplified contact interactions, making them impractical for real-world applications. In this work, we develop a real-time simulation and navigation control framework that integrates hard-magnetic elastic rod theory, formulated within the Discrete Differential Geometry (DDG) framework, with an order-reduced contact handling strategy. Our approach captures large deformations and complex interactions while maintaining computational efficiency. Next, the navigation control problem is formulated as an inverse design task, where optimal magnetic fields are computed in real time by minimizing the constrained forces and enhancing navigation accuracy. We validate the proposed framework through comprehensive numerical simulations and experimental studies, demonstrating its robustness, efficiency, and accuracy. The results show that our method significantly reduces computational costs while maintaining high-fidelity modeling, making it feasible for real-time deployment in clinical settings.

cond-mat.soft↗

arXiv2025

Harnessing Discrete Differential Geometry: A Virtual Playground for the Bilayer Soft Robotics

Soft robots have garnered significant attention due to their promising applications across various domains. A hallmark of these systems is their bilayer structure, where strain mismatch caused by differential expansion between layers induces complex deformations. Despite progress in theoretical modeling and numerical simulation, accurately capturing their dynamic behavior, especially during environmental interactions, remains challenging. This study presents a novel simulation environment based on the Discrete Elastic Rod (DER) model to address the challenge. By leveraging discrete differential geometry (DDG), the DER approach offers superior convergence compared to conventional methods like Finite Element Method (FEM), particularly in handling contact interactions -- an essential aspect of soft robot dynamics in real-world scenarios. Our simulation framework incorporates key features of bilayer structures, including stretching, bending, twisting, and inter-layer coupling. This enables the exploration of a wide range of dynamic behaviors for bilayer soft robots, such as gripping, crawling, jumping, and swimming. The insights gained from this work provide a robust foundation for the design and control of advanced bilayer soft robotic systems.

cs.RO↗

arXiv2024

Bio-Inspired Pneumatic Modular Actuator for Peristaltic Transport

While its biological significance is well-documented, its application in soft robotics, particularly for the transport of fragile and irregularly shaped objects, remains underexplored. This study presents a modular soft robotic actuator system that addresses these challenges through a scalable, adaptable, and repairable framework, offering a cost-effective solution for versatile applications. The system integrates optimized donut-shaped actuation modules and utilizes real-time pressure feedback for synchronized operation, ensuring efficient object grasping and transport without relying on intricate sensing or control algorithms. Experimental results validate the system`s ability to accommodate objects with varying geometries and material characteristics, balancing robustness with flexibility. This work advances the principles of peristaltic actuation, establishing a pathway for safely and reliably manipulating delicate materials in a range of scenarios.

cs.RO↗

arXiv2024

Inverse Design of Snap-Actuated Jumping Robots Powered by Mechanics-Aided Machine Learning

Exploring the design and control strategies of soft robots through simulation is highly attractive due to its cost-effectiveness. Although many existing models (e.g., finite element analysis) are effective for simulating soft robotic dynamics, there remains a need for a general and efficient numerical simulation approach in the soft robotics community. In this paper, we develop a discrete differential geometry-based numerical framework to achieve the model-based inverse design of a novel snap-actuated jumping robot. It is found that the dynamic process of a snapping beam can be either symmetric or asymmetric, such that the trajectory of the jumping robot can be tunable (e.g., horizontal or vertical). By employing this novel mechanism of the bistable beam as the robotic actuator, we next propose a physics-data hybrid inverse design strategy for the snap-jump robot with a broad spectrum of jumping capabilities. We first use the physical engine to study the influences of the robot's design parameters on the jumping capabilities, then generate extensive simulation data to formulate a data-driven inverse design solution. The inverse design solution can rapidly explore the combination of design parameters for achieving a target jump, which provides valuable guidance for the fabrication and control of the jumping robot. The proposed methodology paves the way for exploring the design and control insights of soft robots with the help of simulations.

cs.RO↗

arXiv2024

Inverse Design of Planar Clamped-Free Elastic Rods from Noisy Data

Slender structures, such as rods, often exhibit large nonlinear geometrical deformations even under moderate external forces (e.g., gravity). This characteristic results in a rich variety of morphological changes, making them appealing for engineering design and applications, such as soft robots, submarine cables, decorative knots, and more. Prior studies have demonstrated that the natural shape of a rod significantly influences its deformed geometry. Consequently, the natural shape of the rod should be considered when manufacturing and designing rod-like structures. Here, we focus on an inverse problem: can we determine the natural shape of a suspended 2D planar rod so that it deforms into a desired target shape? We begin by formulating a theoretical framework based on the statics of planar rod equilibrium that can compute the natural shape of a planar rod given its target shape. Furthermore, we analyze the impact of uncertainties (e.g., noise in the data) on the accuracy of the theoretical framework. The results reveal the shortcomings of the theoretical framework in handling uncertainties in the inverse problem, a fact often overlooked in previous works. To mitigate the influence of the uncertainties, we combine the statics of the planar rod with the adjoint method for parameter sensitivity analysis, constructing a learning framework that can efficiently explore the natural shape of the designed rod with enhanced robustness. This framework is validated numerically for its accuracy and robustness, offering valuable insights into the inverse design of soft structures for various applications, including soft robotics and animation of morphing structures.

cond-mat.soft↗

arXiv2024

Bundling and Tumbling in Bacterial-inspired Bi-flagellated Soft Robots for Attitude Adjustment

We create a mechanism inspired by bacterial swimmers, featuring two flexible flagella with individual control over rotation speed and direction in viscous fluid environments. Using readily available materials, we design and fabricate silicone-based helical flagella. To simulate the robot's motion, we develop a physics-based computational tool, drawing inspiration from computer graphics. The framework incorporates the Discrete Elastic Rod method, modeling the flagella as Kirchhoff's elastic rods, and couples it with the Regularized Stokeslet Segments method for hydrodynamics, along with the Implicit Contact Model to handle contact. This approach effectively captures polymorphic phenomena like bundling and tumbling. Our study reveals how these emergent behaviors affect the robot's attitude angles, demonstrating its ability to self-reorient in both simulations and experiments. We anticipate that this framework will enhance our understanding of the directional change capabilities of flagellated robots, potentially stimulating further exploration on microscopic robot mobility.

cs.RO↗

arXiv2023

Modeling, Characterization, and Control of Bacteria-inspired Bi-flagellated Mechanism with Tumbling

Multi-flagellated bacteria utilize the hydrodynamic interaction between their filamentary tails, known as flagella, to swim and change their swimming direction in low Reynolds number flow. This interaction, referred to as bundling and tumbling, is often overlooked in simplified hydrodynamic models such as Resistive Force Theories (RFT). However, for the development of efficient and steerable robots inspired by bacteria, it becomes crucial to exploit this interaction. In this paper, we present the construction of a macroscopic bio-inspired robot featuring two rigid flagella arranged as right-handed helices, along with a cylindrical head. By rotating the flagella in opposite directions, the robot's body can reorient itself through repeatable and controllable tumbling. To accurately model this bi-flagellated mechanism in low Reynolds flow, we employ a coupling of rigid body dynamics and the method of Regularized Stokeslet Segments (RSS). Unlike RFT, RSS takes into account the hydrodynamic interaction between distant filamentary structures. Furthermore, we delve into the exploration of the parameter space to optimize the propulsion and torque of the system. To achieve the desired reorientation of the robot, we propose a tumble control scheme that involves modulating the rotation direction and speed of the two flagella. By implementing this scheme, the robot can effectively reorient itself to attain the desired attitude. Notably, the overall scheme boasts a simplified design and control as it only requires two control inputs. With our macroscopic framework serving as a foundation, we envision the eventual miniaturization of this technology to construct mobile and controllable micro-scale bacterial robots.

cs.RO↗

arXiv2021

Lateral contact yields longitudinal cohesion in active undulatory systems

Many animals and robots move using undulatory motion of their bodies. When in close proximity undulatory motion can lead to novel collective behaviors such as gait synchronization, spatial reconfiguration, and clustering. Here we study the role of contact interactions between model undulatory swimmers: three-link robots in experiment and multi-link robots in simulation. The undulatory gait of each swimmer is generated through a time-dependent sinusoidal-like waveform which has a fixed phase offset, $ϕ$. By varying the phase relationship between neighboring swimmers we seek to study how contact forces and spatial configurations are governed by the phase difference between neighboring swimmers. We find that undulatory actuation in close proximity drives neighboring swimmers into spatial equilibrium configurations that depend on the actuation phase difference. We propose a model for spatial equilibrium of nearest neighbor undulatory swimmers which we call the gait compatibility condition, which is the set of spatial and gait configurations in which no collisions occur. Robotic experiments with two, three, and four swimmers exhibit good agreement with the compatibility model. To probe the interaction potential between undulatory swimmers we perturb the each longitudinally from their equilibrium configurations and we measure their steady-state displacement. These studies reveal that undulatory swimmers in close proximity exhibit cohesive longitudinal interaction forces that drive the swimmers from incompatible to compatible configurations. This system of undulatory swimmers provides new insight into active-matter systems which move through body undulation. In addition to the importance of velocity and orientation coherence in active-matter swarms, we demonstrate that undulatory phase coherence is also important for generating stable, cohesive group configurations.

nlin.AO↗

arXiv2021

Collective synchronization of undulatory movement through contact

Many biological systems synchronize their movement through physical interactions. By far the most well studied examples concern physical interactions through a fluid: beating cilia, swimming sperm and worms, and flapping wings, all display synchronization behavior through fluid mechanical interactions. However, as the density of a collective increases individuals may also interact with each other through physical contact. In the field of "active matter" systems, it is well known that inelastic contact between individuals can produce long-range correlations in position, orientation, and velocity. In this work we demonstrate that contact interactions between undulating robots yield novel phase dynamics such as synchronized motions. We consider undulatory systems in which rhythmic motion emerges from time-independent oscillators that sense and respond to undulatory bending angle and speed. In pair experiments we demonstrate that robot joints will synchronize to in-phase and anti-phase oscillations through collisions and a phase-oscillator model describes the stability of these modes. To understand how contact interactions influence the phase dynamics of larger groups we perform simulations and experiments of simple three-link undulatory robots that interact only through contact. Collectives synchronize their movements through contact as predicted by the theory and when the robots can adjust their position in response to contact we no longer observe anti-phase synchronization. Lastly we demonstrate that synchronization dramatically reduces the interaction forces within confined groups of undulatory robots indicating significant energetic and safety benefits from group synchronization. The theory and experiments in this study illustrate how contact interactions in undulatory active matter can lead to novel collective motion and synchronization.

nlin.AO↗

arXiv2020

Design and Practice of the Regulation in Speed with Flywheel

This paper aims at the teaching contents of "Fluctuation and Regulation in Speed of Machines" for students, explores the effect of the flywheel on the speed regulation of the mechanism system and its influencing factors, designs an experimental device with applied value, and develops a comparatively complete experimental scheme by contrasting experimental principles. Using this experimental device, we can in-depth study of the effect of the flywheel on the speed of different drive shafts, and deepen students' understanding and mastery of knowledge about regulation in speed.

physics.ed-ph↗

arXiv2020

A Simple Fix for Convolutional Neural Network via Coordinate Embedding

Convolutional Neural Networks (CNN) has been widely applied in the realm of computer vision. However, given the fact that CNN models are translation invariant, they are not aware of the coordinate information of each pixel. Thus the generalization ability of CNN will be limited since the coordinate information is crucial for a model to learn affine transformations which directly operate on the coordinate of each pixel. In this project, we proposed a simple approach to incorporate the coordinate information to the CNN model through coordinate embedding. Our approach does not change the downstream model architecture and can be easily applied to the pre-trained models for the task like object detection. Our experiments on the German Traffic Sign Detection Benchmark show that our approach not only significantly improve the model performance but also have better robustness with respect to the affine transformation.

cs.CV↗