arXiv subjects

Yu Tang

Publications and source records attributed to Yu Tang.

At least 37 records · Page 2Linked to original sources

arXiv2025

Geometric Control of Mechanical Systems with Symmetries Based on Sliding Modes

In this paper, we propose a framework for designing sliding mode controllers for a class of mechanical systems with symmetry, both unconstrained and constrained, that evolve on principal fiber bundles. Control laws are developed based on the reduced motion equations by exploring symmetries, leading to a sliding mode control strategy where the reaching stage is executed on the base space, and the sliding stage is performed on the structure group. Thus, design complexity is reduced, and difficult choices for coordinate representations when working with a particular Lie group are avoided. For this purpose, a sliding subgroup is constructed on the structure group based on a kinematic controller, and the sliding variable will converge to the identity of the state manifold upon reaching the sliding subgroup. A reaching law based on a general sliding vector field is then designed on the base space using the local form of the mechanical connection to drive the sliding variable to the sliding subgroup, and its time evolution is given according to the appropriate covariant derivative. Almost global asymptotic stability and local exponential stability are demonstrated using a Lyapunov analysis. We apply the results to a fully actuated system (a rigid spacecraft actuated by reaction wheels) and a subactuated nonholonomic system (unicycle mobile robot actuated by wheels), which is also simulated for illustration.

cs.RO↗

arXiv2025

GOSU: Retrieval-Augmented Generation with Global-Level Optimized Semantic Unit-Centric Framework

Building upon the standard graph-based Retrieval-Augmented Generation (RAG), the introduction of heterogeneous graphs and hypergraphs aims to enrich retrieval and generation by leveraging the relationships between multiple entities through the concept of semantic units (SUs). But this also raises a key issue: The extraction of high-level SUs limited to local text chunks is prone to ambiguity, complex coupling, and increased retrieval overhead due to the lack of global knowledge or the neglect of fine-grained relationships. To address these issues, we propose GOSU, a semantic unit-centric RAG framework that efficiently performs global disambiguation and utilizes SUs to capture interconnections between different nodes across the global context. In the graph construction phase, GOSU performs global merging on the pre-extracted SUs from local text chunks and guides entity and relationship extraction, reducing the difficulty of coreference resolution while uncovering global semantic objects across text chunks. In the retrieval and generation phase, we introduce hierarchical keyword extraction and semantic unit completion. The former uncovers the fine-grained binary relationships overlooked by the latter, while the latter compensates for the coarse-grained n-ary relationships missing from the former. Evaluation across multiple tasks demonstrates that GOSU outperforms the baseline RAG methods in terms of generation quality.

cs.CL↗

arXiv2025

MoE-TTS: Enhancing Out-of-Domain Text Understanding for Description-based TTS via Mixture-of-Experts

Description-based text-to-speech (TTS) models exhibit strong performance on in-domain text descriptions, i.e., those encountered during training. However, in real-world applications, the diverse range of user-generated descriptions inevitably introduces numerous out-of-domain inputs that challenge the text understanding capabilities of these systems. To address this issue, we propose MoE-TTS, a description-based TTS model designed to enhance the understanding of out-of-domain text descriptions. MoE-TTS employs a modality-based mixture-of-experts (MoE) approach to augment a pre-trained textual large language model (LLM) with a set of specialized weights adapted to the speech modality while maintaining the original LLM frozen during training. This approach allows MoE-TTS to effectively leverage the pre-trained knowledge and text understanding abilities of textual LLMs. Our experimental results indicate that: first, even the most advanced closed-source commercial products can be challenged by carefully designed out-of-domain description test sets; second, MoE-TTS achieves superior performance in generating speech that more accurately reflects the descriptions. We encourage readers to listen to the demos at https://welkinyang.github.io/MoE-TTS/.

eess.AS↗

arXiv2025

Split-as-a-Pro: behavioral control via operator splitting and alternating projections

The paper introduces Split-as-a-Pro, a control framework that integrates behavioral systems theory, operator splitting methods, and alternating projection algorithms. The framework reduces dynamic optimization problems - arising in both control and estimation - to efficient projection computations. Split-as-a-Pro builds on a non-parametric formulation that exploits system structure to separate dynamic constraints imposed by individual subsystems from external ones, such as interconnection constraints and input/output constraints. This enables the use of arbitrary system representations, as long as the associated projection is efficiently computable, thereby enhancing scalability and compatibility with gray-box modeling. We demonstrate the effectiveness of Split-as-a-Pro by developing a distributed algorithm for solving finite-horizon linear quadratic control problems and illustrate its use in predictive control. Our numerical case studies show that algorithms obtained using Split-as-a-Pro significantly outperform their centralized counterparts in runtime and scalability across various standard graph topologies, while seamlessly leveraging both model-based and data-driven system representations.

math.OC↗

arXiv2025

Pricing is All You Need to Improve Traffic Routing

We investigate the design of pricing policies that enhance driver adherence to route guidance, ensuring effective routing control. The major novelty lies in that we adopt a Markov chain to model drivers' compliance rates conditioned on both traffic states and tolls. By formulating the managed traffic network as a nonlinear stochastic dynamical system, we can quantify in a more realistic way the impacts of driver route choices and thus determine appropriate tolls. Specially, we focus on a network comprised of one corridor and one local street. We assume that a reasonable routing policy is specified in advance. However, drivers could be reluctant to be detoured. Thus a fixed toll is set on the corridor to give drivers incentives to choose the local street. We evaluate the effectiveness of the given routing and pricing policies via stability analysis. We suggest using the stability and instability conditions to establish lower and upper bounds on throughput. This allows us to select suitable tolls that maximize these bounds.

eess.SY↗

arXiv2025

Observational signature of Lorentz violation in acceleration radiation

In recent years, Lorentz violation (LV) has emerged as a vibrant area of research in fundamental physics. Despite predictions from quantum gravity theories that Lorentz symmetry may break down at Planck-scale energies, which are currently beyond experimental reach, its low-energy signatures could still be detectable through alternative methods. In this paper, we propose a quantum optical approach to investigate potential LV effects on the acceleration radiation of a freely falling atom within a black hole spacetime coupled to a Lorentz-violating vector field. Our proposed experimental setup employs a Casimir-type apparatus, wherein a two-level atom serves as a dipole detector, enabling its interaction with the field to be modeled using principles from quantum optics. We demonstrate that LV can introduce distinct quantum signatures into the radiation flux, thereby significantly modulating particle emission rates. It is found that while LV effects are negligible at high mode frequencies, they become increasingly pronounced at lower frequencies. This suggests that detecting LV at low-energy scales may depend on advancements in low-frequency observational techniques or detectors.

gr-qc↗

arXiv2025

Resilient Control of Dynamic Flow Networks Subject to Stochastic Cyber-Physical Disruptions

Modern network systems, such as transportation and communication systems, are prone to cyber-physical disruptions and thus suffer efficiency loss. This paper studies network resiliency, in terms of throughput, and develops resilient control to improve throughput. We consider single-commodity networks that admit congestion propagation. We also apply a Markov process to model disruption switches. For throughput analysis, we first use insights into congestion spillback to propose novel Lyapunov functions and then exploit monotone network dynamics to reduce computational costs of verifying stability conditions. For control design, we show that (i) for a network with infinite link storage space, there exists an open-loop control that attains the min-expected-cut capacity; (ii) for a network with observable disruptions that restrict maximum sending and/or receiving flows, there exists a mode-dependent control that attains the expected-min-cut capacity; (iii) for general networks, there exists a closed-loop control with throughput guarantees. We also derive lower bounds of resiliency scores for a set of numerical examples and verify resiliency improvement with our method.

eess.SY↗

arXiv2024

Geo-ConvGRU: Geographically Masked Convolutional Gated Recurrent Unit for Bird-Eye View Segmentation

Convolutional Neural Networks (CNNs) have significantly impacted various computer vision tasks, however, they inherently struggle to model long-range dependencies explicitly due to the localized nature of convolution operations. Although Transformers have addressed limitations in long-range dependencies for the spatial dimension, the temporal dimension remains underexplored. In this paper, we first highlight that 3D CNNs exhibit limitations in capturing long-range temporal dependencies. Though Transformers mitigate spatial dimension issues, they result in a considerable increase in parameter and processing speed reduction. To overcome these challenges, we introduce a simple yet effective module, Geographically Masked Convolutional Gated Recurrent Unit (Geo-ConvGRU), tailored for Bird's-Eye View segmentation. Specifically, we substitute the 3D CNN layers with ConvGRU in the temporal module to bolster the capacity of networks for handling temporal dependencies. Additionally, we integrate a geographical mask into the Convolutional Gated Recurrent Unit to suppress noise introduced by the temporal module. Comprehensive experiments conducted on the NuScenes dataset substantiate the merits of the proposed Geo-ConvGRU, revealing that our approach attains state-of-the-art performance in Bird's-Eye View segmentation.

cs.CV↗

arXiv2024

Microheterogeneity and Single-Molecule Mixture Science: An Introduction

Microheterogeneity, as a fundamental natural property, is widely presented in a range of microscopic systems such as polymer systems, biomacromolecular systems, and nanosystems, however, the construction of molecular systems of this form has not yet been realized at the level of organic molecules with well-defined structural compositions. Inspired by the "chemical space" concept, I recently reported a study of the single molecule mixture state, In this preprint, a detailed discussion of microheterogeneity and the concept of "single-molecule mixture" has been provided.

physics.chem-ph↗

arXiv2024

Half-KFN: An Enhanced Detection Method for Subtle Covariate Drift

Detecting covariate drift is a common task of significant practical value in supervised learning. Once covariate drift occurs, the models may no longer be applicable, hence numerous studies have been devoted to the advancement of detection methods. However, current research methods are not particularly effective in handling subtle covariate drift when dealing with small proportions of drift samples. In this paper, inspired by the $k$-nearest neighbor (KNN) approach, a novel method called Half $k$-farthest neighbor (Half-KFN) is proposed in response to specific scenarios. Compared to traditional ones, Half-KFN exhibits higher power due to the inherent capability of the farthest neighbors which could better characterize the nature of drift. Furthermore, with larger sample sizes, the employment of the bootstrap for hypothesis testing is recommended. It is leveraged to calculate $p$-values dramatically faster than permutation tests, with speed undergoing an exponential growth as sample size increases. Numerical experiments on simulated and real data are conducted to evaluate our proposed method, and the results demonstrate that it consistently displays superior sensitivity and rapidity in covariate drift detection across various cases.

stat.ME↗

arXiv2024

Application of Data-Driven Model Predictive Control for Autonomous Vehicle Steering

With the development of autonomous driving technology, there are increasing demands for vehicle control, and MPC has become a widely researched topic in both industry and academia. Existing MPC control methods based on vehicle kinematics or dynamics have challenges such as difficult modeling, numerous parameters, strong nonlinearity, and high computational cost. To address these issues, this paper adapts an existing Data-driven MPC control method and applies it to autonomous vehicle steering control. This method avoids the need for complex vehicle system modeling and achieves trajectory tracking with relatively low computational time and small errors. We validate the control effectiveness of the algorithm in specific scenario through CarSim-Simulink simulation and perform comparative analysis with PID and vehicle kinematics MPC, confirming the feasibility and superiority of it for vehicle steering control.

eess.SY↗

arXiv2024

Semi-Supervised Learning for Visual Bird's Eye View Semantic Segmentation

Visual bird's eye view (BEV) semantic segmentation helps autonomous vehicles understand the surrounding environment only from images, including static elements (e.g., roads) and dynamic elements (e.g., vehicles, pedestrians). However, the high cost of annotation procedures of full-supervised methods limits the capability of the visual BEV semantic segmentation, which usually needs HD maps, 3D object bounding boxes, and camera extrinsic matrixes. In this paper, we present a novel semi-supervised framework for visual BEV semantic segmentation to boost performance by exploiting unlabeled images during the training. A consistency loss that makes full use of unlabeled data is then proposed to constrain the model on not only semantic prediction but also the BEV feature. Furthermore, we propose a novel and effective data augmentation method named conjoint rotation which reasonably augments the dataset while maintaining the geometric relationship between the front-view images and the BEV semantic segmentation. Extensive experiments on the nuScenes and Argoverse datasets show that our semi-supervised framework can effectively improve prediction accuracy. To the best of our knowledge, this is the first work that explores improving visual BEV semantic segmentation performance using unlabeled data. The code is available at https://github.com/Junyu-Z/Semi-BEVseg

cs.CV↗

arXiv2023

How Does Driver Non-compliance Destroy Traffic Routing Control?

Routing control is one of important traffic management strategies against urban congestion. However, it could be compromised by heterogeneous driver non-compliance with routing instructions. In this article we model the compliance in a stochastic manner and investigate its impacts on routing control. We consider traffic routing for two parallel links. Particularly, we consider two scenarios: one ignores congestion spillback while the other one considers it. We formulate the problem as a non-linear Markov chain, given random compliance rates. Then we propose the stability and instability conditions to reveal when the routing is able or unable to stabilize the traffic. We show that for links without congestion spillback there exists a sufficient and necessary stability criterion. For links admiting congestion propagation, we present one stability condition and one instability condition. This stability conditions allow us to quantify the impacts of driver non-compliance on the two-link network in terms of throughput. Finally, we illustrate the results with a set of numerical examples.

eess.SY↗

arXiv2023

Geometric Tracking on $\mathcal{S}^{3}$ Based on Sliding Mode Control

Attitude tracking on the unit sphere of dimension $3$ based on sliding mode is considered in this paper. The tangent bundle of Lagrangian dynamics that describes the rotational motion of a rigid body is first shown to be a Lie group, and then a sliding surface that emerged on it is defined. Next, a sliding-mode controller is designed for attitude tracking that relies on an intrinsic error defined on the Lie group. Almost global asymptotic stability of the closed loop is demonstrated using the Lyapunov analysis. Numerical simulations are included to compare the performance of the sliding mode controller designed on the Lie group with that designed in the embedding Euclidean space.

eess.SY↗

arXiv2023

Physics-informed Machine Learning for Calibrating Macroscopic Traffic Flow Models

Well-calibrated traffic flow models are fundamental to understanding traffic phenomena and designing control strategies. Traditional calibration has been developed base on optimization methods. In this paper, we propose a novel physics-informed, learning-based calibration approach that achieves performances comparable to and even better than those of optimization-based methods. To this end, we combine the classical deep autoencoder, an unsupervised machine learning model consisting of one encoder and one decoder, with traffic flow models. Our approach informs the decoder of the physical traffic flow models and thus induces the encoder to yield reasonable traffic parameters given flow and speed measurements. We also introduce the denoising autoencoder into our method so that it can handles not only with normal data but also with corrupted data with missing values. We verified our approach with a case study of I-210 E in California.

cs.LG↗

arXiv2023

Geometric sliding mode control of mechanical systems on Lie groups

This paper presents a generalization of conventional sliding mode control designs for systems in Euclidean spaces to fully actuated simple mechanical systems whose configuration space is a Lie group for the trajectory-tracking problem. A generic kinematic control is first devised in the underlying Lie algebra, which enables the construction of a Lie group on the tangent bundle where the system state evolves. A sliding subgroup is then proposed on the tangent bundle with the desired sliding properties, and a control law is designed for the error dynamics trajectories to reach the sliding subgroup globally exponentially. Tracking control is then composed of the reaching law and sliding mode, and is applied for attitude tracking on the special orthogonal group SO(3) and the unit sphere S3. Numerical simulations show the performance of the proposed geometric sliding-mode controller (GSMC) in contrast with two control schemes of the literature.

math.OC↗

arXiv2023

Robust Queue Length Estimation for Ramp Metering in a Connected Vehicle Environment

Connected vehicles (CVs) can provide numerous new data via vehicle-to-vehicle or vehicle-to-infrastructure communication. These data can in turn be used to facilitate real-time traffic state estimation. In this paper, we focus on ramp queue length estimation in a connected vehicle environment, which improves control design and implementation of ramp metering algorithms. One major challenge of the estimation problem is that the CV data only represent partial traffic observations and could introduce new uncertainties if real-time CV penetration rates are unknown. To address this, we build our estimation approach on both traditional freeway sensors and new CV data. We first formulate a ramp queue model that considers i) variations in the penetration rate and ii) noise in measurements. Then we develop a robust filter that minimizes the impacts of these two kinds of uncertainties on queue estimation. More importantly, we show that the designed filter has guaranteed long-term estimation accuracy. It allows us to quantify in a theoretical way the relationship between estimation error and fluctuation of CV penetration rates. We also provide a series of simulation results to verify our approach.

eess.SY↗

arXiv2022

Attitude Tracking for Rigid Bodies Using Vector and Biased Gyro Measurements

The rigid-body attitude tracking using vector and biased gyro measurements with unknown inertia matrix is studied in this note. First, a gyro-bias observer with global exponential stability is designed. Then, an attitude tracking controller based on this observer is devised, ensuring almost global asymptotic stability and almost semiglobal exponential stability. A separation property of the combined observer-controller is proved. Lastly, an adaptive attitude tracking controller relying on a modified gyro-bias observer and with no over-parametrization is developed to deal with the unknown inertia matrix. The proposed control schemes require neither an explicit attitude representation nor any attitude estimation, but only the measurement of at least two non-collinear known inertial reference vectors and biased gyro rate, which can be obtained by common low-cost IMU sensors. Simulations are included to illustrate the proposed adaptive controller under noisy measurements.

eess.SY↗