arXiv subjects

Zezhi Tang

Publications and source records attributed to Zezhi Tang.

10 recordsLinked to original sources

arXiv2026

WAM-OPD: On-Policy Distillation for World Action Models

World action models (WAMs) couple visual future prediction with robot action generation, but accelerated students can lose task capabilities during distillation and later encounter states that are poorly represented by offline data. We study whether on-policy distillation (OPD) can repair such a student without requiring sparse-reward reinforcement learning. We introduce WAM-OPD, a deployment-consistent post-training recipe for a video-first WAM. The student acts in the environment and therefore determines the history distribution. A frozen teacher labels those student histories with coherent video and action targets, while the student action branch is trained under its own generated video plan, as it is at deployment. Joint video and action losses update lightweight adapters in the shared backbone, together with an action flow-matching regularizer. In preliminary RoboTwin 2.0 studies on two tasks, the released one-video/one-action-step Flash-WAM improves from 0.0% to 58.3% success on HANDOVER MIC, and from 16.7% to 33.3% on PUT OBJECT CABINET. These task-specific results are an initial capability proof rather than evidence of broad or uniform generalization. They nevertheless suggest that dense teacher supervision on student-induced histories is a promising post-training interface for video-first WAMs.

cs.AI↗

arXiv2026

Guided Action Flow: Q-Guided Inference for Flow-Matching Vision-Language-Action Policies

Deploying a pretrained flow-matching vision-language-action (VLA) policy on a particular robot and workspace often calls for task-specific adaptation, while full- policy fine-tuning is costly and changes the base behavior. We present Guided Action Flow, an inference-time method that keeps a pretrained SmolVLA policy frozen and steers its reverse-time action-flow sampling with gradients from a task-specific action-chunk critic. QGF trains a visual Transformer critic and value model with offline Implicit Q-Learning on 100 real-robot rollouts. The critic conditions on robot state, frozen dual-camera SmolVLA visual tokens, and the policy's normalized 50-step action chunk. On a real-robot water-bottle placement task, QGF with \b{eta} = 2 increases success from 19/40 episodes (47.5%) to 34/40 episodes (85.0%) and reduces timeouts from 13 to 3. With a yellow tape measure added as a visual distractor, QGF completes 6/12 episodes, compared with 0/11 for frozen SmolVLA. These results show that deployment rollouts can provide an effective action-space guidance signal for improving a frozen flow-matching VLA without policy fine-tuning.

cs.RO↗

arXiv2026

SignVLA: Real-Time Sign Language-Guided Robotic Manipulation via Attention LSTM and Vision-Language-Action Models

Vision-Language-Action (VLA) models enable robots to execute manipulation tasks from natural-language instructions grounded in visual observations. However, existing VLA interfaces primarily rely on speech or text input, limiting accessibility for deaf, hard-of-hearing, and speech-impaired users. We present SignVLA, a real-time sign-language-guided VLA framework for accessible human-robot interaction. The system introduces a modular sign-to-text interface that converts visual sign gestures into semantic instructions compatible with downstream VLA policies. Given video streams, SignVLA extracts hand landmark features and employs an attention-enhanced Long Short-Term Memory (LSTM) network to capture temporal gesture dynamics for alphabet- and command-level sign recognition. A temporal stabilization module further improves prediction consistency in real-time interaction settings.The generated instruction sequence is then passed to a downstream VLA policy for sign-conditioned robotic manipulation. Experimental results demonstrate stable real-time sign recognition and successful execution of manipulation tasks driven by sign-language inputs. Our findings suggest that lightweight temporal sign recognition can serve as an effective and practical accessibility layer for multimodal embodied intelligence.

cs.AI↗

arXiv2026

SignVLA: A Gloss-Free Vision-Language-Action Framework for Real-Time Sign Language-Guided Robotic Manipulation

We present, to our knowledge, the first sign language-driven Vision-Language-Action (VLA) framework for intuitive and inclusive human-robot interaction. Unlike conventional approaches that rely on gloss annotations as intermediate supervision, the proposed system adopts a gloss-free paradigm and directly maps visual sign gestures to semantic instructions. This design reduces annotation cost and avoids the information loss introduced by gloss representations, enabling more natural and scalable multimodal interaction. In this work, we focus on a real-time alphabet-level finger-spelling interface that provides a robust and low-latency communication channel for robotic control. Compared with large-scale continuous sign language recognition, alphabet-level interaction offers improved reliability, interpretability, and deployment feasibility in safety-critical embodied environments. The proposed pipeline transforms continuous gesture streams into coherent language commands through geometric normalization, temporal smoothing, and lexical refinement, ensuring stable and consistent interaction. Furthermore, the framework is designed to support future integration of transformer-based gloss-free sign language models, enabling scalable word-level and sentence-level semantic understanding. Experimental results demonstrate the effectiveness of the proposed system in grounding sign-derived instructions into precise robotic actions under diverse interaction scenarios. These results highlight the potential of the framework to advance accessible, scalable, and multimodal embodied intelligence.

cs.RO↗

arXiv2026

Disturbance observer-based tracking control for roll-to-roll slot die coating systems under gap and pump rate disturbances

Roll-to-roll slot die coating is a widely used industrial manufacturing technique applied in a diverse range of applications such as the production of lithium-ion batteries, solar cells and optical films. The efficiency of roll-to-roll slot die coating depends on the precise control of various input parameters such as pump rate, substrate velocity and coating gap. However, these inputs are sensitive to disturbances in process conditions, leading to inconsistencies in the various characteristics of the produced film. To address this challenge, a \gls{DO} is utilized for detecting disturbances, which may occur in the same or different channels as the control signal within the system. A generalized compensator is then implemented to mitigate the impact of these disturbances on the output, thereby enhancing uncertainty suppression. Additionally, integrating the disturbance rejection system with an output tracking controller enables the coating system to maintain the desired thickness under varying input conditions and disturbances. The effectiveness of this approach is then validated using a test rig equipped with a camera system, which facilitates the development of a data-driven model of the dynamic process, represented by state-space equations. The simulation results were demonstrated to showcase the effectiveness of the DOBOTC system, which provides a resilient solution for the output tracking issue in a data-driven model with generalized disturbances.

eess.SY↗

arXiv2025

Deep Reinforcement Learning Optimization for Uncertain Nonlinear Systems via Event-Triggered Robust Adaptive Dynamic Programming

This work proposes a unified control architecture that couples a Reinforcement Learning (RL)-driven controller with a disturbance-rejection Extended State Observer (ESO), complemented by an Event-Triggered Mechanism (ETM) to limit unnecessary computations. The ESO is utilized to estimate the system states and the lumped disturbance in real time, forming the foundation for effective disturbance compensation. To obtain near-optimal behavior without an accurate system description, a value-iteration-based Adaptive Dynamic Programming (ADP) method is adopted for policy approximation. The inclusion of the ETM ensures that parameter updates of the learning module are executed only when the state deviation surpasses a predefined bound, thereby preventing excessive learning activity and substantially reducing computational load. A Lyapunov-oriented analysis is used to characterize the stability properties of the resulting closed-loop system. Numerical experiments further confirm that the developed approach maintains strong control performance and disturbance tolerance, while achieving a significant reduction in sampling and processing effort compared with standard time-triggered ADP schemes.

math.OC↗

arXiv2025

Hierarchical Testing with Rabbit Optimization for Industrial Cyber-Physical Systems

This paper presents HERO (Hierarchical Testing with Rabbit Optimization), a novel black-box adversarial testing framework for evaluating the robustness of deep learning-based Prognostics and Health Management systems in Industrial Cyber-Physical Systems. Leveraging Artificial Rabbit Optimization, HERO generates physically constrained adversarial examples that align with real-world data distributions via global and local perspective. Its generalizability ensures applicability across diverse ICPS scenarios. This study specifically focuses on the Proton Exchange Membrane Fuel Cell system, chosen for its highly dynamic operational conditions, complex degradation mechanisms, and increasing integration into ICPS as a sustainable and efficient energy solution. Experimental results highlight HERO's ability to uncover vulnerabilities in even state-of-the-art PHM models, underscoring the critical need for enhanced robustness in real-world applications. By addressing these challenges, HERO demonstrates its potential to advance more resilient PHM systems across a wide range of ICPS domains.

cs.LG↗

arXiv2025

A temporal scale transformer framework for precise remaining useful life prediction in fuel cells

In exploring Predictive Health Management (PHM) strategies for Proton Exchange Membrane Fuel Cells (PEMFC), the Transformer model, widely used in data-driven approaches, excels in many fields but struggles with time series analysis due to its self-attention mechanism, which yields a complexity of the input sequence squared and low computational efficiency. It also faces challenges in capturing both global long-term dependencies and local details effectively. To tackle this, we propose the Temporal Scale Transformer (TSTransformer), an enhanced version of the inverted Transformer (iTransformer). Unlike traditional Transformers that treat each timestep as an input token, TSTransformer maps sequences of varying lengths into tokens at different stages for inter-sequence modeling, using attention to capture multivariate correlations and feed-forward networks (FFN) to encode sequence representations. By integrating a one-dimensional convolutional layer into the multivariate attention for multi-level scaling of K and V matrices, it improves local feature extraction, captures temporal scale characteristics, and reduces token count and computational costs. Experiments comparing TSTransformer with models like Long Short-Term Memory, iTransformer, and Transformer demonstrate its potential as a powerful tool for advancing PHM in renewable energy, effectively addressing the limitations of pure Transformer models in data-driven time series tasks.

cs.LG↗

arXiv2024

Real-time object detection and robotic manipulation for agriculture using a YOLO-based learning approach

The optimisation of crop harvesting processes for commonly cultivated crops is of great importance in the aim of agricultural industrialisation. Nowadays, the utilisation of machine vision has enabled the automated identification of crops, leading to the enhancement of harvesting efficiency, but challenges still exist. This study presents a new framework that combines two separate architectures of convolutional neural networks (CNNs) in order to simultaneously accomplish the tasks of crop detection and harvesting (robotic manipulation) inside a simulated environment. Crop images in the simulated environment are subjected to random rotations, cropping, brightness, and contrast adjustments to create augmented images for dataset generation. The you only look once algorithmic framework is employed with traditional rectangular bounding boxes for crop localization. The proposed method subsequently utilises the acquired image data via a visual geometry group model in order to reveal the grasping positions for the robotic manipulation.

cs.CV↗

arXiv2024

Discrete-Time Stress Matrix-Based Formation Control of General Linear Multi-Agent Systems

This paper considers the distributed leader-follower stress-matrix-based affine formation control problem of discrete-time linear multi-agent systems with static and dynamic leaders. In leader-follower multi-agent formation control, the aim is to drive a set of agents comprising leaders and followers to form any desired geometric pattern and simultaneously execute any required manoeuvre by controlling only a few agents denoted as leaders. Existing works in literature are mostly limited to the cases where the agents' inter-agent communications are either in the continuous-time settings or the sampled-data cases where the leaders are constrained to constant (or zero) velocities or accelerations. Here, we relax these constraints and study the discrete-time cases where the leaders can have stationary or time-varying velocities. We propose control laws in the study of different situations and provide some sufficient conditions to guarantee the overall system stability. Simulation study is used to demonstrate the efficacy of our proposed control laws.

cs.RO↗