Searcharxiv⌕ Search

arXiv subjects

Minghui Zheng

Publications and source records attributed to Minghui Zheng.

At least 19 recordsLinked to original sources
arXiv2026

ContactWorld: What Representations Matter in Vision-Tactile World Models for Contact-Rich Manipulation

Contact-rich manipulation requires world models to capture complex interaction dynamics from heterogeneous visual and tactile observations, yet the representation properties that enable reliable predictive planning remain poorly understood. We present ContactWorld, a systematic study of vision-tactile representations across 12 contact-rich manipulation tasks. Through controlled evaluation within a unified world-model and planning framework, we find that representations preserving spatial structure and temporal continuity consistently support more accurate prediction and stronger planning performance. Point-cloud observations increase average success from 20.7% and 22.0% with wrist- and front-view RGB, respectively, to 32.1%. Tactile sensing provides further gains only when its representation is compatible with the visual modality, with point clouds and tactile force fields achieving the highest overall success rate of 36.1%. These advantages become more pronounced at increasing goal offsets, where prediction errors and contact uncertainty accumulate. Controlled representation studies and real-world experiments across four manipulation tasks further support these trends. Together, our results establish spatial structure, temporal continuity, and cross-modal compatibility as key principles for designing vision-tactile world models for contact-rich robotic manipulation.

cs.RO↗
arXiv2026

SIMPNet: Spatial-Informed Motion Planning Network

Current robotic manipulators require fast and efficient motion-planning algorithms to operate in cluttered environments. State-of-the-art sampling-based motion planners struggle to scale to high-dimensional configuration spaces and are inefficient in complex environments. This inefficiency arises because these planners utilize either uniform or hand-crafted sampling heuristics within the configuration space. To address these challenges, we present the Spatial-informed Motion Planning Network (SIMPNet). SIMPNet consists of a stochastic graph neural network (GNN)-based sampling heuristic for informed sampling within the configuration space. The sampling heuristic of SIMPNet encodes the workspace embedding into the configuration space through a cross-attention mechanism. It encodes the manipulator's kinematic structure into a graph, which is used to generate informed samples within the framework of sampling-based motion planning algorithms. We have evaluated the performance of SIMPNet using a UR5e robotic manipulator operating within simple and complex workspaces, comparing it against baseline state-of-the-art motion planners. The evaluation results show the effectiveness and advantages of the proposed planner compared to the baseline planners. Project website: \href{https://davoodsz.github.io/simpnet/}{https://davoodsz.github.io/simpnet/}

cs.RO↗
arXiv2026

PHR-VLA: Planning Horizon Reasoning for Vision-Language-Action Models

Vision-language-action models (VLAs) have shown strong promise for general-purpose robotic manipulation by mapping language instructions and vision observations directly to actions. However, most VLAs primarily condition action prediction on current observations and lack an explicit mechanism for reasoning over future task dynamics, which is particularly important for fine-grained, contact-rich manipulation. We present PHR-VLA, a framework that enables planning-horizon reasoning in VLAs through privileged latent representations of future dynamics. PHR-VLA introduces a lightweight auxiliary future head that, during training, aligns the VLA's internal representations with latent dynamics extracted from future observations. Evaluation results demonstrate that local, contact-centric, patch-level latent dynamics supervision from the wrist camera improves success rate on LIBERO from 84.1% to 88.4% and on real-world disassembly tasks from 63.3% to 82.5%. Patch-level supervision from a third-person camera also improves performance on Meta-World from 56.70% to 57.8%. These results demonstrate that privileged latent dynamics alignment provides an effective training signal for improving anticipatory reasoning in VLA policies. Project website: \href{https://davoodsz.github.io/PHR-VLA.github.io/}{https://davoodsz.github.io/PHR-VLA.github.io/}

cs.RO↗
arXiv2026

NeurRAFT: Robot Motion Planning via Anchor-Level Flow Matching with Clearance-Aware Preference Tuning

Recent end-to-end neural motion planners generate trajectories from raw sensor observations, avoiding the privileged geometric models required by classical planners. However, collision-free planning in cluttered environments remains challenging. We present NeurRAFT, a generative planning framework based on anchor-level flow matching and clearance-aware preference tuning. Unlike prior neural planners that model dense waypoint sequences and spend capacity on redundant local details and smoothness, NeurRAFT operates on compact anchor waypoints. We train the planner using a Jacobian-weighted loss that accounts for the task-space impact of each anchor. At inference, the anchors are generated in two integration steps, followed by cubic-spline interpolation to recover a smooth, full-resolution trajectory. Since imitation learning from positive demonstrations cannot distinguish collision-free from near-collision trajectories, collision-prone behaviors persist at test time. Rather than relying on post-hoc corrections, we directly reshape the pretrained planner's distribution toward safer solutions without augmenting inference. Specifically, Direct Preference Optimization shifts probability mass toward trajectories with larger obstacle clearance, with the resulting improvement directly absorbed into the planner parameters. Experiments show substantial improvements over state-of-the-art planners, while real-world experiments demonstrate zero-shot transfer to a Franka robot under noisy and partially occluded depth observations. Video results available at https://neurraft.github.io/.

cs.RO↗
arXiv2026

PerFACT: Motion Policy with LLM-Powered Dataset Synthesis and Fusion Action-Chunking Transformers

Deep learning methods have significantly enhanced motion planning for robotic manipulators by leveraging prior experiences within planning datasets. However, state-of-the-art neural motion planners are primarily trained on small datasets collected in manually generated workspaces, limiting their deployment in various everyday scenarios. Additionally, these planners often rely on monolithic network architectures that struggle to encode critical planning information. To address these challenges, we introduce Motion Policy with Dataset Synthesis powered by large language models (LLMs) and Fusion Action-Chunking Transformers (PerFACT), which incorporates two key components. Firstly, a novel workspace generation method, PerFACT, enables large-scale planning data collection by leveraging procedural primitive generation, and LLM-powered primitive suggestion and placement. Secondly, we introduce Fusion Motion Policy Networks (M$π$NetsFusion), an end-to-end, open-loop neural motion planner that uses a fusion action-chunking transformer to better encode planning signals and attend to multiple feature modalities. Leveraging PerFACT, we collect a dataset of 3.5M trajectories to train and evaluate M$π$NetsFusion against state-of-the-art planners. Results show that M$π$NetsFusion achieves consistently low planning time with sub-second inference, while maintaining competitive performance compared to both sampling-based and end-to-end neural benchmark planners. Project website: \href{https://davoodsz.github.io/perfact.github.io/}{https://davoodsz.github.io/perfact.github.io/}

cs.RO↗
arXiv2026

Flow Motion Policy: Manipulator Motion Planning with Flow Matching Models

Open-loop end-to-end neural motion planners have recently been proposed to improve motion planning for robotic manipulators. These methods enable planning directly from sensor observations without relying on a privileged collision checker during motion planning. However, existing planners produce a single path for a given planning problem and cannot exploit their open-loop nature to propose multiple motion plans. To address this limitation, we introduce Flow Motion Policy, an open-loop neural motion planner that uses flow matching to generate a batch of motion plan proposals by learning a distribution over motion plans conditioned on the planning observation. At inference time, it samples multiple candidate motion plans to enable efficient best-of-$N$ inference while avoiding iterative collision checking during planning. We benchmark the Flow Motion Policy against representative sampling-based, optimization-based and neural motion planning methods. Evaluation results demonstrate that Flow Motion Policy improves planning success and efficiency, highlighting the effectiveness of stochastic generative policies for end-to-end motion planning and best-of-$N$ sampling. Project website: \href{https://davoodsz.github.io/FlowMotionPolicy.github.io/}{https://davoodsz.github.io/FlowMotionPolicy.github.io/}

cs.RO↗
arXiv2026

Learning When to See and When to Feel: Adaptive Vision-Torque Fusion for Contact-Aware Manipulation

Vision-based policies have achieved a good performance in robotic manipulation due to the accessibility and richness of visual observations. However, purely visual sensing becomes insufficient in contact-rich and force-sensitive tasks where force/torque (F/T) signals provide critical information about contact dynamics, alignment, and interaction quality. Although various strategies have been proposed to integrate vision and F/T signals, including auxiliary prediction objectives, mixture-of-experts architectures, and contact-aware gating mechanisms, a comparison of these approaches remains lacking. In this work, we provide a controlled comparison of different F/T-vision integration strategies within diffusion-based manipulation policies. In addition, we propose an adaptive integration strategy that ignores F/T signals during non-contact phases while adaptively leveraging both vision and torque information during contact. Experimental results demonstrate that our method outperforms the strongest baseline by 14% in success rate, highlighting the importance of contact-aware multimodal fusion for robotic manipulation.

cs.RO↗
arXiv2026

Transforming Remanufacturing Automation with Large Language Models: A Forward-Looking Analysis with Case Studies

With growing concerns about resource scarcity and environmental degradation, remanufacturing of end-of-life (EoL) products within the circular economy is attracting increasing attention. Remanufacturing can preserve most of the original manufacturing value and materials while transforming EoL products into like-new condition. However, the variability and uncertainty of EoL products make remanufacturing highly dependent on human expertise. Recently, large language models (LLMs) have demonstrated remarkable capabilities in learning from massive, unstructured datasets, generating expert-level output across various tasks, and communicating with humans in natural language for interpretation. These advantages can align closely with the complex demands of remanufacturing, thereby mitigating the reliance on specialized expertise. However, their roles and research progress in this domain remain underexplored. In this paper, we present a forward-looking review and analysis of the role of LLMs in remanufacturing automation, grounded in a brief critical review of existing LLM-related studies relevant to remanufacturing. Building on this foundation, we introduce ReManGPT as a conceptual framework and use three representative case studies to illustrate selected modules of the framework in practical remanufacturing scenarios. We also analyze three representative remanufacturing applications, electric vehicle batteries, electronic waste, and electric motors, to illustrate how the proposed framework could address their domain-specific challenges. Finally, we discuss the current barriers to deploying this framework in practice and outline future research directions, including LLM-assisted human operation and language-action models for robotic automation.

eess.SY↗
arXiv2026

CONTACT: CONtact-aware TACTile Learning for Robotic Disassembly

Robotic disassembly involves contact-rich interactions in which successful manipulation depends not only on geometric alignment but also on force-dependent state transitions. While vision-based policies perform well in structured settings, their reliability often degrades in tight-tolerance, contact-dominated, or deformable scenarios. In this work, we systematically investigate the role of tactile sensing in robotic disassembly through both simulation and real-world experiments. We construct five rigid-body disassembly tasks in simulation with increasing geometric constraints and extraction difficulty. We further design five real-world tasks, including three rigid and two deformable scenarios, to evaluate contact-dependent manipulation. Within a unified learning framework, we compare three sensing configurations: Vision Only, Vision + tactile RGB (TacRGB), and Vision + tactile force field (TacFF). Across both simulation and real-world experiments, TacFF-based policies consistently achieve the highest success rates, with particularly notable gains in contact-dependent and deformable settings. Notably, naive fusion of TacRGB and TacFF underperforms either modality alone, indicating that simple concatenation can dilute task-relevant force information. Our results show that tactile sensing plays a critical, task-dependent role in robotic disassembly, with structured force-field representations being particularly effective in contact-dominated scenarios.

cs.RO↗
arXiv2026

Imagining the Sense of Touch: Touch-Informed Manipulation via Imagined Tactile Representations

Tactile sensing can substantially improve contact-rich robotic manipulation, yet its practical deployment remains limited by the fragility, calibration requirements, and maintenance burden of tactile hardware. This raises a fundamental question: can robots benefit from tactile knowledge without requiring tactile sensors at deployment? We present TacImag, a tactile imagination framework that predicts tactile observations from vision and proprioception and uses the generated signals to guide manipulation policies. Trained from paired visuotactile demonstrations, TacImag enables touch-informed manipulation using only visual observations at test time. We evaluate TacImag on six simulated and four real-world manipulation tasks. Across simulation and real-world experiments, imagined tactile observations consistently improve manipulation performance without requiring tactile hardware. In real-world experiments, imagined force fields improve contact-sensitive tasks by 44.4% on average, whereas imagined tactile images improve texture-sensitive tasks by 23.3%, revealing that the effectiveness of tactile imagination depends strongly on the relationship between tactile representation and task requirements. Our results further suggest that tactile imagination does not simply recover missing tactile measurements. Instead, it acts as a form of contact-aware supervision that transforms subtle visual interaction cues into representations that are easier for manipulation policies to exploit.

cs.RO↗
arXiv2026

Behavior Uncloning: Distilling Mode Redirection into Policy Weights without Inference-Time Steering

Behavior-cloned policies often learn multiple behavior modes from demonstration datasets, including modes that are unsafe or otherwise undesired at deployment. For example, a policy trained on diverse handover demonstrations may learn to pass a knife blade-first. Standard remedies such as data curation and inference-time steering either require access to the original demonstrations for full retraining or add substantial inference-time overhead. To address this gap, we propose MoRE(Mode Redirection), which redirects policy rollouts toward desired behavior modes through a short "uncloning" step. Specifically, MoRE distills the redirection signal from a temporary mode classifier into the policy weights to steer behavior. A retain loss balances this edit by preserving desired-mode competence, allowing the standalone policy to suppress unwanted modes with zero inference-time overhead. Across eight simulated and real-world tasks, MoRE improves the average deployment success rate (SR) by 44 percentage points over the original mixed-mode policy. Among all compared adaptation and steering baselines, MoRE achieves the strongest SR and approaches the filtered-data retraining reference, while preserving task competence and inference speed. MoRE also generalizes across robot policy backbones, including Diffusion Policy and the Pi0.5 VLA, diverse task categories, and real-world deployments.

cs.RO↗
arXiv2026

HMPO: Hybrid Median-length Policy Optimization for Chain-of-Thought Compression

Large language models achieve remarkable performance via extended chain-of-thought (CoT) reasoning, yet this lengthy process incurs substantial inference overhead. Existing CoT compression methods struggle with inflexible manual length budgets, computationally expensive multi-stage training pipelines, and fragile scalability restricted to small models. We propose HMPO (Hybrid Median-length Policy Optimization), a cost-effective, single-stage reinforcement learning framework. HMPO efficiently compresses CoT via three synergistic components: an adaptive median-based budget derived from successful rollouts to eliminate manual tuning, a cosine-decay token reward for smooth length penalization, and a multiplicative reward formulation that substantially mitigates trivial reward hacking by strictly prioritizing answer correctness. Trained exclusively on mathematical data, HMPO generalizes seamlessly across math, code, science, and instruction-following tasks. Extensive experiments scaling from 9B to 122B parameters across dense and Mixture-of-Experts (MoE) architectures demonstrate that HMPO achieves 19%--46% token compression with negligible accuracy degradation, all while drastically reducing training costs compared to existing multi-stage baselines.

cs.LG↗
arXiv2026

Constrained MPC-Based Motion Planning for Morphing Quadrotors in Ultra-Narrow Passages under Limited Perception

This paper introduces a motion planning framework to plan morphology and trajectory for morphing quadrotors under extremely constrained environments. We develop a novel obstacle avoidance cost function for nonlinear model predictive control (MPC) that enables navigation through extremely narrow gaps under limited perception from a 2D LiDAR. Classical artificial potential field-based costs typically have a high cost in narrow passages, artificially blocking the navigable path. In contrast, we propose a smooth exponential obstacle cost that preserves low traversal cost within narrow gaps while maintaining strong collision avoidance behavior. The formulation avoids hard activation thresholds and introduces a cost reduction factor to reduce the cost within narrow passages. Direct use of 2D LiDAR measurements in MPC allows navigation around arbitrarily shaped obstacles. The method is embedded within an acados-based nonlinear MPC framework. Simulation and experimental results demonstrate successful traversal of narrow corridors where typical repulsive cost functions would fail. The approach provides a computationally efficient and practical solution for navigating through tight spaces while maintaining safety from the obstacles. While we are implementing the framework on the morphing quadrotors, the cost function formulation is general-purpose for any mobile robot application, and is not limited to the morphing quadrotors. The implementation code is available at \href{https://github.com/harshjmodi1996/morphocopter_mpc}{Github Repo} and a short video is available at \href{https://zh.engr.tamu.edu/wp-content/uploads/sites/310/2026/03/MPC_MorphoCopter_video.mp4}{Video Link}.

cs.RO↗
arXiv2026

Redefining End-of-Life: Intelligent Automation for Electronics Remanufacturing Systems

Remanufacturing is fundamentally more challenging than traditional manufacturing due to the significant uncertainty, variability, and incompleteness inherent in end-of-life (EoL) products. At the same time, it has become increasingly essential and urgent for facilitating a circular economy, driven by the growing volume of discarded electronic products and the escalating scarcity of critical materials. In this paper, we review the existing literature and examine the key challenges as well as emerging opportunities in intelligent automation for EoL electronics remanufacturing, providing a comprehensive overview of how robotics, control, and artificial intelligence (AI) can jointly enable scalable, safe, and intelligent remanufacturing systems. This paper starts with the definition, scope, and motivation of remanufacturing within the context of a circular economy, highlighting its societal and environmental significance. Then it delves into intelligent automation approaches for disassembly, inspection, sorting, and component reprocessing in this domain, covering advanced methods for multimodal perception, decision-making under uncertainty, flexible planning algorithms, and force-aware manipulation. The paper further reviews several emerging techniques, including large foundation models, human-in-the-loop integration, and digital twins that have the potential to support future research in this area. By integrating these topics, we aim to illustrate how next-generation remanufacturing systems can achieve robust, adaptable, and efficient operation in the face of complex real-world challenges.

eess.SY↗
arXiv2026

Evaluating Large and Lightweight Vision Models for Irregular Component Segmentation in E-Waste Disassembly

Precise segmentation of irregular and densely arranged components is essential for robotic disassembly and material recovery in electronic waste (e-waste) recycling. This study evaluates the impact of model architecture and scale on segmentation performance by comparing SAM2, a transformer-based vision model, with the lightweight YOLOv8 network. Both models were trained and tested on a newly collected dataset of 1,456 annotated RGB images of laptop components including logic boards, heat sinks, and fans, captured under varying illumination and orientation conditions. Data augmentation techniques, such as random rotation, flipping, and cropping, were applied to improve model robustness. YOLOv8 achieved higher segmentation accuracy (mAP50 = 98.8%, mAP50-95 = 85%) and stronger boundary precision than SAM2 (mAP50 = 8.4%). SAM2 demonstrated flexibility in representing diverse object structures but often produced overlapping masks and inconsistent contours. These findings show that large pre-trained models require task-specific optimization for industrial applications. The resulting dataset and benchmarking framework provide a foundation for developing scalable vision algorithms for robotic e-waste disassembly and circular manufacturing systems.

cs.CV↗
arXiv2026

Toward Generalist Neural Motion Planners for Robotic Manipulators: Challenges and Opportunities

State-of-the-art generalist manipulation policies have enabled the deployment of robotic manipulators in unstructured human environments. However, these frameworks struggle in cluttered environments primarily because they utilize auxiliary modules for low-level motion planning and control. Motion planning remains challenging due to the high dimensionality of the robot's configuration space and the presence of workspace obstacles. Neural motion planners have enhanced motion planning efficiency by offering fast inference and effectively handling the inherent multi-modality of the motion planning problem. Despite such benefits, current neural motion planners often struggle to generalize to unseen, out-of-distribution planning settings. This paper reviews and analyzes the state-of-the-art neural motion planners, highlighting both their benefits and limitations. It also outlines a path toward establishing generalist neural motion planners capable of handling domain-specific challenges. For a list of the reviewed papers, please refer to https://davoodsz.github.io/planning-manip-survey.github.io/.

cs.RO↗
arXiv2026

Learning Actionable Manipulation Recovery via Counterfactual Failure Synthesis

While recent foundation models have significantly advanced robotic manipulation, these systems still struggle to autonomously recover from execution errors. Current failure-learning paradigms rely on either costly and unsafe real-world data collection or simulator-based perturbations, which introduce a severe sim-to-real gap. Furthermore, existing visual analyzers predominantly output coarse, binary diagnoses rather than the executable, trajectory-level corrections required for actual recovery. To bridge the gap between failure diagnosis and actionable recovery, we introduce Dream2Fix, a framework that synthesizes photorealistic, counterfactual failure rollouts directly from successful real-world demonstrations. By perturbing actions within a generative world model, Dream2Fix creates paired failure-correction data without relying on simulators. To ensure the generated data is physically viable for robot learning, we implement a structured verification mechanism that strictly filters rollouts for task validity, visual coherence, and kinematic safety. This engine produces a high-fidelity dataset of over 120k paired samples. Using this dataset, we fine-tune a vision-language model to jointly predict failure types and precise recovery trajectories, mapping visual anomalies directly to corrective actions. Extensive real-world robotic experiments show our approach achieves state-of-the-art correction accuracy, improving from 19.7% to 81.3% over prior baselines, and successfully enables zero-shot closed-loop failure recovery in physical deployments.

cs.RO↗
arXiv2026

Egocentric World Model for Photorealistic Hand-Object Interaction Synthesis

To serve as a scalable data source for embodied AI, world models should act as true simulators that infer interaction dynamics strictly from user actions, rather than mere conditional video generators relying on privileged future object states. In this context, egocentric Human-Object Interaction (HOI) world models are critical for predicting physically grounded first-person rollouts. However, building such models is profoundly challenging due to rapid head motions, severe occlusions, and high-DoF hand articulations that abruptly alter contact topologies. Consequently, existing approaches often circumvent these physics challenges by resorting to conditional video generation with access to known future object trajectories. We introduce EgoHOI, an egocentric HOI world model that breaks away from this shortcut to simulate photorealistic, contact-consistent interactions from action signals alone. To ensure physical accuracy without future-state inputs, EgoHOI distills geometric and kinematic priors from 3D estimates into physics-informed embeddings. These embeddings regularize the egocentric rollouts toward physically valid dynamics. Experiments on the HOT3D dataset demonstrate consistent gains over strong baselines, and ablations validate the effectiveness of our physics-informed design.

cs.CV↗