arXiv subjects

Hao Dong

Publications and source records attributed to Hao Dong.

At least 127 records · Page 7Linked to original sources

arXiv2025

GraspGF: Learning Score-based Grasping Primitive for Human-assisting Dexterous Grasping

The use of anthropomorphic robotic hands for assisting individuals in situations where human hands may be unavailable or unsuitable has gained significant importance. In this paper, we propose a novel task called human-assisting dexterous grasping that aims to train a policy for controlling a robotic hand's fingers to assist users in grasping objects. Unlike conventional dexterous grasping, this task presents a more complex challenge as the policy needs to adapt to diverse user intentions, in addition to the object's geometry. We address this challenge by proposing an approach consisting of two sub-modules: a hand-object-conditional grasping primitive called Grasping Gradient Field~(GraspGF), and a history-conditional residual policy. GraspGF learns `how' to grasp by estimating the gradient from a success grasping example set, while the residual policy determines `when' and at what speed the grasping action should be executed based on the trajectory history. Experimental results demonstrate the superiority of our proposed method compared to baselines, highlighting the user-awareness and practicality in real-world applications. The codes and demonstrations can be viewed at "https://sites.google.com/view/graspgf".

cs.RO↗

arXiv2025

Boosting Universal LLM Reward Design through Heuristic Reward Observation Space Evolution

Large Language Models (LLMs) are emerging as promising tools for automated reinforcement learning (RL) reward design, owing to their robust capabilities in commonsense reasoning and code generation. By engaging in dialogues with RL agents, LLMs construct a Reward Observation Space (ROS) by selecting relevant environment states and defining their internal operations. However, existing frameworks have not effectively leveraged historical exploration data or manual task descriptions to iteratively evolve this space. In this paper, we propose a novel heuristic framework that enhances LLM-driven reward design by evolving the ROS through a table-based exploration caching mechanism and a text-code reconciliation strategy. Our framework introduces a state execution table, which tracks the historical usage and success rates of environment states, overcoming the Markovian constraint typically found in LLM dialogues and facilitating more effective exploration. Furthermore, we reconcile user-provided task descriptions with expert-defined success criteria using structured prompts, ensuring alignment in reward design objectives. Comprehensive evaluations on benchmark RL tasks demonstrate the effectiveness and stability of the proposed framework. Code and video demos are available at jingjjjjjie.github.io/LLM2Reward.

cs.AI↗

arXiv2025

PartRM: Modeling Part-Level Dynamics with Large Cross-State Reconstruction Model

As interest grows in world models that predict future states from current observations and actions, accurately modeling part-level dynamics has become increasingly relevant for various applications. Existing approaches, such as Puppet-Master, rely on fine-tuning large-scale pre-trained video diffusion models, which are impractical for real-world use due to the limitations of 2D video representation and slow processing times. To overcome these challenges, we present PartRM, a novel 4D reconstruction framework that simultaneously models appearance, geometry, and part-level motion from multi-view images of a static object. PartRM builds upon large 3D Gaussian reconstruction models, leveraging their extensive knowledge of appearance and geometry in static objects. To address data scarcity in 4D, we introduce the PartDrag-4D dataset, providing multi-view observations of part-level dynamics across over 20,000 states. We enhance the model's understanding of interaction conditions with a multi-scale drag embedding module that captures dynamics at varying granularities. To prevent catastrophic forgetting during fine-tuning, we implement a two-stage training process that focuses sequentially on motion and appearance learning. Experimental results show that PartRM establishes a new state-of-the-art in part-level motion learning and can be applied in manipulation tasks in robotics. Our code, data, and models are publicly available to facilitate future research.

cs.CV↗

arXiv2025

SpatialBot: Precise Spatial Understanding with Vision Language Models

Vision Language Models (VLMs) have achieved impressive performance in 2D image understanding, however they are still struggling with spatial understanding which is the foundation of Embodied AI. In this paper, we propose SpatialBot for better spatial understanding by feeding both RGB and depth images. Additionally, we have constructed the SpatialQA dataset, which involves multi-level depth-related questions to train VLMs for depth understanding. Finally, we present SpatialBench to comprehensively evaluate VLMs' capabilities in spatial understanding at different levels. Extensive experiments on our spatial-understanding benchmark, general VLM benchmarks and Embodied AI tasks, demonstrate the remarkable improvements of SpatialBot trained on SpatialQA. The model, code and data are available at https://github.com/BAAI-DCAI/SpatialBot.

cs.CV↗

arXiv2025

Foundation Feature-Driven Online End-Effector Pose Estimation: A Marker-Free and Learning-Free Approach

Accurate transformation estimation between camera space and robot space is essential. Traditional methods using markers for hand-eye calibration require offline image collection, limiting their suitability for online self-calibration. Recent learning-based robot pose estimation methods, while advancing online calibration, struggle with cross-robot generalization and require the robot to be fully visible. This work proposes a Foundation feature-driven online End-Effector Pose Estimation (FEEPE) algorithm, characterized by its training-free and cross end-effector generalization capabilities. Inspired by the zero-shot generalization capabilities of foundation models, FEEPE leverages pre-trained visual features to estimate 2D-3D correspondences derived from the CAD model and target image, enabling 6D pose estimation via the PnP algorithm. To resolve ambiguities from partial observations and symmetry, a multi-historical key frame enhanced pose optimization algorithm is introduced, utilizing temporal information for improved accuracy. Compared to traditional hand-eye calibration, FEEPE enables marker-free online calibration. Unlike robot pose estimation, it generalizes across robots and end-effectors in a training-free manner. Extensive experiments demonstrate its superior flexibility, generalization, and performance.

cs.RO↗

arXiv2025

TransDiff: Diffusion-Based Method for Manipulating Transparent Objects Using a Single RGB-D Image

Manipulating transparent objects presents significant challenges due to the complexities introduced by their reflection and refraction properties, which considerably hinder the accurate estimation of their 3D shapes. To address these challenges, we propose a single-view RGB-D-based depth completion framework, TransDiff, that leverages the Denoising Diffusion Probabilistic Models(DDPM) to achieve material-agnostic object grasping in desktop. Specifically, we leverage features extracted from RGB images, including semantic segmentation, edge maps, and normal maps, to condition the depth map generation process. Our method learns an iterative denoising process that transforms a random depth distribution into a depth map, guided by initially refined depth information, ensuring more accurate depth estimation in scenarios involving transparent objects. Additionally, we propose a novel training method to better align the noisy depth and RGB image features, which are used as conditions to refine depth estimation step by step. Finally, we utilized an improved inference process to accelerate the denoising procedure. Through comprehensive experimental validation, we demonstrate that our method significantly outperforms the baselines in both synthetic and real-world benchmarks with acceptable inference time. The demo of our method can be found on https://wang-haoxiao.github.io/TransDiff/

cs.CV↗

arXiv2025

GarmentPile: Point-Level Visual Affordance Guided Retrieval and Adaptation for Cluttered Garments Manipulation

Cluttered garments manipulation poses significant challenges due to the complex, deformable nature of garments and intricate garment relations. Unlike single-garment manipulation, cluttered scenarios require managing complex garment entanglements and interactions, while maintaining garment cleanliness and manipulation stability. To address these demands, we propose to learn point-level affordance, the dense representation modeling the complex space and multi-modal manipulation candidates, while being aware of garment geometry, structure, and inter-object relations. Additionally, as it is difficult to directly retrieve a garment in some extremely entangled clutters, we introduce an adaptation module, guided by learned affordance, to reorganize highly-entangled garments into states plausible for manipulation. Our framework demonstrates effectiveness over environments featuring diverse garment types and pile configurations in both simulation and the real world. Project page: https://garmentpile.github.io/.

cs.RO↗

arXiv2025

ET-SEED: Efficient Trajectory-Level SE(3) Equivariant Diffusion Policy

Imitation learning, e.g., diffusion policy, has been proven effective in various robotic manipulation tasks. However, extensive demonstrations are required for policy robustness and generalization. To reduce the demonstration reliance, we leverage spatial symmetry and propose ET-SEED, an efficient trajectory-level SE(3) equivariant diffusion model for generating action sequences in complex robot manipulation tasks. Further, previous equivariant diffusion models require the per-step equivariance in the Markov process, making it difficult to learn policy under such strong constraints. We theoretically extend equivariant Markov kernels and simplify the condition of equivariant diffusion process, thereby significantly improving training efficiency for trajectory-level SE(3) equivariant diffusion policy in an end-to-end manner. We evaluate ET-SEED on representative robotic manipulation tasks, involving rigid body, articulated and deformable object. Experiments demonstrate superior data efficiency and manipulation proficiency of our proposed method, as well as its ability to generalize to unseen configurations with only a few demonstrations. Website: https://et-seed.github.io/

cs.RO↗

arXiv2025

AdaManip: Adaptive Articulated Object Manipulation Environments and Policy Learning

Articulated object manipulation is a critical capability for robots to perform various tasks in real-world scenarios. Composed of multiple parts connected by joints, articulated objects are endowed with diverse functional mechanisms through complex relative motions. For example, a safe consists of a door, a handle, and a lock, where the door can only be opened when the latch is unlocked. The internal structure, such as the state of a lock or joint angle constraints, cannot be directly observed from visual observation. Consequently, successful manipulation of these objects requires adaptive adjustment based on trial and error rather than a one-time visual inference. However, previous datasets and simulation environments for articulated objects have primarily focused on simple manipulation mechanisms where the complete manipulation process can be inferred from the object's appearance. To enhance the diversity and complexity of adaptive manipulation mechanisms, we build a novel articulated object manipulation environment and equip it with 9 categories of objects. Based on the environment and objects, we further propose an adaptive demonstration collection and 3D visual diffusion-based imitation learning pipeline that learns the adaptive manipulation policy. The effectiveness of our designs and proposed method is validated through both simulation and real-world experiments. Our project page is available at: https://adamanip.github.io

cs.RO↗

arXiv2025

Interactive Symbolic Regression through Offline Reinforcement Learning: A Co-Design Framework

Symbolic Regression (SR) holds great potential for uncovering underlying mathematical and physical relationships from observed data. However, the vast combinatorial space of possible expressions poses significant challenges for both online search methods and pre-trained transformer models. Additionally, current state-of-the-art approaches typically do not consider the integration of domain experts' prior knowledge and do not support iterative interactions with the model during the equation discovery process. To address these challenges, we propose the Symbolic Q-network (Sym-Q), an advanced interactive framework for large-scale symbolic regression. Unlike previous large-scale transformer-based SR approaches, Sym-Q leverages reinforcement learning without relying on a transformer-based decoder. This formulation allows the agent to learn through offline reinforcement learning using any type of tree encoder, enabling more efficient training and inference. Furthermore, we propose a co-design mechanism, where the reinforcement learning-based Sym-Q facilitates effective interaction with domain experts at any stage of the equation discovery process. Users can dynamically modify generated nodes of the expression, collaborating with the agent to tailor the mathematical expression to best fit the problem and align with the assumed physical laws, particularly when there is prior partial knowledge of the expected behavior. Our experiments demonstrate that the pre-trained Sym-Q surpasses existing SR algorithms on the challenging SSDNC benchmark. Moreover, we experimentally show on real-world cases that its performance can be further enhanced by the interactive co-design mechanism, with Sym-Q achieving greater performance gains than other state-of-the-art models. Our reproducible code is available at https://github.com/EPFL-IMOS/Sym-Q.

cs.LG↗

arXiv2025

Interactive Symbolic Regression through Offline Reinforcement Learning: A Co-Design Framework

cs.LG↗

arXiv2025

An Automatic Sound and Complete Abstraction Method for Generalized Planning with Baggable Types

Generalized planning is concerned with how to find a single plan to solve multiple similar planning instances. Abstractions are widely used for solving generalized planning, and QNP (qualitative numeric planning) is a popular abstract model. Recently, Cui et al. showed that a plan solves a sound and complete abstraction of a generalized planning problem if and only if the refined plan solves the original problem. However, existing work on automatic abstraction for generalized planning can hardly guarantee soundness let alone completeness. In this paper, we propose an automatic sound and complete abstraction method for generalized planning with baggable types. We use a variant of QNP, called bounded QNP (BQNP), where integer variables are increased or decreased by only one. Since BQNP is undecidable, we propose and implement a sound but incomplete solver for BQNP. We present an automatic method to abstract a BQNP problem from a classical planning instance with baggable types. The basic idea for abstraction is to introduce a counter for each bag of indistinguishable tuples of objects. We define a class of domains called proper baggable domains, and show that for such domains, the BQNP problem got by our automatic method is a sound and complete abstraction for a generalized planning problem whose instances share the same bags with the given instance but the sizes of the bags might be different. Thus, the refined plan of a solution to the BQNP problem is a solution to the generalized planning problem. Finally, we implement our abstraction method and experiments on a number of domains demonstrate the promise of our approach.

cs.AI↗

arXiv2025

Higher-order multiscale method and its convergence analysis for nonlinear thermo-electric coupling problems of composite structures

This paper proposes a higher-order multiscale computational method for nonlinear thermo-electric coupling problems of composite structures, which possess temperature-dependent material properties and nonlinear Joule heating. The innovative contributions of this work are the novel multiscale formulation with the higher-order correction terms for periodic composite structures and the global error estimation with an explicit rate for higher-order multiscale solutions. By employing the multiscale asymptotic approach and the Taylor series technique, the higher-order multiscale method is established for time-dependent nonlinear thermo-electric coupling problems, which can keep the local balance of heat flux and electric charge for high-accuracy multiscale simulation. Furthermore, an efficient numerical algorithm with off-line and on-line stages is presented in detail, and corresponding convergent analysis is also obtained. Two- and three-dimensional numerical experiments are conducted to showcase the competitive advantages of the proposed method for simulating the time-dependent nonlinear thermo-electric coupling problems in composite structures, not only exceptional numerical accuracy, but also less computational cost.

math.NA↗

arXiv2025

Towards Robust Multimodal Open-set Test-time Adaptation via Adaptive Entropy-aware Optimization

Test-time adaptation (TTA) has demonstrated significant potential in addressing distribution shifts between training and testing data. Open-set test-time adaptation (OSTTA) aims to adapt a source pre-trained model online to an unlabeled target domain that contains unknown classes. This task becomes more challenging when multiple modalities are involved. Existing methods have primarily focused on unimodal OSTTA, often filtering out low-confidence samples without addressing the complexities of multimodal data. In this work, we present Adaptive Entropy-aware Optimization (AEO), a novel framework specifically designed to tackle Multimodal Open-set Test-time Adaptation (MM-OSTTA) for the first time. Our analysis shows that the entropy difference between known and unknown samples in the target domain strongly correlates with MM-OSTTA performance. To leverage this, we propose two key components: Unknown-aware Adaptive Entropy Optimization (UAE) and Adaptive Modality Prediction Discrepancy Optimization (AMP). These components enhance the ability of model to distinguish unknown class samples during online adaptation by amplifying the entropy difference between known and unknown samples. To thoroughly evaluate our proposed methods in the MM-OSTTA setting, we establish a new benchmark derived from existing datasets. This benchmark includes two downstream tasks and incorporates five modalities. Extensive experiments across various domain shift situations demonstrate the efficacy and versatility of the AEO framework. Additionally, we highlight the strong performance of AEO in long-term and continual MM-OSTTA settings, both of which are challenging and highly relevant to real-world applications. Our source code is available at https://github.com/donghao51/AEO.

cs.CV↗

arXiv2025

OmniManip: Towards General Robotic Manipulation via Object-Centric Interaction Primitives as Spatial Constraints

The development of general robotic systems capable of manipulating in unstructured environments is a significant challenge. While Vision-Language Models(VLM) excel in high-level commonsense reasoning, they lack the fine-grained 3D spatial understanding required for precise manipulation tasks. Fine-tuning VLM on robotic datasets to create Vision-Language-Action Models(VLA) is a potential solution, but it is hindered by high data collection costs and generalization issues. To address these challenges, we propose a novel object-centric representation that bridges the gap between VLM's high-level reasoning and the low-level precision required for manipulation. Our key insight is that an object's canonical space, defined by its functional affordances, provides a structured and semantically meaningful way to describe interaction primitives, such as points and directions. These primitives act as a bridge, translating VLM's commonsense reasoning into actionable 3D spatial constraints. In this context, we introduce a dual closed-loop, open-vocabulary robotic manipulation system: one loop for high-level planning through primitive resampling, interaction rendering and VLM checking, and another for low-level execution via 6D pose tracking. This design ensures robust, real-time control without requiring VLM fine-tuning. Extensive experiments demonstrate strong zero-shot generalization across diverse robotic manipulation tasks, highlighting the potential of this approach for automating large-scale simulation data generation.

cs.RO↗

arXiv2024

GarmentLab: A Unified Simulation and Benchmark for Garment Manipulation

Manipulating garments and fabrics has long been a critical endeavor in the development of home-assistant robots. However, due to complex dynamics and topological structures, garment manipulations pose significant challenges. Recent successes in reinforcement learning and vision-based methods offer promising avenues for learning garment manipulation. Nevertheless, these approaches are severely constrained by current benchmarks, which offer limited diversity of tasks and unrealistic simulation behavior. Therefore, we present GarmentLab, a content-rich benchmark and realistic simulation designed for deformable object and garment manipulation. Our benchmark encompasses a diverse range of garment types, robotic systems and manipulators. The abundant tasks in the benchmark further explores of the interactions between garments, deformable objects, rigid bodies, fluids, and human body. Moreover, by incorporating multiple simulation methods such as FEM and PBD, along with our proposed sim-to-real algorithms and real-world benchmark, we aim to significantly narrow the sim-to-real gap. We evaluate state-of-the-art vision methods, reinforcement learning, and imitation learning approaches on these tasks, highlighting the challenges faced by current algorithms, notably their limited generalization capabilities. Our proposed open-source environments and comprehensive analysis show promising boost to future research in garment manipulation by unlocking the full potential of these methods. We guarantee that we will open-source our code as soon as possible. You can watch the videos in supplementary files to learn more about the details of our work. Our project page is available at: https://garmentlab.github.io/

cs.RO↗

arXiv2024

Predictive Inverse Dynamics Models are Scalable Learners for Robotic Manipulation

Current efforts to learn scalable policies in robotic manipulation primarily fall into two categories: one focuses on "action," which involves behavior cloning from extensive collections of robotic data, while the other emphasizes "vision," enhancing model generalization by pre-training representations or generative models, also referred to as world models, using large-scale visual datasets. This paper presents an end-to-end paradigm that predicts actions using inverse dynamics models conditioned on the robot's forecasted visual states, named Predictive Inverse Dynamics Models (PIDM). By closing the loop between vision and action, the end-to-end PIDM can be a better scalable action learner. In practice, we use Transformers to process both visual states and actions, naming the model Seer. It is initially pre-trained on large-scale robotic datasets, such as DROID, and can be adapted to realworld scenarios with a little fine-tuning data. Thanks to large-scale, end-to-end training and the synergy between vision and action, Seer significantly outperforms previous methods across both simulation and real-world experiments. It achieves improvements of 13% on the LIBERO-LONG benchmark, 21% on CALVIN ABC-D, and 43% in real-world tasks. Notably, Seer sets a new state-of-the-art on CALVIN ABC-D benchmark, achieving an average length of 4.28, and exhibits superior generalization for novel objects, lighting conditions, and environments under high-intensity disturbances on real-world scenarios. Code and models are publicly available at https://github.com/OpenRobotLab/Seer/.

cs.RO↗

arXiv2024

Role of Wettability, Adhesion, and Instabilities in Transitions During Lubricated Sliding Friction

Lubricated contacts in soft materials are important in various engineering systems and natural settings. Three major lubrication regimes are boundary (BL), mixed (ML), and elasto-hydrodynamic (EHL) lubrication, where the contact region is dry, partially wetted, or fully wetted, respectively. The transition between these regimes is insufficiently understood, especially for soft contacts, which impedes desired control of lubricated sliding friction. Here, we report on the role of solid wettability and adhesion on these transitions. Wettability of glycerol on polydimethylsiloxane (PDMS) surface, and adhesion between a glass indenter and PDMS, were varied by exposure of the PDMS to an ultraviolet light-ozone (UV-Ozone) cleaner. By combining friction tests and visualization, we demonstrate that the transition from ML to BL regime is dominated by the wettability of the lubricant; increasing wettability of glycerol makes removal of liquid from the contact region more difficult. Transition from EHL to ML is related to a series of events with increasing normal load, which are thinning of the lubricant layer, sudden jump to contact between the glass indenter and solid substrate across a gap of tens to a few hundreds of nanometers, and attendant elastic instabilities such as wrinkling and stick-slip. These results provide a deeper understanding of transitions in lubricated frictional behavior of soft materials which govern the maximum and minimum friction achievable.

cond-mat.soft↗