One-shot imitation learning (OSIL) offers a promising way to teach robots new skills without large-scale data collection. However, current OSIL methods are primarily limited to short-horizon tasks, thus limiting their applicability to complex, long-horizon manipulations. To address this limitation, we propose ManiLong-Shot, a novel framework that enables effective OSIL for long-horizon prehensile manipulation tasks. ManiLong-Shot structures long-horizon tasks around physical interaction events, reframing the problem as sequencing interaction-aware primitives instead of directly imitating continuous trajectories. This primitive decomposition can be driven by high-level reasoning from a vision-language model (VLM) or by rule-based heuristics derived from robot state changes. For each primitive, ManiLong-Shot predicts invariant regions critical to the interaction, establishes correspondences between the demonstration and the current observation, and computes the target end-effector pose, enabling effective task execution. Extensive simulation experiments show that ManiLong-Shot, trained on only 10 short-horizon tasks, generalizes to 20 unseen long-horizon tasks across three difficulty levels via one-shot imitation, achieving a 22.8% relative improvement over the SOTA. Additionally, real-robot experiments validate ManiLong-Shot’s ability to robustly execute three long-horizon manipulation tasks via OSIL, confirming its practical applicability.

Robotic dexterous manipulation is a challenging problem due to high degrees of freedom (DoFs) and complex contacts of multi-fingered robotic hands. Many existing deep reinforcement learning (DRL) based methods aim at improving sample efficiency in high-dimensional output action spaces. However, existing works often overlook the role of representations in achieving generalization of a manipulation policy in the complex input space during the hand-object interaction. In this paper, we propose DexRep, a novel hand-object interaction representation to capture object surface features and spatial relations between hands and objects for dexterous manipulation skill learning. Based on DexRep, policies are learned for three dexterous manipulation tasks, i.e. grasping, in-hand reorientation, bimanual handover, and extensive experiments are conducted to verify the effectiveness. In simulation, for grasping, the policy learned with 40 objects achieves a success rate of 87.9% on more than 5000 unseen objects of diverse categories, significantly surpassing existing work trained with thousands of objects; for the in-hand reorientation and handover tasks, the policies also boost the success rates and other metrics of existing hand-object representations by 20% to 40%. The grasp policies with DexRep are deployed to the real world under multi-camera and single-camera setups and demonstrate a small sim-to-real gap.

Heterogeneous multi-robot systems (HMRS) have emerged as a powerful ap- proach for tackling complex tasks that single robots cannot manage alone. Current large-language-model-based multi-agent systems (LLM-based MAS) have shown success in areas like software development and operating systems, but applying these systems to robot control presents unique challenges. In particular, the ca- pabilities of each agent in a multi-robot system are inherently tied to the physical composition of the robots, rather than predefined roles. To address this issue, we introduce a novel multi-agent framework designed to enable effective collab- oration among heterogeneous robots with varying embodiments and capabilities, along with a new benchmark named Habitat-MAS. One of our key designs is Robot Resume: Instead of adopting human-designed role play, we propose a self- prompted approach, where agents comprehend robot URDF files and call robot kinematics tools to generate descriptions of their physics capabilities to guide their behavior in task planning and action execution. The Habitat-MAS bench- mark is designed to assess how a multi-agent framework handles tasks that require embodiment-aware reasoning, which includes 1) manipulation, 2) perception, 3) navigation, and 4) comprehensive multi-floor object rearrangement. The experi- mental results indicate that the robot’s resume and the hierarchical design of our multi-agent system are essential for the effective operation of the heterogeneous multi-robot system within this intricate problem context.

We aim to develop a model-based planning framework for world models that can be scaled with increasing model and data budgets for general-purpose manipulation tasks with only language and vision inputs. To this end, we present FLow-CentrIc generative Planning (FLIP), a model-based planning algorithm on visual space that features three key modules: 1) a multi-modal flow generation model as the general-purpose action proposal module; 2) a flow-conditioned video generation model as the dynamics module; and 3) a vision-language representation learning model as the value module. Given an initial image and language instruction as the goal, FLIP can progressively search for long-horizon flow and video plans that maximize the discounted return to accomplish the task. FLIP is able to synthesize long-horizon plans across objects, robots, and tasks with image flows as the general action representation, and the dense flow information also provides rich guidance for long-horizon video generation. In addition, the synthesized flow and video plans can guide the training of low-level control policies for robot execution. Experiments on diverse benchmarks demonstrate that FLIP can improve both the success rates and quality of long-horizon video plan synthesis and has the interactive world model property, opening up wider applications for future works. Video demos are on our website: https://nus-lins-lab.github.io/flipweb/.

Dexterous grasping is a fundamental yet challenging skill in robotic manipulation, requiring precise interaction between robotic hands and objects. In this paper, we present$\mathcal{D}(\mathcal{R}, \mathcal{O})$Grasp, a novel framework that models the interaction between the robotic hand in its grasping pose and the object, enabling broad generalization across various robot hands and object geometries. Our model takes the robot hand's description and object point cloud as inputs and efficiently predicts kinematically valid and stable grasps, demonstrating strong adaptability to diverse robot embodiments and object geometries. Extensive experiments conducted in both simulated and real-world environments validate the effectiveness of our approach, with significant improvements in success rate, grasp diversity, and inference speed across multiple robotic hands. Our method achieves an average success rate of$87.53\%$in simulation in less than one second, tested across three different dexterous robotic hands. In real-world experiments using the LeapHand, the method also demonstrates an average success rate of$\mathbf{8 9\%}. \mathcal{D}(\mathcal{R}, \mathcal{O})$Grasp provides a robust solution for dexterous grasping in complex and varied environments. The code, appendix, and videos are available on our project website at https://nus-lins-lab.github.io/drograspweb/.

Garment folding is a common yet challenging task in robotic manipulation. The deformability of garments leads to a vast state space and complex dynamics, which complicates precise and fine-grained manipulation. In this paper, we present MetaFold, a unified framework that disentangles task planning from action prediction and learns each independently to enhance model generalization. It employs language-guided point cloud trajectory generation for task planning and a low-level foundation model for action prediction. This structure facilitates multi-category learning, enabling the model to adapt flexibly to various user instructions and folding tasks. We also construct a large-scale MetaFold dataset comprising folding point cloud trajectories for a total of 1210 garments across multiple categories, each paired with corresponding language annotations. Extensive experiments demonstrate the superiority of our proposed framework. Supplementary materials are available on our website: https://meta-fold.github.io/.

The rapid progress of navigation, manipulation, and vision models has made mobile manipulators capable in many specialized tasks. However, the open-world mobile manipulation (OWMM) task remains a challenge due to the need for generalization to open-ended instructions and environments, as well as the systematic complexity to integrate high-level decision making with low-level robot control based on both global scene understanding and current agent state. To address this complexity, we propose a novel multi-modal agent architecture that maintains multi-view scene frames and agent states for decision-making and controls the robot by function calling. A second challenge is the hallucination from domain shift. To enhance the agent performance, we further introduce an agentic data synthesis pipeline for the OWMM task to adapt the VLM model to our task domain with instruction fine-tuning. We highlight our fine-tuned OWMM-VLM as the first dedicated foundation model for mobile manipulators with global scene understanding, robot state tracking, and multi-modal action generation in a unified model. Through experiments, we demonstrate that our model achieves SOTA performance compared to other foundation models including GPT-4o and strong zero-shot generalization in real world. The project page is at https://hhyhrhy.github.io/owmm-agent-project.

Recent studies on Vision-Language-Action (VLA) models have shifted from the end-to-end action-generation paradigm toward a pipeline involving task planning followed by action generation, demonstrating improved performance on various complex, long-horizon manipulation tasks. However, existing approaches vary significantly in terms of network architectures, planning paradigms, representations, and training data sources, making it challenging for researchers to identify the precise sources of performance gains and determine which component is more difficult to learn. To systematically investigate the impacts of different planning paradigms and representations isolating from network architectures and training data, in this paper, we introduce \name, a unified VLA architecture suite capable of various task planning paradigms, and design a comprehensive suite of controlled experiments across diverse object categories (rigid and deformable), visual modalities (2D and 3D), environments (simulation and real-world), and end-effectors (grippers and dexterous hands). Our results demonstrate that: 1) visually grounded planning representations are generally better than language planning representations; 2) the Hierarchical-VLA paradigm generally achieves superior performance than other paradigms, albeit at the cost of slower training and inference speeds.

Shape assembly composes complex shapes geometries by arranging simple part geometries and has wide applications in autonomous robotic assembly and CAD modeling. Existing works focus on geometry reasoning and neglect the actual physical assembly process of matching and fitting joints, which are the contact surfaces connecting different parts. In this paper, we consider contacting joints for the task of multi-part assembly. A successful joint-optimized assembly needs to satisfy the bilateral objectives of shape structure and joint alignment. We propose a hierarchical graph learning approach composed of two levels of graph representation learning. The part graph takes part geometries as input to build the desired shape structure. The joint-level graph uses part joints information and focuses on matching and aligning joints. The two kinds of information are combined to achieve the bilateral objectives. Extensive experiments demonstrate that our method outperforms previous methods, achieving better shape structure and higher joint alignment accuracy.

Learning reward functions remains the bottleneck to equip a robot with a broad repertoire of skills. Large Language Models (LLM) contain valuable task-related knowledge that can potentially aid in the learning of reward functions. However, the proposed reward function can be imprecise, thus ineffective which requires to be further grounded with environment information. We proposed a method to learn rewards more efficiently in the absence of humans. Our approach consists of two components: We first use the LLM to propose features and parameterization of the reward, then update the parameters through an iterative self-alignment process. In particular, the process minimizes the ranking inconsistency between the LLM and the learnt reward functions based on the execution feedback. The method was validated on 9 tasks across 2 simulation environments. It demonstrates a consistent improvement in training efficacy and efficiency, meanwhile consuming significantly fewer GPT tokens compared to the alternative mutation-based method.

We present Jade, a differentiable physics engine for articulated rigid bodies. Jade models contacts as the Linear Complementarity Problem (LCP). Compared to existing differentiable simulations, Jade offers features including intersection-free collision simulation and stable LCP solutions for multiple frictional contacts. We use continuous collision detection to detect the time of impact and adopt the backtracking strategy to prevent intersection between bodies with complex geometry shapes. We derive the gradient calculation to ensure the whole simulation process is differentiable under the backtracking mechanism. We modify the popular Dantzig’s algorithm to get valid solutions under multiple frictional contacts. We conduct extensive experiments to demonstrate the effectiveness of our differentiable physics simulation over a variety of contact-rich tasks. Supplemental materials and videos are available on our project webpage at https://sites.google.com/view/diffsim

Articulated objects like cabinets and doors are widespread in daily life. However, directly manipulating 3D articulated objects is challenging because they have diverse geometrical shapes, semantic categories, and kinetic constraints. Prior works mostly focused on recognizing and manipulating articulated objects with specific joint types. They can either estimate the joint parameters or distinguish suitable grasp poses to facilitate trajectory planning. Although these approaches have succeeded in certain types of articulated objects, they lack generalizability to unseen objects, which significantly impedes their application in broader scenarios. In this paper, we propose a novel framework of Generalizable Articulation Modeling and Manipulating for Articulated Objects (GAMMA), which learns both articulation modeling and grasp pose affordance from diverse articulated objects with different categories. In addition, GAMMA adopts adaptive manipulation to iteratively reduce the modeling errors and enhance manipulation performance. We train GAMMA with the PartNet-Mobility dataset and evaluate with comprehensive experiments in SAPIEN simulation and real-world Franka robot. Results show that GAMMA significantly outperforms SOTA articulation modeling and manipulation algorithms in unseen and cross-category articulated objects. Images, videos and codes are published on the project website at: sites.google.com/view/gamma-articulation.

Differentiable physics simulation provides an avenue to tackle previously intractable challenges through gradient-based optimization, thereby greatly improving the efficiency of solving robotics-related problems. To apply differentiable simulation in diverse robotic manipulation scenarios, a key challenge is to integrate various materials in a unified framework. We present SoftMAC, a differentiable simulation framework that couples soft bodies with articulated rigid bodies and clothes. SoftMAC simulates soft bodies with the continuum-mechanics-based Material Point Method (MPM). We provide a novel forecast-based contact model for MPM, which effectively reduces penetration without introducing other artifacts like unnatural rebound. To couple MPM particles with deformable and non-volumetric clothes meshes, we also propose a penetration tracing algorithm that reconstructs the signed distance field in local area. Diverging from previous works, SoftMAC simulates the complete dynamics of each modality and incorporates them into a cohesive system with an explicit and differentiable coupling mechanism. The feature empowers SoftMAC to handle a broader spectrum of interactions, such as soft bodies serving as manipulators and engaging with underactuated systems. We conducted comprehensive experiments to validate the effectiveness and accuracy of the proposed differentiable pipeline in downstream robotic manipulation applications. Supplementary materials are available on our project website at https://damianliumin.github.io/SoftMAC.

To substantially enhance robot intelligence, there is a pressing need to develop a large model that enables general-purpose robots to proficiently undertake a broad spectrum of manipulation tasks, akin to the versatile task-planning ability exhibited by LLMs. The vast diversity in objects, robots, and manipulation tasks presents huge challenges. Our work introduces a comprehensive framework to develop a foundation model for general robotic manipulation that formalizes a manipulation task as contact synthesis. Specifically, our model takes as input object and robot manipulator point clouds, object physical attributes, target motions, and manipulation region masks. It outputs contact points on the object and associated contact forces or post-contact motions for robots to achieve the desired manipulation task. We perform extensive experiments both in the simulation and real-world settings, manipulating articulated rigid objects, rigid objects, and deformable objects that vary in dimensionality, ranging from one-dimensional objects like ropes to two-dimensional objects like cloth and extending to three-dimensional objects such as plasticine. Our model achieves average success rates of around 90%. Supplementary materials and videos are available on our project website at https://manifoundationmodel.github.io/.

Detecting 3D keypoints with semantic consistency is widely used in many scenarios such as pose estimation, shape registration and robotics. Currently, most unsupervised 3D keypoint detection methods focus on the rigid-body objects. However, when faced with deformable objects, the keypoints they identify do not preserve semantic consistency well. In this paper, we introduce an innovative unsupervised keypoint detector Key-Grid for both the rigid-body and deformable objects, which is an autoencoder framework. The encoder predicts keypoints and the decoder utilizes the generated keypoints to reconstruct the objects. Unlike previous work, we leverage the identified keypoint in formation to form a 3D grid feature heatmap called grid heatmap, which is used in the decoder section. Grid heatmap is a novel concept that represents the latent variables for grid points sampled uniformly in the 3D cubic space, where these variables are the shortest distance between the grid points and the “skeleton” connected by keypoint pairs. Meanwhile, we incorporate the information from each layer of the encoder into the decoder section. We conduct an extensive evaluation of Key-Grid on a list of benchmark datasets. Key-Grid achieves the state-of-the-art performance on the semantic consistency and position accuracy of keypoints. Moreover, we demonstrate the robustness of Key-Grid to noise and downsampling. In addition, we achieve SE-(3) invariance of keypoints though generalizing Key-Grid to a SE(3)-invariant backbone.

Robot in-hand manipulation is an important skill for robots to carry out sophisticated tasks that require moving the grasped object within hand. In this work, we present the Roller Grasper V3, a nonanthropomorphic robot grasper with a steerable roller on each of its four fingertips, and a manipulation architecture that enables the Roller Grasper V3 to achieve full 6-DoF manipulation of the grasped object in$SE(3)$. The manipulation architecture consists of a high-level planner that searches for a feasible path with waypoints for the object to be manipulated, and a low-level control policy that is used to navigate the object in between the adjacent waypoints. The method was experimentally validated on the Roller Grasper V3 to manipulate multiple objects with different geometries and topologies.

We present ClothesNet: a large-scale dataset of 3D clothes objects with information-rich annotations. Our dataset consists of around 4400 models covering 11 categories annotated with clothes features, boundary lines, and keypoints. ClothesNet can be used to facilitate a variety of computer vision and robot interaction tasks. Using our dataset, we establish benchmark tasks for clothes perception, including classification, boundary line segmentation, and keypoint detection, and develop simulated clothes environments for robotic interaction tasks, including rearranging, folding, hanging, and dressing. We also demonstrate the efficacy of our ClothesNet in real-world experiments. Supplemental materials and dataset are available on our project webpage at https://sites.google.com/view/clothesnet.

Robotic dexterous grasping is a challenging problem due to the high degree of freedom (DoF) and complex contacts of multi-fingered robotic hands. Existing deep re-inforcement learning (DRL) based methods leverage human demonstrations to reduce sample complexity due to the high dimensional action space with dexterous grasping. However, less attention has been paid to hand-object interaction representations for high-level generalization. In this paper, we propose a novel geometric and spatial hand-object interaction representation, named DexRep, to capture object surface features and the spatial relations between hands and objects during grasping. DexRep comprises Occupancy Feature for rough shapes within sensing range by moving hands, Surface Feature for changing hand-object surface distances, and LocalGeo Feature for local geometric surface features most related to potential contacts. Based on the new representation, we propose a dexterous deep reinforcement learning method DexRepNet to learn a generalizable grasping policy. Experimental results show that our method outperforms baselines using existing representations for robotic grasping dramatically both in grasp success rate and convergence speed. It achieves a 93% grasping success rate on seen objects and higher than 80% grasping success rates on diverse objects of unseen categories in both simulation and real-world experiments.

Differentiable Simulations have recently proven useful for various robotic manipulation tasks, including cloth manipulation. In robotic cloth simulation, it is crucial to maintain intersection-free properties. We present DiffClothAI, a differentiable cloth simulation with intersection-free friction contact and two-way coupling with articulated rigid bodies. DiffClothAI integrates the Project Dynamics and Incremental Potential Contact coherently and proposes an effective method to derive gradients in the Cloth Simulation. It also establishes the differentiable coupling mechanism between articulated rigid bodies and cloth. We conduct a comprehensive evaluation of DiffClothAI's effectiveness and accuracy and perform a variety of experiments in downstream robotic manipulation tasks. Supplemental materials and videos are available on our project webpage at https://sites.google.com/view/diffsimcloth.

Building general-purpose robots to perform a diverse range of tasks in a large variety of environments in the physical world at the human level is extremely challenging. According to [1], it requires the robot learning to be sample-efficient, generalizable, compositional, and incremental. In this work, we introduce a systematic learning framework called SAGCI-system towards achieving these above four requirements. Our system first takes the raw point clouds gathered by the camera mounted on the robot's wrist as the inputs and produces initial modeling of the surrounding environment represented as a file of Unified Robot Description Format (URDF). Our system adopts a learning-augmented differentiable simulation that loads the URDF. The robot then utilizes the interactive perception to interact with the environment to online verify and modify the URDF. Leveraging the differentiable simulation, we propose a model-based learning algorithm combining object-centric and robot-centric stages to efficiently produce policies to accomplish manipulation tasks. We apply our system to perform articulated object manipulation tasks, both in the simulation and the real world. Extensive experiments demonstrate the effectiveness of our proposed learning framework. Supplemental materials and videos are available on our project webpage https://sites.google.com/view/egci.

In this paper, we explore whether a robot can learn to hang arbitrary objects onto a diverse set of supporting items such as racks or hooks. Endowing robots with such an ability has applications in many domains such as domestic services, logistics, or manufacturing. Yet, it is a challenging manipulation task due to the large diversity of geometry and topology of everyday objects. In this paper, we propose a system that takes partial point clouds of an object and a supporting item as input and learns to decide where and how to hang the object stably. Our system learns to estimate the contact point correspondences between the object and supporting item to get an estimated stable pose. We then run a deep reinforcement learning algorithm to refine the predicted stable pose. Then, the robot needs to find a collision-free path to move the object from its initial pose to stable hanging pose. To this end, we train a neural network based collision estimator that takes as input partial point clouds of the object and supporting item. We generate a new and challenging, large-scale, synthetic dataset annotated with stable poses of objects hung on various supporting items and their contact point correspondences. In this dataset, we show that our system is able to achieve a 68.3% success rate of predicting stable object poses and has a 52.1% F1 score in terms of finding feasible paths. Supplemental material and videos are available on our project webpage https://sites.google.com/view/hangingobject.

Learning contact-rich, robotic manipulation skills is a challenging problem due to the high-dimensionality of the state and action space as well as uncertainty from noisy sensors and inaccurate motor control. To combat these factors and achieve more robust manipulation, humans actively exploit contact constraints in the environment. By adopting a similar strategy, robots can also achieve more robust manipulation. In this paper, we enable a robot to autonomously modify its environment and thereby discover how to ease manipulation skill learning. Specifically, we provide the robot with fixtures that it can freely place within the environment. These fixtures provide hard constraints that limit the outcome of robot actions. Thereby, they funnel uncertainty from perception and motor control and scaffold manipulation skill learning. We propose a learning system that consists of two learning loops. In the outer loop, the robot positions the fixture in the workspace. In the inner loop, the robot learns a manipulation skill and after a fixed number of episodes, returns the reward to the outer loop. Thereby, the robot is incentivised to place the fixture such that the inner loop quickly achieves a high reward. We demonstrate our framework both in simulation and in the real world on three tasks: peg insertion, wrench manipulation and shallow- depth insertion. We show that manipulation skill learning is dramatically sped up through this way of scaffolding.

The ability to perform in-hand manipulation still remains an unsolved problem; having this capability would allow robots to perform sophisticated tasks requiring repositioning and reorienting of grasped objects. In this work, we present a novel non-anthropomorphic robot grasper with the ability to manipulate objects by means of active surfaces at the fingertips. Active surfaces are achieved by spherical rolling fingertips with two degrees of freedom (DoF) - a pivoting motion for surface reorientation - and a continuous rolling motion for moving the object. A further DoF is in the base of each finger, allowing the fingers to grasp objects over a range of size and shapes. Instantaneous kinematics was derived and objects were successfully manipulated both with a custom handcrafted control scheme as well as one learned through imitation learning, in simulation and experimentally on the hardware.

Autonomous part assembly is a challenging yet crucial task in 3D computer vision and robotics. Analogous to buying an IKEA furniture, given a set of 3D parts that can assemble a single shape, an intelligent agent needs to perceive the 3D part geometry, reason to propose pose estimations for the input parts, and finally call robotic planning and control routines for actuation. In this paper, we focus on the pose estimation subproblem from the vision side involving geometric and relational reasoning over the input part geometry. Essentially, the task of generative 3D part assembly is to predict a 6-DoF part pose, including a rigid rotation and translation, for each input part that assembles a single 3D shape as the final output. To tackle this problem, we propose an assembly-oriented dynamic graph learning framework that leverages an iterative graph neural network as a backbone. It explicitly conducts sequential part assembly refinements in a coarse-to-fine manner, exploits a pair of part relation reasoning module and part aggregation module for dynamically adjusting both part features and their relations in the part graph. We conduct extensive experiments and quantitative comparisons to three strong baseline methods, demonstrating the effectiveness of the proposed approach.

We present an algorithm for transferring physical attributes between webpages and 3D shapes. We crawl product catalogues and other webpages with structured metadata containing physical attributes such as dimensions and weights. Then we transfer physical attributes between shapes and real-world objects using a joint embedding of images and 3D shapes and a view-based weighting and aspect ratio filtering scheme for instance-level linking of 3D models and real-world counterpart objects. We evaluate our approach on a large-scale dataset of unscaled 3D models, and show that we outperform prior work on rescaling 3D models that considers only category-level size priors.