distance-profile grasping

Existing methods for generating 3D hand–object interactions struggle to simultaneously achieve user controllability and generalization across diverse object geometries. To address this, this work proposes a controllable generation framework based on a distance-aware grasping energy term. The approach first employs a diffusion transformer to generate a distance field and an initial hand pose, followed by a pose refinement step within near-contact regions to enforce physical plausibility. The key innovation lies in the introduction of the distance field and its perception-weighted mechanism, which effectively captures semantically similar interaction patterns while remaining invariant to specific hand and object identities. This method enables high-quality, physically plausible, and user-controllable grasp synthesis, demonstrating strong generalization across a wide range of objects and hand shapes.

This work addresses the challenge of insufficient robustness in grasp pose generation under partial point cloud observations by proposing a novel approach that integrates data-driven energy priors with geometric optimization. The method uniquely combines a learned energy-based model (EBM) and the Iterative Closest Point (ICP) algorithm within a Stein Variational Gradient Descent (SVGD) framework, enabling energy-guided iterative refinement for efficient grasping of unseen objects. Experimental results demonstrate that the proposed approach achieves a success rate of 60.9% across 5,360 grasp attempts on 67 objects, significantly outperforming state-of-the-art methods including AnyGrasp (31.1%), GPD (48.4%), and AS-ICP (56.6%), thereby confirming its superior generalization capability under partial observability.

Existing contact-based grasp generation methods often neglect physical attributes such as contact forces, resulting in insufficient grasp stability. To address this, we propose a force-aware contact modeling framework: normal contact forces are discretized into one-hot encodings, enabling joint integration of contact geometry and physical stability constraints; gradient-based optimization is then performed with acceleration minimization as the objective. This work is the first to explicitly model contact force distribution and geometric structure in an end-to-end manner for stable grasp pose generation. Evaluated on two public benchmarks, our method improves stability metrics by approximately 20% and demonstrates strong generalization to unseen objects. Moreover, it accurately identifies critical stable contact points essential for robust grasping.

This work addresses the limitations of traditional multi-finger grasping, which relies on trajectory planning and is prone to failure under object pose perturbations, often necessitating costly replanning. The paper presents the first end-to-end, planner-free grasping controller that leverages a Grasping Distance Field (GDF) in configuration space, driving execution via its negative gradient. Safety is rigorously enforced through a Control Barrier Function–Control Lyapunov Function (CBF-CLF) quadratic program, while hysteresis-based mode switching and a force-closure quality barrier preserve grasp performance. The method provides theoretical guarantees on GDF approximation accuracy and forward invariance of the safe set. In simulation across 50 diverse objects, the approach successfully grasped 46, achieving a 94% grasp-quality retention rate, with each QP solve requiring only 0.09 ms within a 20 ms control cycle.

Modeling multi-contact dynamical systems remains challenging due to modeling complexity and poor real-time performance. To address this, we propose an end-to-end differentiable dynamics modeling framework based on Signed Distance Functions (SDFs). Our method introduces a novel dual-SDF architecture: one SDF encodes the supporting plane for efficient collision detection, while the other—coupled with contact dual cones—enables physically consistent, time-stepped state prediction. The entire model is fully differentiable, enabling gradient-based optimization and seamless integration into learning-based control and real-time closed-loop optimization. In simulation, the framework achieves high-fidelity dynamics fitting; on the Allegro dexterous hand, it accomplishes in-hand object reorientation within ≈2 minutes of online learning, operating at 30–60 Hz control frequency. To our knowledge, this is the first fully differentiable, closed-form, and computationally efficient multi-contact dynamics model, establishing a new paradigm for model-based real-time dexterous manipulation control.

Existing dexterous grasp synthesis methods struggle to simultaneously achieve high success rates and diverse contact patterns across objects of varying scales. To address this challenge, this work proposes the HUGS framework, which innovatively leverages a small-scale human grasp dataset to learn a conditional prior that adaptively recommends both contact modes and initial wrist poses. Integrated with force-closure-aware optimization, HUGS unifies the generation of multimodal grasps—ranging from precision pinches to bimanual coordination. The approach overcomes limitations of traditional heuristic strategies, synthesizing 3.2 million grasp samples across 157,000 scenes covering objects from 2 to 30 cm in size. Real-world experiments demonstrate that the system autonomously selects appropriate contact modes, successfully grasping diverse objects—from small screws to large boxes—with high reliability.

This work addresses the disconnect between dexterous hand design and task-driven control, as well as the limited optimization dimensions in existing approaches, by proposing a unified parametric co-design framework that jointly optimizes palm structure, finger kinematics, fingertip geometry, and surface curvature. Innovatively incorporating fine-grained geometric features—such as surface curvature—into the design space, the method employs a parametric surface deformation kernel to directly model contact interactions, enabling end-to-end simulation-to-reality optimization. Integrating multi-degree-of-freedom finger modeling, task-oriented optimization algorithms, and manufacturability constraints, the approach significantly enhances grasp stability in dynamic manipulation tasks and produces dexterous hand models ready for both simulation and physical fabrication. The resulting designs are open-sourced to advance research in co-design and cross-platform policy transfer.

为解决手-物体交互的统一表示问题，提出InterMASH方法，利用球固定锚点和低度球谐波编码几何与接触信息，通过条件扩散变换器生成一致且物理上合理的抓握。

This work addresses the challenge of grasp generation for heterogeneous dexterous hands, which arises from differences in kinematic topology, actuation dimensionality, and command spaces. The authors propose a unified graph-based modeling approach that represents a robotic hand as a kinematic graph derived from its URDF, integrating hierarchical object surface encoding, differentiable forward kinematics, and dynamic graph message passing to directly generate feasible grasps in palm pose and joint space. Innovatively leveraging conditional flow matching, the method circumvents inverse kinematics, post-hoc optimization, and retargeting, thereby enabling cross-hand generalization. Evaluated on Barrett, Allegro, and Shadow hands, the approach achieves an average success rate of 83.48% with a single inference time of 40 ms, and without retraining, it attains a 72.70% success rate even on finger-joint–missing variants.

本文提出CoToGrasp框架，通过学习规范工作空间来生成稳定且多样的抓握方式，解决了现有灵巧抓握规划器仅优化物理稳定性的问题。