Skip to content
#Grasping and manipulation
Oct 20Tue
  1. Guyuehome35

    Guoli Intelligence wraps up 2026 CIIF: Dexterous hand applications across multiple scenarios

    At the 2026 CIIF, Guoli Intelligence showcased multiple dexterous hands and multi-scenario application solutions, and the exhibition has now concluded successfully. Three live demonstrations—dual-arm teleoperation, autonomous parts sorting, and vision-based gesture recognition—presented coordinated perception, control, and execution capabilities, drawing visitors, industry experts, and partners for observation and exchange.

  2. 人形大讲堂 · 微信公众号66

    FA Robotics completes consecutive C+ and D funding rounds, with nearly 1 billion yuan earmarked for core component capacity

    Collaborative robot company FA Robotics announced it has completed consecutive C+ and D funding rounds totaling nearly 1 billion yuan. The rounds were led by Meituan Longzhu and Meituan respectively, with participation from Yongxin Capital, Shishan Phase II, and Suzhou Venture Capital, while existing shareholders Guolian Minsheng Capital, Jiangsu Production Investment, and Qingliu Capital followed on.

  3. 高工咨询 · 微信公众号64

    Casia Hand Robotics debuts 7 industry-grade dexterous hands and 4 industrial solutions at CIIF

    Casia Hand Robotics exhibited for the first time at the 26th China International Industry Fair, showing 7 industry-grade dexterous hands across the M, X, and G series and 4 industrial-grade solutions. The Casia Hand-M6 weighs only 410 g with a 10 kg payload, a payload-to-weight ratio of more than 24 times; the flagship X12 uses a 12-DOF fully direct-drive architecture, combining a 30 kg handling payload with 150 N grip force and an operating life of more than 1 million cycles at rated load; the G4 three-finger dexterous hand reaches a life of 500,000 cycles.

  4. OFweek Robotics72

    Fa'ao Robotics completes nearly 1 billion yuan in C+ and D rounds to expand humanoid force-controlled arm and core component capacity

    Fa'ao (Suzhou) Robotics Technology Co., Ltd. announced the consecutive completion of its C+ and D funding rounds, totaling nearly 1 billion yuan. The C+ round was led by Meituan Longzhu, and the D round was led by Meituan. The funds will go toward building production capacity for core components, as well as expanding production and R&D for the ART series industrial-grade humanoid force-controlled arm and the AIRLab embodied intelligence platform.

  5. PaperWeekly · 微信公众号74

    SenseTime SenseNova-RoboRSI unified Harness raises GPT-6-Astra success rate on RoboDojo from 15.83% to 54.17%

    A joint research team from SenseTime, Nanyang Technological University and other institutions has introduced the SenseNova-RoboRSI unified Harness, which raises GPT-6-Astra's manipulation success rate on 42 RoboDojo tasks from 15.83% in the baseline configuration to 54.17%, with model parameters unchanged.

  7. Beijing Humanoid Robot Innovation Center58

    Renmin University GeWu-Lab and Beijing Humanoid Innovation Center to livestream Robo-ValueRL millimeter-level VLA reinforcement learning solution

    Renmin University of China's GeWu-Lab and the Beijing Humanoid Innovation Center will livestream a presentation of the Robo-ValueRL millimeter-level VLA reinforcement learning solution on October 21, 2026 at 19:00 Beijing time, with talks by Hu Di and Xia Wenke.

  8. 机器人全球资讯 · 微信公众号31

    First Real-World Intelligent Robot Skills Exhibition (Nanjing Stop) Held on October 20, with 27 Jiangning Companies Including Estun and Zhongke Guiji Exhibiting

    The First Real-World Intelligent Robot Skills Exhibition (Nanjing Stop) was held on October 20 at Southeast University's Jiulonghu Campus, where ten typical application scenarios for intelligent manufacturing in Nanjing were released, showcasing equipment such as special robots, intelligent inspection robots, dexterous hands, and underwater operation robots.

  9. Tech Xplore Robotics62

    Seoul National University's Dynin-Robotics robot foundation model guides actions by predicting future scenes, achieving 78.4% average success rate in real-world operation

    Professor He Jiong's team at Seoul National University's AIDAS Lab proposed Dynin-Robotics, a robot foundation model that unifies language, images, video, and robot actions in a single model, generating and selecting actions by predicting future scene images after an action; the paper has been accepted by NeurIPS 2026.

  10. Robotics & Automation News47

    Primepick brings NSF-certified MealMate unattended hot-food robotic kiosk to the US market

    Primepick is bringing the MealMate unattended hot-food robotic kiosk to the US market. It has already been deployed at manufacturing plants in New Jersey and Connecticut and at Molloy University, and is being rolled out to hospitals and nursing homes in the Tri-State area, with three new sites in Huntsville, Chicago, and Dallas in preparation.

  11. Robotics & Automation News55

    Tesollo dexterous hands cumulative sales expected to surpass 1,000 units in 2026

    South Korean robotic hand company Tesollo expects cumulative sales of its robotic hands and grippers to exceed 1,000 units by the end of 2026. The company expanded its production facilities in 2026, and its product line ranges from single-finger mechanisms to five-finger dexterous hands, with customers expanding from research and proof-of-concept to industrial deployment.

  13. Robotics & Automation News38

    Why Inconsistent Parts Bring Down Automated Production Lines

    Small differences in part dimensions and shape can cause automated production lines to stop frequently, misjudge good parts, or miss defective ones, because each station sets its path, fixtures, and force limits for a single nominal part. Differences such as stamping burrs, injection-molding flash, and machining dimensional drift accumulate; at Cp 1.00, about 0.27% of parts are out of tolerance, and at Cp 1.33 this falls to about 64 per million.

  14. 深圳市机器人协会 · 微信公众号44

    FAO showcases embodied intelligence production line deployment results at the Industrial Fair

    At the 26th China International Industry Fair Robotics Show (RS2026), FAO demonstrated its embodied intelligence deployment capabilities through a full-scale live production line and reached ecosystem strategic cooperation agreements with five partners: Aitu Technology, Boyin Hechuang, Yingshen Intelligence, Xingyuanzhi, and Lingqiao Intelligence.

Oct 19Mon
  1. 中国机器人网 · 微信公众号47

    2026 China Robot Competition Adds Embodied Dexterous Manipulation Track for the First Time

    The 2026 China Robot Competition (RoboCup China Open) was held at the Hangzhou Baimahu International Convention and Exhibition Center, featuring an embodied dexterous manipulation track for the first time, with two events: "Three-Finger Dexterous Manipulation — Ball Tower Challenge" and "Five-Finger Dexterous Manipulation — Artistic Skills Challenge."

  2. 高工咨询 · 微信公众号66

    Fa'ao Robotics completes nearly 1 billion yuan in C+ and D rounds, showcases force-controlled humanoid arm and intelligent workstation at CIIF

    Fa'ao Robotics announced at the 2026 China International Industry Fair that it has completed its C+ and D funding rounds, with a cumulative scale of nearly 1 billion yuan. Investors include Meituan Dragon Ball, Yongxin Capital, Shishan Phase II, and Suzhou Venture Capital, among others. The funds will be invested in expanding production capacity for core components and in scaling up R&D for industrial embodied intelligence.

  3. arXiv cs.RO78

    Evaluating a Frozen Frontier VLM Directly as a Robot Policy on the RoboMME Memory-Dependent Manipulation Benchmark

    The study uses a frozen frontier vision-language model, GPT-6-Astra, directly as the robot policy and tests it on RoboMME, a benchmark of 16 memory-dependent manipulation tasks. The model keeps the entire episode context throughout and directly outputs each end-effector waypoint and gripper command, with no training specific to this benchmark.

    Why it matters: The paper provides complete comparative data on a frozen frontier VLM used directly as a policy on a memory-dependent manipulation benchmark, which can be used to assess how much of the embodied stack a general-purpose model can take over.

  4. arXiv cs.RO67

    RoboHarness: Simple Robot Harness Lets LLM Agents Outperform VLA and World Action Models on BEHAVIOR Challenge 2025

    An arXiv paper presents RoboHarness, a robot harness that gives LLM agents a visual-geometric control panel so they can directly understand and invoke embodied tasks. The authors argue that coding agents have the potential to achieve generalist embodied manipulation without the route of large-scale VLA training. Experiments show that, with RoboHarness, Qwen3.8-Flash-Next outperforms state-of-the-art VLA models on BEHAVIOR Challenge 2025 tasks, and the implementation is open source.

  5. arXiv cs.RO55

    DexBioLab: A dexterous manipulation simulation benchmark for real biological laboratory scenarios

    The research team proposes DexBioLab, a simulation benchmark and framework for dexterous manipulation in real biological laboratory environments, comprising 14 tasks across 4 robot embodiments, covering functional tool use, precision manipulation, language-conditioned visual grounding, and instrument operation.

  8. arXiv cs.RO72

    ADEPT: Accelerating Dexterous Manipulation via Reinforcement Learning Pre-Training and Post-Training

    ADEPT is a large-scale reinforcement learning framework that first pre-trains a dexterous policy on a generic object-reposing task, then post-trains downstream policies with this prior, enabling long-horizon tasks to be solved directly from raw visual-tactile perception across high-DoF embodiments. To prevent naive fine-tuning from degrading zero-shot reposing ability, the authors propose a stable post-training scheme combining behavior-cloning distillation, critic warm-up, and conservative on-policy updates, and introduce a joint-space Geometric Fabric that coordinates the policy with the robot.

  9. arXiv cs.RO62

    MANGO-Grasp: A Cross-Embodiment Dexterous Grasping Framework Based on Geometry-Oriented 3D Gaussians and Mahalanobis Fields

    MANGO-Grasp proposes an anisotropic interaction framework that represents objects with geometry-oriented 3D Gaussian primitives to preserve local surface geometry, encodes hands as pretrained descriptors that capture both morphology and kinematics, and models hand-object contact through anisotropic Mahalanobis fields.

  10. arXiv cs.RO64

    FACT post-training method improves VLA success rate in contact-rich manipulation tasks

    The paper investigates why vision-language-action (VLA) models fail in contact-rich manipulation tasks that require precise physical interaction, and proposes two failure modes: precision failures (policies struggle to produce the small, precise corrective actions needed during contact) and force failures (policies fail to respond to contact forces).

  11. arXiv cs.RO55

    FELT: Generating Tactile Signals from Vision for Visuo-Tactile Manipulation

    FELT is a framework that uses existing paired visuo-tactile data to learn contact-aware representations from RGB observations, providing them to downstream policies through two interfaces: generated tactile images and contact-aware features. It combines a frozen visual encoder with separate query decoders for the left and right fingers, and the two branches exchange information through gated connections, with a convolutional readout producing spatial contact representations. On four contact-intensive real-robot tasks, both interfaces achieved higher success rates than vision-only policies, and directly integrating contact-aware features into policies raised success rates by 15-22.5 percentage points.

  12. arXiv cs.RO48

    EDAR: Learning Environment-Dependent Action Representations for Robotic Manipulation

    EDAR (Environment-Dependent Action Representation) anchors action tokens in both executable control structure and expected visual consequences. By coupling motor commands with their environment-conditioned effects, it enables the learned action space to capture interaction semantics rather than merely command-level patterns.

  13. arXiv cs.RO62

    NativeMEM: A Native Memory Compression Method for Long-Horizon Robotic Manipulation

    NativeMEM is a vision-language-action (VLA) policy with long-horizon, real-time updated memory. Its core is the Native Memory Compression encoding scheme, which reuses the VLA's own vision encoder to compress each historical frame from each camera view into a single token. Appended to the input sequence, these tokens let the pretrained VLA attend to long-term history with negligible latency overhead, requiring neither an external planner nor a newly initialized memory module.

  14. arXiv cs.RO59

    WireCraft: A Simulation Benchmark for Industrial Deformable Linear Object Manipulation

    WireCraft is a simulation benchmark for industrial deformable linear object (DLO) manipulation, covering three task families: connector insertion, clip routing, and channel seating. It supports two physics models, articulated and deformable, and the released trajectories come from both simulation and a real UR5 robotic arm.

  15. arXiv cs.RO62

    HydroShear: Fluid-elastic shear simulation for tactile sim-to-real reinforcement learning

    HydroShear is a non-holonomic fluid-elastic tactile simulator accepted to RSS 2026, addressing tactile sim-to-real policy transfer for contact-intensive tasks. It models stick-slip transitions, path-dependent force and shear accumulation, and full SE(3) object-sensor interaction, and uses signed distance functions (SDFs) to track the displacement of indenter surface points as they interact with the sensor membrane, generating physically grounded and computationally efficient force fields from arbitrary watertight geometries.

  16. arXiv cs.RO70

    Blind assembly strategy: peg-in-hole assembly of diverse parts using proprioception alone

    This study trains blind assembly policies that observe only joint positions, velocities, and a short history of their own actions, without using visual, force, torque, or tactile sensing, and uses reinforcement learning in simulation to complete peg insertion, gear meshing, and nut tightening. The peg insertion policy transfers zero-shot to the real world and successfully inserts pegs of different shapes and sizes in 73% of trials; controlled experiments show that the blind policy generalizes better to unseen part shapes and sizes but may be slower.

  17. arXiv cs.RO58

    PRIME: Learning Precise Progress Estimation from Imprecise Demonstrations

    The paper proposes PRIME, a progress estimator for generating dense reward signals in reinforcement learning for robotic manipulation, learning precise task progress estimation from imprecise demonstration data. PRIME builds a weighted graph over observed states, connecting states across trajectories based on spatial proximity and visual consistency; edge weights encode spatial distance as transition cost, and shortest-path search approximates the geodesic distance to success, whose changes can distinguish progress, stagnation, and regression without assuming that progress is monotonic over time.

  18. arXiv cs.RO53

    Embodiment-Aware Planning for Heterogeneous Dual-Arm Manipulation

    This study proposes an embodiment-aware planning method in which two Franka FR3 arms equipped with different simple grippers (a suction cup and a parallel-jaw gripper) cooperate to complete manipulation tasks that a single gripper cannot perform reliably. Given a desired object motion specified by a human demonstration, the planner optimizes arm-trajectory assignments over a finite horizon, determining segment by segment whether single-arm manipulation or a coordinated dual-arm handover is required, while jointly considering heterogeneous grasp feasibility, motion constraints, manipulability, and cross-gripper compatibility; it also computes globally consistent assignments and adds explicit switching regularization.

  20. arXiv cs.RO70

    No Vision Needed: Touch-Only Dexterous Hand Baoding Ball Manipulation Achieves Zero-Shot Transfer

    The research team achieved Baoding ball rotation on the Shadow Dexterous Hand Lite using only proprioception and binary tactile feedback, enabling zero-shot sim-to-real transfer. In a single continuous trial on the physical hand, it completed 225 180° rotations (112.5 full rotations) at about 0.36 rotations per second, with no vision or object-state information throughout. The authors also compared the roles of touch and pure proprioception, identifying the hand regions where tactile information is most critical.

  21. arXiv cs.RO68

    ActiveAlign bridges active perception and VLA policies via observation-aligned data adaptation

    The paper proposes ActiveAlign, a data adaptation framework that addresses the mismatch between the initial observation state produced by active camera motion and the manipulation policy's training distribution (first-frame distribution sensitivity): on LIBERO, directly prepending camera repositioning to pi_0.5 reduces the success rate from 46.3% to 11.5%.

  22. arXiv cs.RO55

    SCOPE: A Selective Program Edit Certification Method for Embodied Heuristic Learning

    SCOPE (Selective Certification Of Program Edits) uses failed and successful executions as complementary evidence to guide program revision: frontier-guided program search annotates a shared diagnostic graph with execution evidence to distinguish verified progress from unresolved checks, while selective behavioral certification uses stateful API replay to restore recorded responses and mutations and checks whether edits preserve selected historical successful behaviors.

  23. arXiv cs.RO45

    GRASP: Global-Local Representation Alignment and Spatial-Physical Grounding for Bimanual Robotic Manipulation

    GRASP proposes a training-phase prior injection framework for bimanual robotic manipulation. A physics-grounded flow-matching loss uses differentiable forward kinematics (FK) to supervise tool center point (TCP) poses and inter-arm relative poses, and the framework introduces a global query group distilled from DINOv3 together with local interaction queries supervised by Grounding DINO; the teacher models are removed at inference. Simulation and real-world long-horizon bimanual task experiments show that the two priors are complementary under different deployment conditions: the full model performs better in in-distribution deployment, while the distilled query branch is more suitable under strong visual randomization.

  24. arXiv cs.RO58

    VLA-Fabric: Learning Composable Interaction Mechanisms for Multi-Arm VLA Agents

    VLA-Fabric assigns a complete vision-language-action (VLA) model to each robotic arm, preserving each arm's local perception and action ownership, and achieves coordination through explicitly learned interaction across policy boundaries. The paper systematically studies where interaction should occur, how peer information should be aggregated, and how interaction should be learned, finding that shared context, peer-specific latent variables, and action-stage interaction play complementary coordination roles, and reproduces this principle on structurally different VLA architectures.

  25. arXiv cs.RO60

    Robotic Spatiotemporal Editing: A Unified and Controllable Action Generation Interface via Masked Action Reconstruction

    The paper proposes the MAP (Masked Action Space) interface, which unifies spatiotemporal references such as keypoints, flows, orientations, and key poses as partial constraints in the same temporal Cartesian action space, and models controllable manipulation as an action inpainting problem.

  26. arXiv cs.RO57

    Hardware-as-Regularization: Improving Imitation Learning Data Consistency with an Adjustable Passive Compliant Leader Arm

    An arXiv paper proposes the Hardware-as-Regularization method, which uses a low-cost passive joint module (an adjustable mechanical clutch plus a 3D-printed spiral spring) to pull each joint toward a user-defined neutral posture, so that the passive restoring force acts as a physical prior for joint-space L2 regularization.