KORTHOS / ECOSYSTEM / PRODUCT

Shadow Dexterous Hand Series

About

About Shadow Dexterous Hand Series

Shadow Dexterous Hand uses 20 motors to control 24 joints across five fingers, palm and wrist. Tendon-driven actuation supports finger abduction, opposition, palm flexion and detailed in-hand manipulation within an anthropomorphic structure weighing approximately 4.3 kilograms.

More than 100 sensors provide joint, tendon and tactile information at rates reaching one kilohertz. EtherCAT and ROS integration connect the hand with robot arms, teleoperation gloves and learning systems. Shadow supplies lighter four-, three- and two-finger variants within the same product family.

Capabilities

Source-backed capabilities

  • Capability areas: Tactile Dexterity, Robotic grasping, Force control
  • Workflows: End-effector manipulation
  • Operating environments: Robotics labs, AI research settings, teleoperation systems and humanoid or arm-based manipulation testbeds.
  • Target markets: Researchers, labs, humanoid builders and AI teams needing a canonical high-DoF hand platform.
  • Deployment model: Commercial specialist end-effector in Shadow Robot's Dexterous Hand series.
  • Tactile interaction sensing

    Claimed - Shadow Robot presents its Dexterous Hand for precise tool and object manipulation, tactile research and glove-based teleoperation in remote or hazardous environments.

    Source
  • Object manipulation

    Research demonstration - 2022-10-11 - The paper evaluates multi-agent reinforcement learning for in-hand manipulation tasks on the Shadow Dexterous Hand.

    Source
  • Robotic grasping

    Claimed - Shadow Robot presents its Dexterous Hand for precise tool and object manipulation, tactile research and glove-based teleoperation in remote or hazardous environments.

    Source
  • Dexterous manipulation

    Claimed - Shadow Robot presents its Dexterous Hand for precise tool and object manipulation, tactile research and glove-based teleoperation in remote or hazardous environments.

    Source
  • Remote and shared control

    Research demonstration - 2023-04-10 - The paper introduces a 6D object-in-hand pose dataset collected through teleoperation with a Shadow Dexterous Hand.

    Source
  • Remote and shared control

    Research demonstration - 2018-09-17 - Imitation and grasping trials with novice users show faster and more reliable teleoperation than the compared vision-based method.

    Source
  • Remote and shared control

    Claimed - Shadow Robot presents its Dexterous Hand for precise tool and object manipulation, tactile research and glove-based teleoperation in remote or hazardous environments.

    Source

Supply

Supply / Components

  • Shadow Dexterous Hand

    Moulded Fingers And Cnc Aluminium Tactile Parts - Manufacturing - Protolabs - Protolabs moulded finger parts and CNC-machined aluminium tactile-sensor parts for Shadow Dexterous Hand development - source: protolabs.com

    Source

Research

  • FIRMGrasp: A Friction-Informed Risk Margin for Robust Grasp Synthesis

    2026-07-27 - Research Publication - Simulation studies test friction-robust grasp ranking on LEAP, Allegro and Shadow hand models.

    Source
  • Cross-Embodiment Robot Manipulation via a Unified Hand Action Space

    2026-07-03 - Research Publication - Cross-embodiment transfer is physically tested on LEAP Hand and simulated on Allegro and Shadow hands.

    Source
  • DMFC-GraspNet: Differentiable Multi-Fingered Robotic Grasp Generation in Cluttered Scenes

    2023-08-01 - Research Publication - Robotic grasping is a fundamental skill required for object manipulation in robotics.

    Source
  • PoseFusion: Robust Object-in-Hand Pose Estimation with SelectLSTM

    2023-04-10 - Research Publication - Accurate estimation of the relative pose between an object and a robot hand is critical for many manipulation tasks.

    Source
  • A Multi-Agent Approach for Adaptive Finger Cooperation in Learning-based In-Hand Manipulation

    2022-10-11 - Research Publication - In-hand manipulation is challenging for a multi-finger robotic hand due to its high degrees of freedom and the complex interaction with the object.

    Source
  • GradTac: Spatio-Temporal Gradient Based Tactile Sensing

    2022-03-14 - Research Publication - GradTac processes BioTac SP signals on a Shadow Dexterous Hand to improve tactile tracking under fluid-sensor noise.

    Source
  • Grasping in the Dark: Zero-Shot Object Grasping Using Tactile Feedback

    2020-11-02 - Research Publication - A tactile slip controller enables a BioTac-equipped Shadow Dexterous Hand to grasp previously unseen objects without object models.

    Source
  • Demonstration-Guided Deep Reinforcement Learning of Control Policies for Dexterous Human-Robot Interaction

    2019-06-27 - Research Publication - Demonstration-guided reinforcement learning controls a Shadow Dexterous Hand for handshake, hand-clap, and finger-touch interactions.

    Source
  • Shadow Dexterous Hand teleoperation paper maps hand depth images to robot joint angles

    2019-05-01 - Research Publication - This paper covers Shadow Dexterous Hand teleoperation paper maps hand depth images to robot joint angles for Shadow Dexterous Hand in dexterity.

    Source
  • Computational Tactile Flow for Anthropomorphic Grippers

    2019-03-19 - Research Publication - Computational tactile flow extracts motion and surface cues from pressure sequences recorded with a Shadow Dexterous Hand.

    Source
  • Vision-based Teleoperation of Shadow Dexterous Hand using End-to-End Deep Neural Network

    2018-09-17 - Research Publication - TeachNet maps human-hand depth images directly to Shadow C6 joint angles for markerless dexterous-hand teleoperation.

    Source
  • Learning Dexterous In-Hand Manipulation

    2018-08-01 - Research Publication - Simulation-trained reinforcement-learning policies transfer to a physical Shadow Dexterous Hand for vision-based object reorientation.

    Source
  • Multi-goal RL benchmark paper includes in-hand manipulation with Shadow Dexterous Hand

    2018-02-26 - Research Publication - This paper covers Multi-goal RL benchmark paper includes in-hand manipulation with Shadow Dexterous Hand for Shadow Dexterous Hand in dexterity.

    Source
  • Multi-Goal Reinforcement Learning: Challenging Robotics Environments and Request for Research

    2018-02-26 - Research Publication - First of all, it introduces a suite of challenging continuous control tasks (integrated with OpenAI Gym) based on currently existing robotics hardware.

    Source

Resources

Media library

Official Shadow Robot specifications, integration references, research guidance and exact product media for the Shadow Dexterous Hand.

Timeline

  • Innovate UK awards Shadow Robot GBP 650k innovation loan for touch-transmitting teleoperation

    2019-04-17 - Funding - Innovate UK awarded Shadow Robot a GBP 650k innovation loan for remote-touch teleoperation technology. The announcement centered on the Shadow Dexterous Hand, tactile fingertip sensing and haptic feedback for remote operation over long distances. Shadow positioned the work for hazardous or controlled environments including nuclear decommissioning, bomb disposal, space exploration and pharma labs.

    Source
  • Shadow Robot, SynTouch and HaptX demonstrate Tactile Telerobot

    2018-06-24 - Partnership - Shadow Robot said the Tactile Telerobot integrated its dexterous hand with SynTouch BioTac sensors and HaptX haptic gloves. The team demonstrated an operator in California controlling a Shadow hand in London while receiving touch feedback from the robot fingertips. ANA Holdings funded and facilitated the integration as part of its Avatar initiative.

    Source
  • Shadow Robot starts Innovate UK semi-autonomous teleoperation project

    2017-03-21 - Partnership - Shadow Robot said an Innovate UK-backed semi-autonomous teleoperation demonstrator launched on March 1, 2017. The project paired Shadow Dexterous Hand hardware with an OC Robotics snake-arm robot, Cambrian Intelligence augmented-VR work and UCL vision/testing support. The goal was to let operators command higher-level teleoperation tasks rather than directly controlling every robot movement.

    Source

Context

Market context

Anthropomorphic five-finger dexterous robot hand for manipulation research, AI/ML experimentation, tactile-sensing work and teleoperation.

  • Target environment: Robotics labs, AI research settings, teleoperation systems and humanoid or arm-based manipulation testbeds.
  • Workflow context: Dexterous grasping, tactile sensing, teleoperation, sim-to-real learning and manipulation research.
  • Customer context: Researchers, labs, humanoid builders and AI teams needing a canonical high-DoF hand platform.
  • Deployment model: Commercial specialist end-effector in Shadow Robot's Dexterous Hand series.
  • Commercial maturity: Commercial and long-running research platform with public specs and broad research recognition.
  • Adoption constraints: Adoption depends on cost, integration complexity, maintenance, control software, tactile requirements and task fit.
  • Market position: Canonical dexterous-hand research platform for testing high-DoF manipulation, tactile sensing and learning workflows.
  • Adjacent products: AGILINK SkillHand, Linker Hand L30, DexRobot DexHand021 Pro, OceanTrix hand

Comparable Systems