DEXTEROUS MANIPULATION

The world of dexterous hands.

Compare hardware. Discover applications. Share what works. An open reference and community for dexterous manipulation.

Shadow Robot Dexterous Hand Series — Dexterous Hand, L/R

Shadow’s full Dexterous Hand is a five-finger tendon-driven research system. Its manufacturer distinguishes 20 actuated DOF from 24 total joints and offers tactile configurations. Record source date: 2026-09-16.

Understand this hand

Shadow’s full Dexterous Hand is a five-finger tendon-driven research system. Its manufacturer distinguishes 20 actuated DOF from 24 total joints and offers tactile configurations.

What distinguishes it
The full hand and the Lite variants differ in fingers, actuators and overall assembly mass. Select the exact configuration before comparing hardware.
Before you build
Confirm supplied tactile options, wrist and forearm scope, and the software release. Keep the overall assembly weight distinct from any hand-only figure.

Editorial interpretation of public information · reviewed 2026-09-24. Manufacturer / project source ↗

Shadow Robot Dexterous Hand Series — Dexterous Hand, L/R
Evidence review · 2026-09-25

Manufacturer, author and seller statements checked against public sources. This is not an independent hardware test. Fields outside this review retain their earlier evidence dates.

Model / DOF
20 actuated axes, 24 joints, 5 fingers; series comparison checked. Source ↗
Weight
4.3 kg overall assembly; isolated hand mass unreported. Excluded from hand-only mass sorting. Source ↗
Sensing / software
STF configuration and ROS are manufacturer-reported; installed options and supported releases need confirmation. Source ↗
Price / availability
Quote route; no numeric manufacturer price on this page. Stock and lead time unverified. Source ↗

Specifications

Record source date: 2026-09-16

Active / total DOF: 20 / 24

Joint position; 40 tendon-load sensors; 1 IMU; 2 standard / up to 5 Shadow Tactile Fingertips (STF)

Weight basis: 4.3 kg for one hand + forearm. Hand-only weight is not separately disclosed.

Availability / status: Current / marketed

Evidence limits: 24 joints include 2 wrist joints and coupled motion; 20 active DoF. 4.3 kg is overall assembly, not isolated palm.

INTERACTIVE 3D

Explore the hand. Right here.

Rotate, zoom and adjust individual model joints. Downloads begin when you choose Load 3D model. This is geometry, not a physical performance test.

Open larger view ↗ · 27 licensed models available

From model to your first experiment

Browser geometry is verified here. A hardware driver, physics simulation and task policy require separate setup and testing.

  1. Inspect one joint. Load the model, choose a finger joint, move its slider slightly, then Reset. Record the joint name, limits and which segment moved. A slider is a model coordinate, not necessarily a motor.
  2. Get the same source revision. Use the version-specific links below. Check the right/left hand, fingertip and sensor variant before using the geometry in your own project.
  3. Validate your integration. Check metres versus millimetres, radians versus degrees, joint order, limits and mimic/coupling rules. Start with visual inspection; add collisions, inertias, actuator settings and a supported controller before attempting simulation or hardware motion.
Shadow Hand E3M5 right — PST fingertips · BSD-3-Clause · 0.3 MB viewer download

Original manufacturer visual geometry and expanded default URDF; illustrative kinematics only. Right Hand E3M5 with PST fingertips. Catalog H001 is a family association, not exact current unit equivalence. Twenty-four model joint coordinates are not 24 independent actuators; tendon coupling is not implemented by the independent sliders.

Open original model file ↗ · Publisher repository & setup documentation ↗

Source revision: 172c3d9c3715ee75eb10dfe4ee57e26a342b4529
Model path: sr_description/robots/sr_hand.urdf.xacro

git clone https://github.com/shadow-robot/sr_common.git shadow-e3m5-source
cd shadow-e3m5-source
git checkout --detach 172c3d9c3715ee75eb10dfe4ee57e26a342b4529

This checks out source files only; it does not install a simulator or run a robot. Follow the publisher’s dependency and package setup instructions.

This source uses Xacro. Expand it with the publisher’s ROS package dependencies and configuration arguments before loading the resulting URDF.

Find this hand’s software and project references → · Learn joint mapping and action interfaces →

Keep an experiment record: model version, source commit, tool/version, joint mapping, observed result and unresolved problems. Do not describe a geometry preview as a successful hardware test.

Software, projects & datasets

What is ready to build with?

01 · Published supportSource statement

Manufacturer lists ROS integration; release compatibility not tested.

Product source ↗
02 · Developer references11 linked references

Check each link’s model scope, release, files and license. A reference is not a tested integration.

03 · Reproduced hereNot independently tested

No independent hardware performance or SDK integration test is recorded here. Browser model previews show geometry only.

Curated public references, checked 2026-10-25. Links describe a model or family relationship, not hardware-tested compatibility.

Dexterous hand URDF collection

DexSuite · Family resource

Research-maintained visual and collision models for multiple hand families.

Inspect the model directory and upstream asset license; revision, size and sensor geometry may differ from your hand.

Hand motion retargeting

DexSuite · Family resource

Human-to-robot motion optimizers with configurations for the listed hand families.

Retargeting generates joint targets; it is not a device driver. Verify model geometry and joint order.

Shadow Hand ROS interface

Shadow Robot · Family resource

Manufacturer control interfaces and integration examples.

Select the matching hand configuration and ROS branch; Lite variants have different joints.

Shadow robot descriptions

Shadow Robot · Family resource

URDF descriptions and common packages for Shadow hardware.

Check joint count, hand side and hardware revision before using a model.

Shadow Hand in MuJoCo

Google DeepMind · Model-specific

Simulation assets for the E3M5 Shadow Hand, with left and right versions.

A simulation model, not a hardware driver. Match the E3M5 geometry to your unit.

DexGraspNet

PKU EPIC · Research use

Synthetic grasp data and generation code using a Shadow Hand model.

Simulation-generated grasps are not validated hardware executions.

UniDexGrasp

PKU EPIC · Research use

Vision-based dexterous grasp learning with Shadow Hand simulation tasks.

Follow the authors’ environment and model versions; deployment requires additional work.

Human-to-robot hand embodiment

DFKI Robotics Innovation Center · Research use

Research code for transferring human motions to several supported robot hands.

Use the project’s specific model, calibration and motion-capture assumptions; mesh licenses vary.

DexLearn grasp-synthesis baselines

DexLearn contributors · Research use

Learning-based grasp synthesis with documented Shadow-hand configurations.

Research model association only, not a verified driver for a current Shadow hardware revision. The README states noncommercial terms.

AffordDex: task-appropriate grasp contacts

AffordDex research team · Research use

Affordance-aware policy learning with released Shadow-hand simulation tasks.

Simulation research use; hardware support and reuse permissions must be checked separately.

Shadow Hand E3M5 right — PST fingertips — original model package

shadow-robot · Family resource

Original publisher geometry used by the native viewer. BSD-3-Clause; source 172c3d9c3715ee75eb10dfe4ee57e26a342b4529.

Original manufacturer visual geometry and expanded default URDF; illustrative kinematics only. Right Hand E3M5 with PST fingertips. Catalog H001 is a family association, not exact current unit equivalence. Twenty-four model joint coordinates are not 24 independent actuators; tendon coupling is not implemented by the independent sliders.

Learn to audit a model-to-hand interface →

Frequently asked questions

What is the Shadow Robot Dexterous Hand Series — Dexterous Hand, L/R?

Shadow Robot Dexterous Hand Series — Dexterous Hand, L/R is a dexterous robotic hand record in the Dexhands directory. Record source date: 2026-09-16. Shadow’s full Dexterous Hand is a five-finger tendon-driven research system. Its manufacturer distinguishes 20 actuated DOF from 24 total joints and offers tactile configurations.

What are the specifications of the Shadow Robot Dexterous Hand Series — Dexterous Hand, L/R?

Active / total DOF: 20 / 24. Sensing: Joint position; 40 tendon-load sensors; 1 IMU; 2 standard / up to 5 Shadow Tactile Fingertips (STF). Weight basis: 4.3 kg for one hand + forearm. Hand-only weight is not separately disclosed.. Availability / status: Current / marketed.

How much does the Shadow Robot Dexterous Hand Series — Dexterous Hand, L/R cost?

Public price observations are not established for this record. Use the original source or send a supplier inquiry for current pricing.

Where can I buy or source the Shadow Robot Dexterous Hand Series — Dexterous Hand, L/R?

Check the original source links on this page, or use the Dexhands sourcing page to send one request to multiple suppliers.

Product sources

Original source

Original source

Suggest a correction

Start a discussion

Published posts are public. First contributions are reviewed. Only share material you have permission to disclose.

Report a post