KORTHOS / ECOSYSTEM / PRODUCT

DEX-EE Hand

About

About DEX-EE Hand

DEX-EE is Shadow Robot’s three-fingered research hand, built with 12 degrees of freedom for repeated manipulation experiments. Position, force and inertial sensing track its motion and contact with objects. Optical fingertip sensors provide detailed tactile feedback, supplemented by sensors along the fingers, so learning systems can respond to contact during a grasp.

High-bandwidth torque and position control and ROS integration connect the hand to experimental software. The symmetric DEX-EE geometry differs from DEX-EE Chiral, which offsets its third finger like a thumb and is offered in left, right and paired configurations. The hardware is designed for user maintenance and includes protective shutdown routines for demanding learning experiments.

Capabilities

Source-backed capabilities

  • Capability areas: Robotic grasping, Dexterous manipulation
  • Workflows: End-effector manipulation, Dexterous grasping
  • Operating environments: Research labs, robotics development environments and advanced manipulation research facilities.
  • Target markets: Robotics researchers, manipulation labs and AI research teams working on dexterous manipulation.
  • Deployment model: Commercial product sold to research and advanced development customers.
  • Dexterous manipulation

    Research demonstration - 2025-06-13 - ExoStart transfers vision policies to a physical DEX-EE across seven grasping, reorientation and tool-use tasks; six reached at least 50 percent success over 50 physical trials.

    Source
  • Robotic grasping

    Research demonstration - 2025-06-13 - ExoStart transfers vision policies to a physical DEX-EE across seven grasping, reorientation and tool-use tasks; six reached at least 50 percent success over 50 physical trials.

    Source
  • Precision insertion

    Research demonstration - 2025-06-13 - ExoStart transfers vision policies to a physical DEX-EE across seven grasping, reorientation and tool-use tasks; six reached at least 50 percent success over 50 physical trials.

    Source
  • Doors and articulated objects

    Research demonstration - 2025-06-13 - ExoStart transfers vision policies to a physical DEX-EE across seven grasping, reorientation and tool-use tasks; six reached at least 50 percent success over 50 physical trials.

    Source

Observed use

Exact-product configurations are shown without inferring supply, purchase, customer status, standard fit, deployment or partnership.

  • MANUS gloves

    2026-06-15 - Physical Demo - Controlled By - Physical - Shadow Robot demonstrated both Shadow Hand and DEX-EE Chiral at ICRA 2026 using MANUS gloves.

    Source

Research

  • Exploiting Policy Idling for Dexterous Manipulation

    2025-08-21 - Research Publication - The Pause-Induced Perturbations study evaluates policy improvement on a real precision connector-insertion task using a KUKA iiwa 14 with a three-finger Shadow DEX-EE hand and fingertip tactile sensing. The method improves absolute real-world success by 15 to 35 percentage points by perturbing policies when idling is detected.

    Source
  • ExoStart: Efficient learning for dexterous manipulation with sensorized exoskeleton demonstrations

    2025-06-13 - Research Publication - ExoStart uses a sensorized exoskeleton matched to the Shadow DEX-EE kinematics, dynamics filtering and auto-curriculum reinforcement learning to train vision policies for a physical DEX-EE on a KUKA iiwa. Seven real tasks include key turning, nut unscrewing, peg insertion, cube flipping, case opening and bulb installation; six achieved 50 percent or higher success over 50 physical trials.

    Source
  • DemoStart: Demonstration-led auto-curriculum applied to sim-to-real with multi-fingered robots

    2024-09-10 - Research Publication - DemoStart trains multi-finger manipulation policies from sparse rewards and a small set of simulated demonstrations, then transfers pixel-based policies zero-shot to physical KUKA iiwa systems fitted with Shadow DEX-EE hands. The paper reports using a fleet of six real setups and 100 times fewer demonstrations than the compared real-robot approach.

    Source

Resources

Media library

Official Shadow Robot product specifications, an exact simulation model and rights-conscious manufacturer media for the DEX-EE Hand.

Timeline

  • Demonstration: Shadow Robot demonstrated both Shadow Hand and DEX-EE Chiral at ICRA 2026 using MANUS gloves

    2026-06-15 - Demonstration - MANUS gloves were used to control Shadow Hand during its ICRA 2026 demonstration. The DEX-EE Chiral hand was demonstrated separately with the same glove interface.

    Source
  • Shadow Robot unveils DEX-EE hand developed with Google DeepMind

    2024-05-09 - Product Launch - Shadow Robot announced DEX-EE on May 9, 2024, positioning it as a robust dexterous hand for AI research. The company said DEX-EE was built by Shadow Robot and developed with research and insights from the Google DeepMind robotics team. Shadow said the hand would be demonstrated publicly at ICRA in May 2024.

    Source

Context

Market context

Three-finger dexterous robot hand developed with Google DeepMind for demanding real-world machine-learning and manipulation experiments.

  • Target environment: Research labs, robotics development environments and advanced manipulation research facilities.
  • Workflow context: Dexterous grasping, precision manipulation and tactile interaction on advanced robotic platforms.
  • Customer context: Robotics researchers, manipulation labs and AI research teams working on dexterous manipulation.
  • Deployment model: Commercial product sold to research and advanced development customers.
  • Commercial maturity: Commercially available product with strong positioning in the dexterous manipulation research market.
  • Adoption constraints: End-effector level product requiring integration into a complete robotic system. Priced for the research market rather than broad commercial use.
  • Market position: Extends Shadow Robot's specialist hand lineage into a current dexterous manipulation product for research and advanced control work.
  • Adjacent products: Ability Hand, Shadow Hand series, dexterous robotic hands

Comparable Systems