Sharpa Wave

Source-backed capabilities

• Capability areas: Robotic grasping, Dexterous manipulation
• Workflows: End-effector manipulation
• Operating environments: Manipulation labs, robot hand research, mobile manipulator integration and tactile-sensing testbeds.
• Target markets: Researchers, robot builders and integrators evaluating tactile robotic hands.
• Deployment model: Prototype dexterous hand product associated with Sharpa Robotics.
• Tactile interaction sensing

  Research demonstration - 2026-06-15 - T-Rex evaluates two physical 22-DoF Sharpa Wave hands with per-fingertip tactile deformation and six-axis wrench sensing across 12 force-sensitive and deformable-object tasks, reporting more than 30% higher average success than the strongest baseline.

  Source
• Dexterous manipulation

  Research demonstration - 2026-09-27 - The Sharpa Wave and xArm7 system demonstrated dexterous manipulation by solving all ten scrambled 2×2×2 cubes in approximately 137 seconds on average, including pickup, table-supported changes of grasp and final placement.

  Source
• Force-controlled manipulation

  Research demonstration - 2026-06-15 - T-Rex evaluates two physical 22-DoF Sharpa Wave hands with per-fingertip tactile deformation and six-axis wrench sensing across 12 force-sensitive and deformable-object tasks, reporting more than 30% higher average success than the strongest baseline.

  Source
• Dexterous manipulation

  Research demonstration - 2026-06-15 - T-Rex evaluates two physical 22-DoF Sharpa Wave hands with per-fingertip tactile deformation and six-axis wrench sensing across 12 force-sensitive and deformable-object tasks, reporting more than 30% higher average success than the strongest baseline.

  Source
• Deformable-object manipulation

  Research demonstration - 2026-06-15 - T-Rex evaluates two physical 22-DoF Sharpa Wave hands with per-fingertip tactile deformation and six-axis wrench sensing across 12 force-sensitive and deformable-object tasks, reporting more than 30% higher average success than the strongest baseline.

  Source
• Bimanual manipulation

  Research demonstration - 2026-06-15 - T-Rex evaluates two physical 22-DoF Sharpa Wave hands with per-fingertip tactile deformation and six-axis wrench sensing across 12 force-sensitive and deformable-object tasks, reporting more than 30% higher average success than the strongest baseline.

  Source
• Object manipulation

  Research demonstration - 2026-06-15 - CAIP study evaluates dual Sharpa Wave hands across six manipulation tasks

  Source
• Remote and shared control

  Claimed - 2026-06-15 - Teleoperate a dexterous hand with tactile feedback.

  Source

• Production Setup

  mass production announced - 2026-01-06 - robotic-hand production - Sharpa said its SharpaWave human-sized dexterous robotic hand entered mass production, moving the component from prototype status into scalable production.

  Source
• Production Setup

  robotic hand entered mass production - 2025-12-19 - dexterous robotic hand production - Third-party coverage reported Sharpa’s advanced robotic hand entering mass production, moving the dexterous-hand product from prototype toward scaled hardware availability.

  Source

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

• MANUS gloves

  2026-08-18 - Research Experiment - Controlled By - Physical - Pre-training Visual Dexterity used MANUS gloves and Quest tracking to teleoperate two YAM Pro arms fitted with Sharpa Wave hands, with three RealSense D405 cameras.

  Source
• dual UR7e arms, Quest 3 and DexTeleop-0

  2026-07-14 - Research Experiment - Controlled By - Mixed - DexTeleop-0 used dual UR7e arms and Sharpa Wave hands with Quest 3 tracking and an Isaac Sim digital twin for force-aware bimanual teleoperation.

  Source
• Franka arm and Heterogeneous Tactile Transformer

  2026-07-14 - Research Experiment - Tested With - Physical - A Sharpa Wave hand on a Franka arm ran camera-free HTT tactile experiments, reaching 95% toy-screw tightening success.

  Source
• FTP-1 tactile foundation policy

  2026-07-03 - Research Experiment - Trained With - Mixed - FTP-1 used Sharpa Wave tactile data in pretraining and evaluated long-horizon tasks on a Sharpa North platform with dual Wave hands.

  Source
• dual UR3e arms and Do as I Do pipeline

  2026-07-03 - Research Experiment - Controlled By - Physical - Do as I Do retargeted internet video to dual UR3e arms with Sharpa Wave hands and executed ten real-world motions.

  Source
• MANUS gloves

  2026-06-18 - Research Experiment - Controlled By - Physical - T-Rex research used two Dexmate Vega-1 arms, Sharpa Wave hands, multi-view cameras, MANUS gloves and VIVE trackers to collect a 100-hour tactile dataset across 12 tasks.

  Source
• Dexmate Vega bimanual platform and CAIP

  2026-06-15 - Research Experiment - Tested With - Physical - CAIP was evaluated on a Dexmate Vega bimanual platform with dual Sharpa Wave hands, reporting 76% average success across six tasks.

  Source
• H2 Plus

  2026-06-01 - Reference Design - Mounted - Physical - H2 Plus was announced in a reference configuration with two Sharpa Wave tactile hands. Unitree's current product table lists the hands as optional.

  Source
• IMCopilot and MoDE-VLA

  2026-03-09 - Research Experiment - Tested With - Physical - A Sharpa Wave configuration tested IMCopilot and MoDE-VLA on apple peeling, gear, charger and tube manipulation tasks.

  Source
• Isaac Lab

  2026-03-09 - Digital Simulation - Connected - Simulated - A MANUS demonstration teleoperated a simulated 22-DoF Sharpa Wave inside NVIDIA Isaac Lab 2.3 and fed demonstrations into Isaac Lab Mimic.

  Source
• DexEMG and EMG2Pose retargeting

  2026-03-06 - Research Experiment - Controlled By - Physical - DexEMG used EMG2Pose retargeting to control Sharpa Wave on unfamiliar-object and longer-horizon manipulation tasks.

  Source
• MANUS gloves

  2026-03-04 - Research Experiment - Trained With - Mixed - NVIDIA EgoScale used a Sharpa hand action space and MANUS/Vive capture for aligned human–robot data, with transfer tests on a Unitree G1 tri-finger hand.

  Source
• Tacmap tactile simulator

  2026-02-19 - Research Experiment - Trained With - Mixed - Tacmap policies trained in tactile simulation transferred to physical Sharpa Wave hardware without real-world fine-tuning.

  Source
• KUKA arm and SimToolReal

  2026-02-18 - Research Experiment - Tested With - Physical - SimToolReal used a KUKA arm with Sharpa Wave and 120 real rollouts for zero-shot dexterous tool manipulation.

  Source
• Astribot S1

  Digital Simulation - Mounted - Simulated - BiDex-100 uses a simulated Astribot S1 embodiment with Sharpa Wave hands for ten generalisation tasks.

  Source
• Franka Research 3

  Digital Simulation - Mounted - Simulated - BiDex-100 simulates two Franka Research3 arms with22DoF Sharpa Wave hands for96tasks.

  Source
• Vega 1 Pro

  Research Experiment - Mounted - Physical - FTP-1 evaluates Dexmate Vega with two mounted Sharpa Wave tactile hands in a contact-rich manipulation research configuration.

  Source
• DITTO-X actuated exoskeleton

  Research Experiment - Controlled By - Physical - Sharpa Wave is physically used for battery insertion, tong use and raspberry placement. DITTO-X supplies forward force-feedback teleoperation and reverse hand-pose matching before human intervention.

  Source
• UFACTORY xArm 7

  Research Experiment - Mounted - Physical - FINGR used a Sharpa Wave hand mounted on an xArm7 for autonomous 2×2 cube turns and solves, with marker tracking and table-assisted regrasping.

  Source
• Unitree H2 Plus in NVIDIA Isaac GR00T Reference Humanoid Robot

  Reference Design - Mounted - Physical - NVIDIA's Isaac GR00T reference design combines a Unitree H2 Plus body with dual Sharpa Wave hands and Jetson AGX Thor T5000 compute.

  Source

Supply / Components

• Sharpa Wave

  Robotic Actuation Modules - Actuation - Hesai Technology - Hesai actuation modules are installed in Sharpa Wave robotic hands - source: whbl.com

  Source

• BiDex-100: A Comprehensive Benchmark for Bimanual Dexterous Manipulation Across 100+ Simulation Tasks

  2026-10-18 - Research Publication - BiDex-100 is a simulation benchmark with 106 bimanual dexterous tasks across five capability axes. It supplies8,080 validated teleoperated demonstrations across dual-Franka and Astribot S1 simulation embodiments.

  Source
• FACT: A Multimodal Force-Action-Track Foundation Model for Human-to-Robot Dexterity

  2026-10-13 - Research Publication - FACT combines visual planning with force-aware action refinement for dexterous manipulation. Its ten-task physical evaluation uses YAM Ultra arms and Sharpa Wave hands; robot pretraining also reuses the T-Rex collection from a Dexmate Vega-1.

  Source
• Forward and Reverse Teleoperation for Dexterous Manipulation and Human Intervention

  2026-09-30 - Research Publication - An actuated exoskeleton renders force and vibrotactile feedback during forward teleoperation. Reverse teleoperation aligns the operator’s hand with the policy-controlled robot before takeover. The described DITTO-Human collection spans six tasks, three hands and more than 16 hours; public downloads remain pending.

  Source
• FINGR: Learning Dexterous Hand Control for Real-World Rubik’s Cube Solving

  2026-09-27 - Research Publication - UC San Diego’s FINGR learns contact-rich finger control from demonstrations using finger-relative geometry and future-interaction prediction. Physical evaluation uses a Sharpa Wave hand on an xArm7, with coded-marker cube tracking and table-assisted regrasping.

  Source
• DexTacWAM: A Visuo-Tactile World-Action Model for Dexterous Manipulation

  2026-09-21 - Research Publication - DexTacWAM extends a video world-action model with per-fingertip tactile prediction and a hand-aware tactile compressor. Across six contact-rich tasks on a bimanual Dexmate and Sharpa setup, it averages 70.6 against 38.0 for the strongest baseline.

  Source
• Robotic Origami Challenge: Fold Plane Demonstrations

  2026-06-21 - Research Publication - A Sharpa Wave dataset for contact-rich, long-horizon imitation learning, with synchronized vision, tactile video, fingertip force/torque data and bimanual joint actions.

  Source
• MagicSim: A Unified Infrastructure for Executable Embodied Interaction

  2026-06-16 - Research Publication - MagicSim proposes a shared runtime for simulated robot control, skills and planning, combining executable robot embodiments with multiple drivers. Its registry includes Unitree G1 variants and a Vega-1P mobile manipulator with Sharpa Wave hands.

  Source
• T-Rex: Tactile-Reactive Dexterous Manipulation

  2026-06-15 - Research Publication - T-Rex evaluates two physical 22-DoF Sharpa Wave hands with per-fingertip tactile deformation and six-axis wrench sensing. The work introduces 100 hours of tactile-rich robot data and reports more than 30% higher average success than the strongest baseline on 12 force-sensitive and deformable-object tasks.

  Source
• Contrastive Action-Image Pre-training for Visuomotor Control

  2026-06-15 - Research Publication - CAIP learns action-centred visual representations from 32,041 hours of egocentric human video and 88 hours of robot manipulation data. Physical evaluation uses a Dexmate Vega with two 22-DoF Sharpa Wave hands and reports more than 30% gains on folding, pouring and fine-grained manipulation tasks.

  Source
• FTP-1: A Generalist Foundation Tactile Policy Across Tactile Sensors for Contact-Rich Manipulation

  2026-06-11 - Research Publication - FTP-1 trains a generalist tactile policy across heterogeneous sensors and embodiments using about 3,000 hours from 26 data sources. Physical evaluations include Sharpa North and a Sharpa-equipped Dexmate platform, alongside a separate simulated benchmark.

  Source
• MoDE-VLA

  2026-03-09 - Research Publication - Sharpa released a paper describing RL-augmented teleoperation and Mixture-of-Dexterous-Experts VLA for contact-rich bimanual manipulation The work uses shared-autonomy primitives for high-DoF control and combines vision language force and touch inputs for dexterous execution The system was demonstrated on apple peeling, a contact-rich manipulation task used to stress test dual dexterous hand control Sharpa released a paper describing RL-augmented teleoperation and Mixture-of-Dexterous-Experts VLA for contact-rich bimanual manipulation The work uses shared-autonomy primitives for high-DoF control and combines vision language force and touch inputs for dexterous execution The system was demonstrated on apple peeling, a contact-rich manipulation task used to stress test dual dexterous hand control

  Source
• EgoScale: Scaling Dexterous Manipulation with Diverse Egocentric Human Data

  2026-02-18 - Research Publication - EgoScale pretrains a vision-language-action policy on 20,854 hours of action-labelled egocentric video and evaluates physical bimanual manipulation on Galaxea R1Pro with two 22-DoF Sharpa Wave hands. Its final policy reports a 54% average success-rate improvement over no pretraining across five dexterous tasks.

  Source
• SimToolReal: An Object-Centric Policy for Zero-Shot Dexterous Tool Manipulation

  2026-02-18 - Research Publication - SimToolReal trains one object-centric reinforcement-learning policy in simulation and deploys it on a physical 22-DoF Sharpa five-finger hand. Across 120 real-world rollouts spanning 24 tasks, 12 objects and six tool categories, it reports 37% higher task progress than retargeting and fixed-grasp baselines while matching specialist policies.

  Source

• Sharpa Wave completes autonomous cube solves in UC San Diego’s FINGR study

  2026-09-27 - Research Publication - FINGR achieved 297 successful layer turns out of 300 attempts, compared with 79.7% success for the base flow policy. The Sharpa Wave and xArm7 system demonstrated dexterous manipulation by solving all ten scrambled 2×2×2 cubes in approximately 137 seconds on average, including pickup, table-supported changes of grasp and final placement. The released dataset contains 36 operator-recorded sequences and 576 turns, with tactile measurements, joint states, tracked cube geometry and turn annotations.

  Source
• Sharpa launches Wave SE dexterous hand and opens pre-orders

  2026-08-19 - Product Launch - Wave SE retains 22 independently actuated degrees of freedom in a human-scale hand intended for embodied-AI research and robot integration. The SE configuration is rated for a 24 kg payload, 90 N grip force, 12 N fingertip force and operation at 2 Hz. Its tactile configuration provides a 60 × 60 array, 0.05 N force sensitivity, 2 mm spatial resolution and onboard inference at up to 30 frames per second.

  Source
• Sharpa Wave tactile hands join NVIDIA Isaac GR00T humanoid reference design

  2026-05-31 - Partnership - NVIDIA announced the Isaac GR00T Reference Humanoid Robot using a Unitree H2 Plus body, Sharpa Wave tactile five-finger hands, Jetson Thor onboard compute, and Isaac GR00T software workflows. Sharpa Wave supplies dexterous tactile manipulation hardware for academic humanoid robotics research and physical-AI development. The reference design gives Sharpa a high-visibility integration point in the humanoid research ecosystem.

  Source

Market context

Sharpa Wave is Sharpa's dexterous robotic hand for high-resolution tactile sensing and fine manipulation.

• Target environment: Manipulation labs, robot hand research, mobile manipulator integration and tactile-sensing testbeds.
• Workflow context: Fine manipulation, contact-rich sensing, object handling, fingertip feedback and dexterous end-effector research.
• Customer context: Researchers, robot builders and integrators evaluating tactile robotic hands.
• Deployment model: Prototype dexterous hand product associated with Sharpa Robotics.
• Commercial maturity: Prototype-stage product with public positioning and limited deployment evidence.
• Adoption constraints: Adoption depends on sensing resolution, grip force, durability, control interface and integration support.
• Market position: Tactile dexterous-hand lane where sensing quality is the central differentiator.
• Adjacent products: Shadow Dexterous Hand, Linker Hand L20, AGILINK OmniHand, OceanTrix hand

Comparable Systems

• Aero Hand Open

  Chestnut Robotics - Dexterous Hand

• OmniHand O10 2025

  AgiBot - Hand

• OmniHand Pro 2025

  AgiBot - Hand

• AIDIN-Hand

  Aidin Robotics - Robotic Hand

• Amazing Hand

  Pollen Robotics - Robotic Hand

• Apex Hand

  Dexcel Robotics - Dexterous Hand