The Wuji Glove is a data glove by Shenzhen Wuji Technology Co., Ltd. for precise hand motion capture. It was developed specifically for teleoperation of the Wuji Hand as well as general motion capture applications in robotics. Motion capture modules on each finger record joint angles in real time and transmit the data via Ethernet to the host computer.
The glove is available for both left- and right-handed users and integrates seamlessly into the Wuji software environment (Wuji Studio, Wuji SDK). It supports a pose retargeting system based on the DexRetargeting algorithm and is compatible with Apple Vision Pro for live teleoperation.
Versions
Wuji Glove – Tactile Sensing Version
This version features tactile sensing only and is designed to be worn by the Wuji Hand in order to capture tactile data.
Wuji Glove – Motion Capture Version
This version offers the same tactile sensing capabilities, supplemented by EMF- and IMU-based motion capture functions. It is designed to be worn by the human hand and enables both hand pose and tactile data collection as well as teleoperation of dexterous robotic hands.
| Wuji Glove – Technical Specifications |
| Function |
Data glove for hand motion capture |
| Sensors |
Motion capture modules on each finger (IMU-based) |
| Data Transmission |
Ethernet (USB-C to Ethernet adapter) |
| Default IP Address |
192.168.1.100 (left glove), 192.168.1.101 (right glove) |
| Network Segment |
192.168.1.x (subnet mask 255.255.255.0) |
| Power Supply |
Via USB-C (data + power combined) |
| Versions |
Left-handed and right-handed version |
| Compatible Software |
Wuji Studio, Wuji SDK (Python), ROS2 driver |
| Teleoperation |
Compatible with Apple Vision Pro (DexRetargeting algorithm) |
| Operating Systems |
Windows, Linux |
| Applications |
Teleoperation, robotics research, embodied AI, data collection |
The Wuji Glove is an intelligent data glove designed for robot teleoperation and research in the field of human-computer interaction. It enables real-time capture of tactile perception and posture data of the human hand.
Core Functions
| Function |
Specification |
Output |
| Tactile Sensing |
24 × 32 pressure sensor matrix (534 active points) |
120 FPS, normalized pressure values 0–1 |
| EMF Pose Tracking |
5 RX modules |
120 FPS, 3D position and pose of the fingers |
| IMU |
High-frequency sampling at 1 kHz |
Acceleration + angular velocity |
Typical Application Scenarios
- Teleoperation of robotic dexterous hands
- Tactile data acquisition and research
- Human-computer interaction experiments
- Gesture control for VR/AR
Specifications
Repeatability Test Conditions: Glove samples are mounted on a hand mold fixture. Seven distinct areas of the tactile palm surface are selected. A pressure applicator with a flat disc probe (20 mm diameter) repeatedly applies the same force to each area. The maximum percentage deviation is then calculated.
Press Life / Flex Life Test Conditions: Glove samples are worn by a test operator who repeatedly performs open and gripping motions (up to full scale) on a cylindrical force measurement device.
Glove Parameters
| Category |
Parameter |
Value |
| Dimensions & Weight |
Glove size (L × W) |
215 × 195 mm |
| |
Thickness (material / with structure) |
6.55 mm / 35 mm |
| |
Total length (with cable) |
340 mm |
| |
Glove weight |
67.5 g |
| |
Total weight (with cable) |
192.6 g |
| Tactile Sensors |
Sensor type |
Pressure sensor matrix |
| |
Array resolution |
24 × 32 (534 active points) |
| |
Sampling rate |
120 FPS |
| |
Pressure range |
0 – 20 N (20 mm disk probe) |
| |
Pressure resolution |
0.1 N |
| |
Spatial resolution |
4 mm |
| |
Coverage |
Entire palm |
| |
Repeatability |
±6 % |
| |
Pressure lifetime |
100,000+ |
| |
Bending lifetime |
100,000+ |
| Spatial Tracking (EMF) |
Receiver modules |
5 RX modules, each with 6 degrees of freedom |
| |
Sampling rate |
120 FPS |
| |
Position accuracy |
≤ 2 mm rms |
| |
Orientation accuracy |
< 5° rms |
| |
Tracking range |
≤ 30 cm |
| IMU |
Number / type |
6 × 6-axis IMU |
| |
Sampling rate |
1000 Hz |
| |
Accelerometer |
±16 g |
| |
Gyroscope |
±2000 °/s |
| Power & Communication |
Power supply |
USB Type-C, 5 V / 1 A |
| |
Data interface |
RJ45 (via adapter), 100BASE-TX |
| |
Glove connection |
USB Type-C |
| Environment |
Operating temperature |
-5 to 40 °C |
| |
Storage temperature |
-10 to 70 °C |
| |
Operating humidity |
< 55 % RH |
| |
Storage humidity |
< 80 % RH |
EMF Specifications
| Category |
Parameter |
Value |
| Electromagnetic Sensor |
Sampling rate |
120 Hz |
| |
Motion-to-data latency (wired) |
≤ 10 ms |
| Hand Kinematics |
Sensor distribution |
Thumb, index finger, middle finger, ring finger, little finger, back of the hand |
| Sensor Data |
Fingertips |
3-axis position + 3-axis rotation |
| |
Back of the hand |
3-axis rotation |
| Joint Calculation |
|
Real-time inverse kinematics for the entire hand + joint angles |
| Position Tracking |
External tracker mounting |
Mechanical interface available |
| |
Optical marker mounting |
Mechanical interface available |
| Environment & Reliability |
Interference susceptibility |
Sensitive to metallic / ferromagnetic materials; recommended distance ≥ 3 cm |
| |
Reliability |
No mechanical sensor wear, no electromagnetic hysteresis |
| Calibration |
Calibration steps |
8 steps |
| |
Objective |
Priority on precise joint angles for improved object interaction and gesture recognition |
| |
Optimization parameters |
Bone length, joint pose, fingertip pose |
Components
Palm
- More than 500 pressure sensors are distributed across the palm, forming a high-resolution tactile sensing system.
- Captures pressure changes in real time when interacting with objects and generates high-resolution pressure distribution maps of the hand.
- During grasping, the system provides feedback on force distribution across different areas of the palm – from light touch to firm grip.
Back of the Hand
- Six motion-capture modules form a complete tracking system for hand movement.
- Five modules are located at the fingertips and one in the center of the back of the hand.
- By combining these sensor data with kinematic algorithms, the glove calculates finger flexion angles, finger spreading, and the spatial pose of the palm.
Data Acquisition Board
The cables from the motion-capture modules run through a protected cable sleeve to the data acquisition board located on the arm. An LED indicator on the board displays the current status of the glove.
Scope of Delivery
- 1 × Wuji Glove (left or right)
- 1 × Data acquisition board with arm strap
- 1 × USB-C to Ethernet adapter (with power supply)
- 1 × USB-C to USB-C cable
- Documentation / quickstart guide (download)
