Skip to main content

Python

2F-85 gripper

SDK

Supported by Robotiq

Example of TSF usage via python are available developped and maintained by Robotiq team is available the the folowing repository:

https://github.com/robotiq/tactile_sensors/tree/main/sensor_quickstart

Are below are the related instructions to use this driver.

Intro​

Lightweight cross-platform tool to test TSF-85 connections.

Quick Start — Terminal​

Linux​

cd sensor_quickstart
./run_quick_connect.sh

Windows​

cd sensor_quickstart
run_quick_connect.bat

That's it! The script handles everything automatically.

Quick Start — Web Viewer​

A browser-based dashboard with real-time heatmaps, dynamic time-series with FFT, and IMU plots.

Linux​

cd sensor_quickstart
./run_web_viewer.sh

Windows​

cd sensor_quickstart
run_web_viewer.bat

The script sets up the environment, connects to the sensor, and opens your browser to http://localhost:8080. The dashboard has three tabs:

  • Static — live tactile heatmap (7x4 grid per finger) with baseline subtraction
  • Dynamic — dynamic tactile time-series and FFT spectrum
  • IMU — accelerometer and gyroscope plots

The server shuts down automatically when you close the browser tab.

Requirements​

  • Python 3.7+: Download Python
    • ✅ Check "Add Python to PATH" during installation
    • ✅ After installing, restart your terminal/command prompt
  • pyserial: Installed automatically by the script

What It Does​

  1. Checks for Python installation
  2. Creates virtual environment (.venvSimpleCheck)
  3. Installs dependencies
  4. Detects sensor
  5. Displays real-time sensor data

Expected Output​

================================================================================
                        Robotiq Tactile Sensor Monitor
================================================================================
Data Rate: 160.234 KB/s  |  Refresh Rate: 1000.1 Hz  |  Total Packets: 15234

FINGER 0
--------------------------------------------------------------------------------
  Static Tactile (7 rows × 4 columns):
      0     1     2     3
      4     5     6     7
      8     9    10    11
     12    13    14    15
     16    17    18    19
     20    21    22    23
     24    25    26    27

  Dynamic Tactile:    123

  Accelerometer: X=    12  Y=   -45  Z=  1024
  Gyroscope:     X=     3  Y=    -2  Z=     1
  Magnetometer:  X=   -15  Y=    23  Z=   -87

  Open Byte:    0

FINGER 1
--------------------------------------------------------------------------------
  [... same format ...]

================================================================================
Press Ctrl+C to exit

Troubleshooting​

Sensor Not Found​

Linux:

  • Check USB connection
  • Verify sensor is plugged in
  • Try different USB port

Windows:

  • Check Device Manager (Win+X → Device Manager → Ports)
  • Sensor appears as "USB Serial Device" or "Cypress USB UART"
  • Try different USB port
  • VM users: USB passthrough may not work reliably for serial devices

No Data Displayed​

  • Unplug and replug the sensor
  • Close terminal and rerun script
  • Sensor may need to be reset

Python Not Found (Windows)​

  • Install Python from python.org
  • Must check "Add Python to PATH" during installation
  • Restart command prompt after installing

Learn More​

Sensor Details​

  • Baud rate: 115200
  • Format: 8N1 (8 data bits, no parity, 1 stop bit)
  • USB VID:PID: 16d0:14cc (Robotiq) or 04b4:f232 (Cypress, older units)
  • Data: 28 tactile sensors per finger (7×4 grid) + IMU + dynamic sensor

File Structure​

sensor_quickstart/
├── quick_connect.py         # Terminal-based sensor monitor
├── web_viewer.py            # Web-based visualization server
├── protocol.py              # USB protocol implementation
├── requirements.txt         # Dependencies (pyserial, websockets)
├── run_quick_connect.sh     # Linux launcher (terminal)
├── run_quick_connect.bat    # Windows launcher (terminal)
├── run_web_viewer.sh        # Linux launcher (web UI)
├── run_web_viewer.bat       # Windows launcher (web UI)
├── web/                     # Web UI assets
│   ├── index.html
│   ├── app.js
│   └── style.css
└── README.md

Press Ctrl+C to stop the sensor monitor