winter term 2025/2026
Technical University of Applied Sciences Augsburg, Prof. Andreas Muxel
- install Docker Desktop https://docs.docker.com/desktop/
- start Docker Desktop
The following steps just needs to be done once, when you start simulation for the first time:
- on the bottom of Docker Desktop open ">_ Terminal"
- type in terminal
docker run --name ursim -it --network bridge -p 2222:2222 -p 29999:29999 -p 30001-30004:30001-30004 -p 5900:5900 universalrobots/ursim_e-series ROBOT_MODEL=UR5
Otherwise, if executed the next time:
- select "VNC Viewer" on the left side (view install instrucions below)
- no credentials needed
- select session and press "Connect"
- open https://hub.docker.com/extensions/pgmystery/docker-extension-vnc
- select "Open in Docker Desktop" and install extension
- create a new session using the "+" on the topleft
- Give a "Session Name", for example "ursim"
- Connection-Type is "Docker Container"
- refresh the next field and look for "/ursim" and select it
- set VNC port to 5900
- select option "Stop Container after disconnect"
- press "Connect"
- power on virtual robot (very similar to real robot)
- when finished"Disconnect"
- install Apple's Command Line Developer Tools with Terminal:
xcode-select --install
- see next steps here: https://www.macports.org/install.php
- run in Terminal:
sudo port install boost
- download and install Anaconda here: https://www.anaconda.com/
- once Anaconda is installed, open it, create a new environment and call it "robot" (make sure you are using a python version that is supported by the ur-rtde package here: https://pypi.org/project/ur-rtde/ -> Python version 3.12 should work for Windows and Mac
- video tutorial Prof. Michael Kipp: https://www.youtube.com/watch?v=sDKVHmPsEEY&t=29s (just Anaconda needed, skip Tensorflow, Seaborn & Jupiter Noteook)
- start environment with play button and select "Open Terminal"
- type the following commands in the Terminal (might take some time to install, esprecially on Mac):
pip install ur_rtde
pip install python-osc
- download code repository, https://github.com/HybridThingsLab/course-robotics/archive/refs/heads/2025.zip and unzip
- launch VS Code from Anaconda at "Home" with "robot" selected in the top of the Home tab
- add folder "course-robotics-2025" to your workspace with "File/Add Folder to Workspace"
- save workspace
- find scripts to read and write data to robot viaOpen Sound Control (OSC) in subfolder of folder "/Python"
- view Folder in "Explorer" (icon documents left)
- in the "Python/RTDE_OSC_read" folder in VSCode navigate to -> src\config.json and change the “ip” to the IP Address of the robot (simulation or real robot) (burger menue top right "About")
- in the "Python/RTDE_OSC_write" folder in VSCode navigate to -> src\RobotController.py and change the “ip” to your IP Address of the robot (simulation or real robot)
- open two terminal windows in VSCode side-by-side by using the split option in the terminal (icon window left from trash can)
- in one terminal window change the directory to RTDE_OSC_read\src by typing "cd " and drag and drop "src" folder to terminal
- run the script by typing
python main.py
- in the other terminal window change the directory to RTDE_OSC_write\src by typing "cd " and drag and drop "src" folder to terminal
- if you run the real robot type in the terminal
python main.py --driverobot
- if you run robot as a simulation with Docker type in the terminal
python main.py --driverobot --docker
- you should get an message "INFO:root:connected!"
- open one example provided in the repository (p.ex. "TouchDesigner/02_RobotController/robotController.toe")
- to stop a Python script press "ctrl+c" in the terminal
- to restart use UP or DOWN key to recall last prompts and press RETURN
- any tool with an OSC interface can be used to control the robot
- see also code examples provided in this repository
- downloads:
- TouchDesigner, https://derivative.ca/product/touchdesigner-non-commercial/77
- Processing, https://processing.org/download
- p5js, https://p5js.org/download/
- ...