Spes Robotics is proud to support the release of Webots R2025a , bringing major enhancements to robot simulation.
What's New:
- New robots & demos: Heron USV, ROSbot, ROSbot XL, Crazyflie
- New W3D format replaces X3D/VRML
- Improved Supervisor API with better PROTO access
- ROS 2 Jazzy support, Docker for ROS 1, Iron dropped
- Performance boosts & bug fixes
Built for educators, researchers, and developers — this release marks another step forward for open-source robotics simulation.
🔗
Download Webots R2025a
📄
View Full Changelog
💡
Spes Robotics – Official Support Partner
We recently had the pleasure of collaborating with a talented master's student, Marija Golubović, on an exciting imitation learning project! Her work focused on developing end-to-end ML visuomotor control for the UFACTORY Lite6 robot arm, allowing it to autonomously pick up a screwdriver with precision.
Marija’s project included building an NVIDIA Omniverse Isaac Sim simulation environment for training and development and creating a ROS 2 system to support real-time servoing and sim2real transfer. This is our second master project this year, and we are continually amazed by the depth of innovation students bring to our team.
A huge thanks to Marija for her hard work and dedication! Her project delivered promising results in imitation learning, and we are excited to keep expanding our R&D in this direction.
🔗 Check out her work here: https://github.com/MarijaGolubovic/robo_imitate/
We decided to open-source a robot arm teleoperation tool we developed as a part of an internal project. It allows you to move your smartphone in free space, the web app tracks the smartphone movements (essentially ARCore that fuses RGB and IMU) and reports the movements to a Python server to servo the robot arm in 6 DoF. It is easy to install (just a Python package) and it doesn't require any special hardware.
We hope this tool will be valuable to the community!
We provide examples with + Universal Robots and + UFACTORY Lite6.
Check out the project!
https://github.com/SpesRobotics/teleop
We are excited to share a remarkable project by Miloš Nikolić, who recently defended his master's thesis. Miloš developed a high-fidelity simulation and control system for an agricultural robot designed to navigate field rows with precision.
Key Features:
- Powered by ROS 2 and Isaac Sim: Ensuring robust simulation and control
- High-Fidelity Graphics: Delivering realistic and detailed visuals
- Computer Vision and ML: For accurate row detection
- Path Generation with Nav2: Enabling smooth U-turns and seamless navigation
- Synthetic Dataset Generation: Enhancing training and overall performance
We are proud to support the growth and development of the Memristor Robotics team for the second year in a row. Our commitment to advancing technology and fostering innovation makes this support a perfect match. Congrats Memristor on the national Eurobot results and we wish Memristor luck in France! P.S. Memristor Robotics is committed to open-source development, you can check their ROS 2 based software stack at: https://github.com/memristor/mep3
