Robotiq Releases Open-Source Software For Physical AI And Robotics

(MENAFN- Robotics & Automation News) Robotiq has launched three open-source software releases for its adaptive grippers: a C++ SDK, an official ROS 2 package, and updated Nvidia Isaac Sim assets. All three are maintained and optimized by Robotiq and available now on github/robotiq.

Robotiq adaptive grippers have run in ROS 2 and Isaac Sim for years, mainly driven by the community of robotic developers who wrote the drivers, built the simulation models, and maintained the grippers packages on their own time so the hardware would work inside the tools they already used. That work carried the ecosystem a long way, and a good deal of it is still visible in what shipped this month.

“What changed is the demand. A physical AI application asks far more of a gripper than a conventional one ever did,'' said Vincent Duchaine, CTO AI at Robotiq.

“It calls for higher fidelity in simulation, tighter control loops on real hardware, much more of what the gripper can report about what it is holding, and deployments that keep collecting this data across millions of cycles.

“The tools move fast on their own account and staying compatible is continuous work. Packages maintained in spare hours were never meant to carry that kind of weight.”

To better meet the growing demand resulting from physical AI, Robotiq is taking ownership by optimizing, maintaining, and meeting developers where they already work.

The ROS 2 packages are a drop-in replacement for the community driver most labs already run, matching its interfaces so switching costs a single line. The C++ SDK gives the same control as a stand-alone library for teams building their own stack. The Isaac Sim assets now enable closed-loop kinematics, which is essential for the high-fidelity simulation of the gripper.

These new open-source software releases also point at something wider. They are the first pieces of Contact Core, Robotiq's software layer for physical AI.

Contact Core gives model builders the simulation assets, control packages, and synchronized contact-rich data interfaces they need to train manipulation models on Robotiq components and deploy them on real robots.

The next step is embedded intelligence at the point of contact, expected for release in Q4.

See Robotiq at IROS 2026

Robotiq will show its new innovation exclusively at the IEEE/RSJ International Conference on Intelligent Robots & Systems (IROS) 2026 in Pittsburgh on September 28 at booth 427.

MENAFN23092026005532012229ID1111705726