Bridging the Sim-To-Real Gap with multipanda_ros2: A Real-Time ROS2 Framework for Multimanual Systems
Jon �kerlj, Seongjin Bien, Abdeldjallil Naceri, Sami Haddadin
AI summary
Problem
Existing multi-robot control architectures lack the real-time performance, flexible controller switching, and high-fidelity simulation integration required for safe, contact-rich physical human-robot interaction and dual-arm tasks.
Approach
The authors developed an open-source ROS 2 architecture featuring a controllet design pattern to manage arbitrary Franka robots at 1 kHz, integrated with a MuJoCo simulator and quantitative sim-to-real metrics, alongside a methodology for real-world inertial parameter identification.
Key results
- Achieves ≤2 ms controllet-switching delays for seamless controller transitions
- Sustains a stable 1 kHz real-time torque control loop across multiple Franka robots
- Quantifies sim-to-real fidelity using kinematic and dynamic consistency metrics
- Demonstrates that real-world inertial parameter identification significantly improves force and torque accuracy
Why it matters
Provides robotics researchers and engineers with a robust, reproducible, open-source platform for developing and benchmarking advanced multi-robot interaction and sim-to-real transfer algorithms.
Abstract
We present multipanda ros2, a novel open- source ROS 2 architecture for multi-robot control of Franka Robotics robots. Leveraging ros2 control, this framework provides native ROS 2 interfaces for controlling any number of robots from a single process. Our core contributions address key challenges in real-time torque control, including interaction control and robot-environment modeling. A central focus of this work is sustaining a 1kHz control frequency, a necessity for real-time control and a minimum frequency required by safety standards. Moreover, we introduce a controllet-feature design pattern that enables controller-switching delays of ≤ 2 ms, facilitating reproducible benchmarking and complex multi-robot interaction scenarios. To bridge the simulation-to- reality (sim2real) gap, we integrate a high-fidelity MuJoCo simulation with quantitative metrics for both kinematic ac- curacy and dynamic consistency (torques, forces, and control errors). Furthermore, we demonstrate that real-world inertial parameter identification can significantly improve force and torque accuracy, providing a methodology for iterative physics refinement. Our work extends approaches from soft robotics to rigid dual-arm, contact-rich tasks, showcasing a promising method to reduce the sim2real gap and providing a robust, reproducible platform for advanced robotics research.