Skip to main content

🦾 Meet the YuMi Robot

The ABB IRB 14000 YuMi is a dual-arm collaborative robot designed for small-part assembly, precision manipulation, and safe human-robot interaction.
At the RAIL Lab, YuMi is primarily used for research on manipulation, perception, robot learning, and interactive robotic systems.


🌟 Key Details

  • Manufacturer: ABB Robotics
  • Model: IRB 14000 YuMi
  • Robot Type: Dual-arm collaborative robot (COBOT) designed for safe human-robot interaction
  • Degrees of Freedom: 7 DOF per arm (14 axes total)
  • Payload: ~0.5 kg per arm
  • Reach: ~559 mm
  • Repeatability: ~±0.02 mm
  • Safety: Intrinsically collaborative design, rounded geometry, limited force operation
  • Controller: IRC5C compact robotic controller (supports MultiMove & dual-arm coordination)
  • Programming Interfaces:
    • RAPID (ABB native robot language)
    • ROS support via kth-ros-pkg/yumi
    • Python control via Yumipy (galou/robotware6_06 fork)
  • Network Interface: Ethernet (Service Port XP23) for control & communication
  • Operating Modes:
    • Manual Mode (slow/jog/teaching)
    • Auto Mode (full-speed program execution, ~10× faster)
  • Ideal Use Cases: Precision assembly, manipulation research, vision-guided robotics

🛠️ Why It’s a Lab Favorite

  • All-Terrain Robustness: Handles rocky ground, mud, and slopes up to 30° with ease
  • High Payload: Supports heavy sensors and manipulator arms for custom experiments
  • ROS-Ready: Includes tutorials for mapping, SLAM, AMCL, Gmapping, and more
  • Full 3D Perception: VLP‑16 enables 360° point clouds—great for obstacle avoidance, mapping, and visual odometry

Useful Official Documentation

NameLink
YuMi Product Pagehttps://new.abb.com/products/robotics/industrial-robots/irb-14000-yumi
ABB Technical Libraryhttps://library.abb.com/
RAPID Language Handbookhttps://library.abb.com/

Recent Research Using YuMi

TitleYearLink
Redundancy Parameterization of the YuMi Arm2025https://arxiv.org/abs/2505.23111
Real-Time Failure Handling w/ Vision-Language Models2025https://arxiv.org/abs/2503.15202
Digital Twin for Human-Robot Interaction2025https://www.researchgate.net/publication/381909584