AsymEEF: Bimanual Manipulation with Asymmetric End Effectors
Organizations: Georgia Institute of Technology · Tsinghua University · University of California San Diego
Abstract
Teleoperated demonstrations are costly to collect and rarely reusable across end-effectors with different kinematics and action spaces. This burden grows in bimanual manipulation, where each end-effector pair may require new data. We present AsymEEF, a framework for executable cross-end-effector demonstration generation and unified bimanual policy learning. Given a small source dataset, AsymEEF transfers demonstrations to heterogeneous combinations of two-, three-, and five-fingered end-effectors without additional teleoperation. Kinematic teacher anchors provide geometric supervision, while sparse real-world X-Calib anchors correct discrepancies in grasp execution. A shared encoder maps wrist poses and actuator states to a morphology-agnostic latent space, and a morphology-conditioned decoder reconstructs target commands. Generated trajectories undergo replay filtering for task-level executability. The representation also supports a unified bimanual visuomotor policy trained across multiple pairs. Across four simulated and six real-world tasks, AsymEEF improves replay success over adapted transfer baselines, achieves a mean replay-valid expansion gain of 16.6 times in simulation, and is evaluated through downstream policy training on real robots. Project website: https://asymeef.vercel.app/.
Figures & tables
| Variant | Avg. normalized error | Native MAE | Cross MAE | Real Cross MAE |
|---|---|---|---|---|
| Full | 4.185 | 0.0967 | 0.3069 | 0.2088 |
| w/o Self | 8.325 | 0.5617 | 0.2504 | 0.1766 |
| w/o Cross | 6.858 | 0.0973 | 0.5813 | 0.3345 |
| w/o X-Calib | 5.697 | 0.0992 | 0.2937 | 0.6181 |
Bimanual Robotic Manipulation
Bimanual robotic manipulation coordinates two robot arms to handle objects, tools, and tasks that require collaboration.
See all 213 papers