Cross-embodiment manipulation paper contributes a multi-modal Unitree G1 dataset
Evidence notes
- The paper studies modality-augmented fine-tuning of foundation robot policies across GR1 and Unitree G1 embodiments.
- For G1, the authors contribute a new multi-modal dataset with cuRobo motion planning, inverse kinematics, and ground-truth contact-force measurements, then report a 94 percent success rate on the Pick Apple to Bowl task for the contact-augmented model.
Company context
Develops quadruped and humanoid robots, with strength in dynamic locomotion, vertically integrated hardware, and relatively low-cost commercial deployment. Unitree is one of the clearest examples of a legged robotics company moving from research visibility into real productisation and broader market distribution.