Event
Amazon demonstrates ResMimic training for humanoid box-carry tasks
Evidence notes
- Amazon FAR and academic collaborators introduced ResMimic, a residual-learning framework for humanoid whole-body loco-manipulation.
- A Unitree G1 carried a 4.5 kg box and handled irregular objects using whole-body contact.
- The work was published as arXiv paper 2510.05070 and demonstrated on physical hardware.
Company context
Develops physical AI systems for dexterous robotic manipulation, combining learned control models with robotic hands and fixed robot stations. Systems are designed to learn from demonstrations and execute tasks in industrial environments.
Context
- Company
- mimic
- Segment
- Industrial
- Event type
- Research Publication
- Added to Korthos
- Feb 14, 2026
- Geography
- Zürich · Switzerland