A naukri.com initiative
The Robot Report
1M
68
Image Credit: The Robot Report
NVIDIA adds open AI and simulation tools for robot learning, humanoid development
- NVIDIA has announced new AI and simulation tools aimed at accelerating the development of robot humanoid, including the NVIDIA Isaac Lab robot learning framework and six new robot learning workflows for the Project GR00T initiative to accelerate humanoid development.
- Hugging Face and NVIDIA are collaborating to accelerate open-source robotics research with LeRobot, NVIDIA Isaac Lab, and NVIDIA Jetson.
- The Cosmos tokenizer is designed to minimize distortion, running up to 12 times faster than current tokenizers.
- Curating video data poses challenges, with the massive size requiring scalable pipelines and efficient orchestration for load balancing across GPUs.
- NeMo Curator streamlines data curation by reducing video processing time and enables robot developers to improve their world-model accuracy.
- Hugging Face's AI platform, which includes APIs with more than 1.5 million models, datasets, and applications now works with NVIDIA Isaac Lab on Isaac Sim, enabling training by demonstration or trial and error in realistic simulation.
- Initial steps in the collaboration have shown a physical-picking set up with LeRobot software running on NVIDIA Jetson Orin Nano, providing a compact compute platform for deployment.
- NVIDIA has added six new workflows intended to help robots perceive, move, and interact with people and their environments as part of the Project GR00T blueprints.
- NVIDIA Cosmos tokenizer provides high-quality encoding and decoding to simplify the development of world models with minimal distortion and temporal instability.
- NVIDIA NeMo Curator reduces video processing time and enables robot developers to improve their world-model accuracy by processing large-scale text, image, and video data.
Read Full Article
4 Likes
For uninterrupted reading, download the app