Copied from :- Meta AI
New robotics research from Meta AI and Carnegie Mellon University Robotics — RoboAgent can acquire a wide diversity of non-trivial skills and generalize them to hundreds of diverse unseen scenarios — all with an order of magnitude less data than prior works in this space.
Additionally we’re open-sourcing RoboSet, a large, high-quality robotics dataset collected with commodity hardware, to support and accelerate open source research in robot learning.
More details ➡️
Get the dataset ➡️ https://bit.ly/3DZk8vI
RoboAgent can exhibit a diverse set of 12 non-trivial manipulation skills across 38 tasks and can generalize them to hundreds of diverse unseen scenarios — beyond that, it can also evolve its capabilities with new experiences.
RoboAgent.
Towards Sample Efficient Robot Manipulation with Semantic Augmentations and Action Chunking.
Leave a reply