Training general robotic policies from heterogeneous data for different tasks is a significant challenge. Existing robotic datasets vary in different modalities such as color, depth, tactile …
The success of deep learning depends heavily on the availability of large datasets, but in robotic manipulation there are many learning problems for which such datasets do not exist …
Data-driven robotic manipulation has been gaining traction. However, creating synthetic large-scale datasets for training, validation and benchmarks often relies on random …
Z Liu, S Tian, M Guo, K Liu, J Wu - 7th Annual Conference on Robot …, 2023 - openreview.net
When limited by their own morphologies, humans and some species of animals have the remarkable ability to use objects from the environment toward accomplishing otherwise …
JA Caley, GA Hollinger - 2020 IEEE International Conference …, 2020 - ieeexplore.ieee.org
Robots often require a model of their environment to make informed decisions. In unknown environments, the ability to infer the value of a data field from a limited number of samples is …
Model-free control strategies such as reinforcement learning have shown the ability to learn control strategies without requiring an accurate model or simulator of the world. While this is …
Large-scale training have propelled significant progress in various sub-fields of AI such as computer vision and natural language processing. However, building robot learning systems …
A Khazatsky, K Pertsch, S Nair, A Balakrishna… - arXiv preprint arXiv …, 2024 - arxiv.org
The creation of large, diverse, high-quality robot manipulation datasets is an important stepping stone on the path toward more capable and robust robotic manipulation policies …
T Zhang, Y Hu, J You, Y Gao - arXiv preprint arXiv:2406.10615, 2024 - arxiv.org
Given the high cost of collecting robotic data in the real world, sample efficiency is a consistently compelling pursuit in robotics. In this paper, we introduce SGRv2, an imitation …