Octo: An open-source generalist robot policy

OM Team, D Ghosh, H Walke, K Pertsch… - arxiv preprint arxiv …, 2024‏ - arxiv.org
Large policies pretrained on diverse robot datasets have the potential to transform robotic
learning: instead of training new policies from scratch, such generalist robot policies may be …

Open X-Embodiment: Robotic Learning Datasets and RT-X Models : Open X-Embodiment Collaboration0

A O'Neill, A Rehman, A Maddukuri… - … on Robotics and …, 2024‏ - ieeexplore.ieee.org
Large, high-capacity models trained on diverse datasets have shown remarkable successes
on efficiently tackling downstream applications. In domains from NLP to Computer Vision …

[HTML][HTML] Robotic disassembly for end-of-life products focusing on task and motion planning: A comprehensive survey

ME Asif, A Rastegarpanah, R Stolkin - Journal of Manufacturing Systems, 2024‏ - Elsevier
The rise of mass production and the resulting accumulation of end-of-life (EoL) products
present a growing challenge in waste management and highlight the need for efficient …

Lotus: Continual imitation learning for robot manipulation through unsupervised skill discovery

W Wan, Y Zhu, R Shah, Y Zhu - 2024 IEEE International …, 2024‏ - ieeexplore.ieee.org
We introduce LOTUS, a continual imitation learning algorithm that empowers a physical
robot to continuously and efficiently learn to solve new manipulation tasks throughout its …

Octo: An open-source generalist robot policy

O Mees, D Ghosh, K Pertsch, K Black… - First Workshop on …, 2024‏ - openreview.net
Large policies pretrained on diverse robot datasets have the potential to transform robotic
learning: instead of training new policies from scratch, such generalist robot policies may be …

Prime: Scaffolding manipulation tasks with behavior primitives for data-efficient imitation learning

T Gao, S Nasiriany, H Liu, Q Yang… - IEEE Robotics and …, 2024‏ - ieeexplore.ieee.org
Imitation learning has shown great potential for enabling robots to acquire complex
manipulation behaviors. However, these algorithms suffer from high sample complexity in …

Igor: Image-goal representations are the atomic control units for foundation models in embodied ai

X Chen, J Guo, T He, C Zhang, P Zhang… - arxiv preprint arxiv …, 2024‏ - arxiv.org
We introduce Image-GOal Representations (IGOR), aiming to learn a unified, semantically
consistent action space across human and various robots. Through this unified latent action …

Dexskills: Skill segmentation using haptic data for learning autonomous long-horizon robotic manipulation tasks

X Mao, G Giudici, C Coppola… - 2024 IEEE/RSJ …, 2024‏ - ieeexplore.ieee.org
Effective execution of long-horizon tasks with dexterous robotic hands remains a significant
challenge in real-world problems. While learning from human demonstrations has shown …

Precise and Dexterous Robotic Manipulation via Human-in-the-Loop Reinforcement Learning

J Luo, C Xu, J Wu, S Levine - arxiv preprint arxiv:2410.21845, 2024‏ - arxiv.org
Reinforcement learning (RL) holds great promise for enabling autonomous acquisition of
complex robotic manipulation skills, but realizing this potential in real-world settings has …

Cloud-based hierarchical imitation learning for scalable transfer of construction skills from human workers to assisting robots

H Yu, VR Kamat, CC Menassa - Journal of Computing in Civil …, 2024‏ - ascelibrary.org
Assigning repetitive and physically demanding construction tasks to robots can alleviate
human workers' exposure to occupational injuries, which often result in significant downtime …