扩散策略：通过动作扩散学习视觉运动策略
Diffusion Policy: Visuomotor Policy Learning via Action Diffusion
将机器人控制建模为条件动作序列去噪，把扩散模型连接到视觉运动学习。
THE RESEARCH ATLAS
π0：面向通用机器人控制的视觉—语言—动作流模型
将预训练视觉语言主干与基于流的动作生成结合，用于多种机器人任务。
π0: A Vision-Language-Action Flow Model for General Robot Control
为什么值得读
比较连续动作生成与离散动作 Token 预测的差异。
引用
@misc{black2024vision,
title = {π0: A Vision-Language-Action Flow Model for General Robot Control},
author = {Kevin Black and Noah Brown and Danny Driess and Adnan Esmail and Michael Equi and Chelsea Finn and Niccolo Fusai and Lachy Groom and Karol Hausman and Brian Ichter and Szymon Jakubczak and Tim Jones and Liyiming Ke and Sergey Levine and Adrian Li-Bell and Mohith Mothukuri and Suraj Nair and Karl Pertsch and Lucy Xiaoyang Shi and James Tanner and Quan Vuong and Anna Walling and Haohuan Wang and Ury Zhilinsky},
year = {2024},
eprint = {2410.24164},
archivePrefix = {arXiv},
primaryClass = {cs.LG},
url = {https://arxiv.org/abs/2410.24164}
}论文与代码版权归原作者；引用及许可请以原始来源为准。
相关研究
RT-2：视觉—语言—动作模型将网络知识迁移至机器人控制
RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control
让视觉语言模型输出机器人动作，探索网络规模学习向机器人控制的迁移。
OpenVLA：开源视觉—语言—动作模型
OpenVLA: An Open-Source Vision-Language-Action Model
开源视觉—语言—动作模型及适配框架，为通用机器人策略研究提供基础。