Recoleta
GitHub
趋势

机器人基础模型正在在迁移、规划和接触上变得更落地

日 · 2026-04-22 · Embodied AI

RoboticsVision Language ActionWorld ModelsCross EmbodimentTactile SensingMedical Robotics
Markdown 源文件

概览

这一天最强的主线是：机器人论文正在把广泛预训练、显式规划和接触反馈，和具体的执行指标连接起来。JoyAI-RA、Cortex 2.0 和 Open-H-Embodiment 构成了核心。它们在跨本体迁移、长时程规划和医疗机器人上都给出了提升，而且结果已经落到真实机器人、工业部署，或两者兼有。

研究发现

Cross-embodiment training gets real benchmark wins

视觉-语言-动作工作在数据上覆盖更广，也更明确说明动作如何在不同本体之间迁移。JoyAI-RA 用网页数据、第一人称人类视频、仿真和机器人轨迹训练一个策略，然后把动作映射到共享表示中。报告的提升在仿真和真实人形基准上都很大：RoboTwin Easy 和 Hard 分别是 90.48% 和 89.28%，RoboCasa GR1 Tabletop 是 63.2%，AgiBot 真实世界基准的平均成功率是 0.74，π0.5 是 0.62。Open-H-Embodiment 在医疗机器人里用同样的规模论证，数据量是 770 小时、20 个平台，GR00T-H 策略在端到端缝合上达到 25% 成功率，而 ACT、GR00T-N1.6 和 LingBot-VA 都是 20 次试验里 0 次成功。当天最强的证据表明，跨本体学习正在变成明确的训练目标，而不只是对泛化能力的主张。

资料来源

  • JoyAI-RA 0.1: A Foundation Model for Robotic Autonomy
    Tianle Zhang, Zhihao Yuan, Dafeng Chi, Peidong Liu, Dongwei Li, Kejun Hu, Likui Zhang, Junnan Nie, Ziming Wei, Zengjue Chen, Yili Tang, Jiayi Li, Zhiyuan Xiang, Mingyang Li, Tianci Luo, Hanwen Wan, Ao Li, Linbo Zhai, Zhihao Zhan, Xiaodong Bai, Jiakun Cai, Peng Cao, Kangliang Chen, Siang Chen, Yixiang Dai, Shuai Di, Yicheng Gong, Chenguang Gui, Yucheng Guo, Peng Hao, Qingrong He, Haoyang Huang, Kunrui Huang, Zhixuan Huang, Shibo Jin, Yixiang Jin, Anson Li, Dongjiang Li, Jiawei Li, Ruodai Li, Yihang Li, Yuzhen Li, Jiaming Liang, Fangsheng Liu, Jing Long, Mingxi Luo, Xing Pan, Hui Shen, Xiaomeng Tian, Daming Wang, Song Wang, Junwu Xiong, Hang Xu, Wanting Xu, Zhengcheng Yu, He Zhang, Jiyao Zhang, Lin Zhao, Chen Zhou, Nan Duan, Yuzheng Zhuang, Liang Lin
  • Open-H-Embodiment: A Large-Scale Dataset for Enabling Foundation Models in Medical Robotics
    Open-H-Embodiment Consortium, :, Nigel Nelson, Juo-Tung Chen, Jesse Haworth, Xinhao Chen, Lukas Zbinden, Dianye Huang, Alaa Eldin Abdelaal, Alberto Arezzo, Ayberk Acar, Farshid Alambeigi, Carlo Alberto Ammirati, Yunke Ao, Pablo David Aranda Rodriguez, Soofiyan Atar, Mattia Ballo, Noah Barnes, Federica Barontini, Filip Binkiewicz, Peter Black, Sebastian Bodenstedt, Leonardo Borgioli, Nikola Budjak, Benjamin Calmé, Fabio Carrillo, Nicola Cavalcanti, Changwei Chen, Haoxin Chen, Sihang Chen, Qihan Chen, Zhongyu Chen, Ziyang Chen, Shing Shin Cheng, Meiqing Cheng, Min Cheng, Zih-Yun Sarah Chiu, Xiangyu Chu, Camilo Correa-Gallego, Giulio Dagnino, Anton Deguet, Jacob Delgado, Jonathan C. DeLong, Kaizhong Deng, Alexander Dimitrakakis, Qingpeng Ding, Hao Ding, Giovanni Distefano, Daniel Donoho, Anqing Duan, Marco Esposito, Shane Farritor, Jad Fayad, Zahi Fayad, Mario Ferradosa, Filippo Filicori, Chelsea Finn, Philipp Fürnstahl, Jiawei Ge, Stamatia Giannarou, Xavier Giralt Ludevid, Frederic Giraud, Aditya Amit Godbole, Ken Goldberg, Antony Goldenberg, Diego Granero Marana, Xiaoqing Guo, Tamás Haidegger, Evan Hailey, Pascal Hansen, Ziyi Hao, Kush Hari, Kengo Hayashi, Jonathon Hawkins, Shelby Haworth, Ortrun Hellig, S. Duke Herrell, Zhouyang Hong, Andrew Howe, Junlei Hu, Ria Jain, Mohammad Rafiee Javazm, Howard Ji, Rui Ji, Jianmin Ji, Zhongliang Jiang, Dominic Jones, Jeffrey Jopling, Britton Jordan, Ran Ju, Michael Kam, Luoyao Kang, Fausto Kang, Siddhartha Kapuria, Peter Kazanzides, Sonika Kiehler, Ethan Kilmer, Ji Woong, Kim, Przemysław Korzeniowski, Chandra Kuchi, Nithesh Kumar, Alan Kuntz, Federico Lavagno, Yu Chung Lee, Hao-Chih Lee, Hang Li, Zhen Li, Xiao Liang, Xinxin Lin, Jinsong Lin, Chang Liu, Fei Liu, Pei Liu, Yun-hui Liu, Wanli Liuchen, Eszter Lukács, Sareena Mann, Miles Mannas, Brett Marinelli, Sabina Martyniak, Francesco Marzola, Lorenzo Mazza, Xueyan Mei, Maria Clara Morais, Luigi Muratore, Chetan Reddy Narayanaswamy, Michał Naskręt, David Navarro-Alarcon, Cyrus Neary, Chi Kit Ng, Christopher Nguan, David Noonan, Ki Hwan Oh, Tom Christian Olesch, Allison M. Okamura, Justin Opfermann, Matteo Pescio, Doan Xuan Viet Pham, Tito Porras, Hongliang Ren, Ariel Rodriguez Jimenez, Ferdinando Rodriguez y Baena, Septimiu E. Salcudean, Asmitha Sathya, Preethi Satish, Lalithkumar Seenivasan, Jiaqi Shao, Yiqing Shen, Yu Sheng, Lucy XiaoYang Shi, Zoe Soulé, Stefanie Speidel, Mingwu Su, Jianhao Su, Idris Sunmola, Kristóf Takács, Yunxi Tang, Patrick Thornycroft, Yu Tian, Jordan Thompson, Mehmet K. Turkcan, Mathias Unberath, Pietro Valdastri, Carlos Vives, Quan Vuong, Martin Wagner, Farong Wang, Wei Wang, Lidian Wang, Chung-Pang Wang, Guankun Wang, Junyi Wang, Erqi Wang, Ziyi Wang, Tanner Watts, Wolfgang Wein, Yimeng Wu, Zijian Wu, Hongjun Wu, Luohong Wu, Jie Ying Wu, Junlin Wu, Victoria Wu, Kaixuan Wu, Mateusz Wójcikowski, Yunye Xiao, Nan Xiao, Wenxuan Xie, Hao Yang, Tianqi Yang, Yinuo Yang, Menglong Ye, Ryan S. Yeung, Nural Yilmaz, Chim Ho Yin, Michael Yip, Rayan Younis, Chenhao Yu, Sayem Nazmuz Zaman, Milos Zefran, Han Zhang, Yuelin Zhang, Yidong Zhang, Yanyong Zhang, Xuyang Zhang, Yameng Zhang, Joyce Zhang, Ning Zhong, Peng Zhou, Haoying Zhou, Xiuli Zuo, Nassir Navab, Mahdi Azizian, Sean D. Huver, Axel Krieger

Planning over futures moves into deployed robot systems

世界模型越来越多地出现在长时程里错误会不断累积的场景中。Cortex 2.0 给工业 VLA 栈加入潜在轨迹展开规划，在动作执行前对进展、风险和完成情况打分。摘要称它在四个已部署仓储任务上拿到最好的成功率，评估时没有人工介入，不过摘录里没有给出逐任务的具体数字。在更严格的安全场景里，面向血管取栓的 TD-MPC2 世界模型智能体在留出的仿真解剖上平均成功率达到 58%，SAC 是 36%，路径比率从 22% 提高到 49%，最大导丝尖端力保持在 0.55 N，远低于论文给出的 1.5 N 破裂阈值。医疗机器人也通过 Open-H 的动作条件手术模拟器加入了这条路线，这个模拟器是在九个平台上训练的。

资料来源

  • Cortex 2.0: Grounding World Models in Real-World Industrial Deployment
    Adriana Aida, Walid Amer, Katarina Bankovic, Dhruv Behl, Fabian Busch, Annie Bhalla, Minh Duong, Florian Gienger, Rohan Godse, Denis Grachev, Ralf Gulde, Elisa Hagensieker, Junpeng Hu, Shivam Joshi, Tobias Knobloch, Likith Kumar, Damien LaRocque, Keerthana Lokesh, Omar Moured, Khiem Nguyen, Christian Preyss, Ranjith Sriganesan, Vikram Singh, Carsten Sponner, Anh Tong, Dominik Tuscher, Marc Tuscher, Pavan Upputuri
  • Toward Safe Autonomous Robotic Endovascular Interventions using World Models
    Harry Robertshaw, Nikola Fischer, Han-Ru Wu, Andrea Walker Perez, Weiyuan Deng, Benjamin Jackson, Christos Bergeles, Alejandro Granados, Thomas C Booth
  • Open-H-Embodiment: A Large-Scale Dataset for Enabling Foundation Models in Medical Robotics
    Open-H-Embodiment Consortium, :, Nigel Nelson, Juo-Tung Chen, Jesse Haworth, Xinhao Chen, Lukas Zbinden, Dianye Huang, Alaa Eldin Abdelaal, Alberto Arezzo, Ayberk Acar, Farshid Alambeigi, Carlo Alberto Ammirati, Yunke Ao, Pablo David Aranda Rodriguez, Soofiyan Atar, Mattia Ballo, Noah Barnes, Federica Barontini, Filip Binkiewicz, Peter Black, Sebastian Bodenstedt, Leonardo Borgioli, Nikola Budjak, Benjamin Calmé, Fabio Carrillo, Nicola Cavalcanti, Changwei Chen, Haoxin Chen, Sihang Chen, Qihan Chen, Zhongyu Chen, Ziyang Chen, Shing Shin Cheng, Meiqing Cheng, Min Cheng, Zih-Yun Sarah Chiu, Xiangyu Chu, Camilo Correa-Gallego, Giulio Dagnino, Anton Deguet, Jacob Delgado, Jonathan C. DeLong, Kaizhong Deng, Alexander Dimitrakakis, Qingpeng Ding, Hao Ding, Giovanni Distefano, Daniel Donoho, Anqing Duan, Marco Esposito, Shane Farritor, Jad Fayad, Zahi Fayad, Mario Ferradosa, Filippo Filicori, Chelsea Finn, Philipp Fürnstahl, Jiawei Ge, Stamatia Giannarou, Xavier Giralt Ludevid, Frederic Giraud, Aditya Amit Godbole, Ken Goldberg, Antony Goldenberg, Diego Granero Marana, Xiaoqing Guo, Tamás Haidegger, Evan Hailey, Pascal Hansen, Ziyi Hao, Kush Hari, Kengo Hayashi, Jonathon Hawkins, Shelby Haworth, Ortrun Hellig, S. Duke Herrell, Zhouyang Hong, Andrew Howe, Junlei Hu, Ria Jain, Mohammad Rafiee Javazm, Howard Ji, Rui Ji, Jianmin Ji, Zhongliang Jiang, Dominic Jones, Jeffrey Jopling, Britton Jordan, Ran Ju, Michael Kam, Luoyao Kang, Fausto Kang, Siddhartha Kapuria, Peter Kazanzides, Sonika Kiehler, Ethan Kilmer, Ji Woong, Kim, Przemysław Korzeniowski, Chandra Kuchi, Nithesh Kumar, Alan Kuntz, Federico Lavagno, Yu Chung Lee, Hao-Chih Lee, Hang Li, Zhen Li, Xiao Liang, Xinxin Lin, Jinsong Lin, Chang Liu, Fei Liu, Pei Liu, Yun-hui Liu, Wanli Liuchen, Eszter Lukács, Sareena Mann, Miles Mannas, Brett Marinelli, Sabina Martyniak, Francesco Marzola, Lorenzo Mazza, Xueyan Mei, Maria Clara Morais, Luigi Muratore, Chetan Reddy Narayanaswamy, Michał Naskręt, David Navarro-Alarcon, Cyrus Neary, Chi Kit Ng, Christopher Nguan, David Noonan, Ki Hwan Oh, Tom Christian Olesch, Allison M. Okamura, Justin Opfermann, Matteo Pescio, Doan Xuan Viet Pham, Tito Porras, Hongliang Ren, Ariel Rodriguez Jimenez, Ferdinando Rodriguez y Baena, Septimiu E. Salcudean, Asmitha Sathya, Preethi Satish, Lalithkumar Seenivasan, Jiaqi Shao, Yiqing Shen, Yu Sheng, Lucy XiaoYang Shi, Zoe Soulé, Stefanie Speidel, Mingwu Su, Jianhao Su, Idris Sunmola, Kristóf Takács, Yunxi Tang, Patrick Thornycroft, Yu Tian, Jordan Thompson, Mehmet K. Turkcan, Mathias Unberath, Pietro Valdastri, Carlos Vives, Quan Vuong, Martin Wagner, Farong Wang, Wei Wang, Lidian Wang, Chung-Pang Wang, Guankun Wang, Junyi Wang, Erqi Wang, Ziyi Wang, Tanner Watts, Wolfgang Wein, Yimeng Wu, Zijian Wu, Hongjun Wu, Luohong Wu, Jie Ying Wu, Junlin Wu, Victoria Wu, Kaixuan Wu, Mateusz Wójcikowski, Yunye Xiao, Nan Xiao, Wenxuan Xie, Hao Yang, Tianqi Yang, Yinuo Yang, Menglong Ye, Ryan S. Yeung, Nural Yilmaz, Chim Ho Yin, Michael Yip, Rayan Younis, Chenhao Yu, Sayem Nazmuz Zaman, Milos Zefran, Han Zhang, Yuelin Zhang, Yidong Zhang, Yanyong Zhang, Xuyang Zhang, Yameng Zhang, Joyce Zhang, Ning Zhong, Peng Zhou, Haoying Zhou, Xiuli Zuo, Nassir Navab, Mahdi Azizian, Sean D. Huver, Axel Krieger

Compact VLA systems add task structure and confidence signals

更小的模型在动作学习前获得了更多结构。PokeVLA 围绕具身预训练、来自基座和腕部视角的目标分割，以及训练中的几何对齐构建了一个紧凑 VLA。报告中的提升之所以重要，是因为它们在迁移和扰动下都能保持：在 LIBERO-Plus 变体上比 OpenVLA-OFT 高 9.7%，比 VLA-Adapter 高 20.2%，在真实世界扰动下还有 20.0% 的提升。另一篇论文处理的是另一个薄弱点：置信度。Temporal Difference Calibration 把整个回合的成功预测定义出来，用时序差分目标训练，并在 OpenVLA、π0、π0-FAST 和 UniVLA 上报告了更好的校准和更早的失败检测。它还在 LIBERO 上为 OpenVLA 带来 15% 的成功率提升，条件是用学到的价值预测器给采样动作排序。共同点是，轻量或黑盒 VLA 系统现在拿到的是围绕感知和可靠性的任务特定监督，而不只是更大的主干网络。

资料来源

Tactile work gets cheaper to build and faster to train

接触丰富的机器人工作正在从两个方向同时提升感知质量：更好的硬件和更快的仿真。FingerEye 用一个很小的指尖模块，把接触前、接触开始和接触后都保持在一个视觉流里，材料成本大约 60 美元。论文报告了六轴力矩传感灵敏度，并把这个信号用于精细抓取，包括在脆弱物体上做接触感知停止。ETac 从训练侧处理同一个问题。它用快速传播模型加一个小的学习残差来近似 FEM 质量的触觉形变，达到 4,096 个并行环境和单张 RTX 4090 上 869 FPS，并训练出在四类物体上平均成功率为 84.45% 的盲抓取策略。当天的触觉论文给出的结论很直接：更丰富的接触反馈正在变得更容易搭建，也更容易在大规模上训练。

资料来源

较新编码代理研究正在按保留代码、控制面和可执行证明来评判较早代码研究正在围绕生成内容收紧行为检查