Long-horizon precision manipulation in laboratory automation, such as pipette tip attachment and liquid transfer, requires policies that respect strict procedural logic while operating in continuous, high-dimensional state spaces. However, existing approaches struggle with reward sparsity, multi-stage structural constraints, and noisy or imperfect demonstrations, leading to inefficient exploration and unstable convergence. We propose a Keyframe-Guided Reward Generation Framework that automatically extracts kinematics-aware keyframes from demonstrations, generates stage-wise targets via a diffusion-based predictor in latent space, and constructs a geometric progress-based reward to guide online reinforcement learning. The framework integrates multi-view visual encoding, latent similarity-based progress tracking, and human-in-the-loop reinforcement fine-tuning on a Vision-Language-Action backbone to align policy optimization with the intrinsic stepwise logic of biological protocols. Across four real-world laboratory tasks, including high-precision pipette attachment and dynamic liquid transfer, our method achieves an average success rate of 82% after 40--60 minutes of online fine-tuning. Compared with HG-DAgger (42%) and Hil-ConRFT (47%), our approach demonstrates the effectiveness of structured keyframe-guided rewards in overcoming exploration bottlenecks and providing a scalable solution for high-precision, long-horizon robotic laboratory automation.
RoboMemArena: A Comprehensive and Challenging Robotic Memory Benchmark
The Hong Kong University of Science and Technology(香港科学与技术大学);Zhejiang University(浙江大学);Westlake University(西湖大学);Tsinghua University(清华大学);Zhejiang University of Technology(浙江工业大学);Shanghai Jiao Tong University(上海交通大学)
Improving Vision-Language-Action Model Fine-Tuning with Structured Stage and Keyframe Supervision
School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院);New Laboratory of Pattern Recognition (NLPR), State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS), Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所模式识别国家重点实验室(NLPR)多模态人工智能系统国家重点实验室(MAIS));FiveAges
TGM-VLA: Task-Guided Mixup for Sampling-Efficient and Robust Robotic Manipulation
Shenzhen International Graduate School, Tsinghua University, Shenzhen, China(清华大学深圳国际研究生院);The National and Local Co-Build Humanoid Robotics Innovation Center(国家级与地方共建人形机器人创新中心)
BioProVLA-Agent: An Affordable, Protocol-Driven, Vision-Enhanced VLA-Enabled Embodied Multi-Agent System with Closed-Loop-Capable Reasoning for Biological Laboratory Manipulation
Key Laboratory of Smart Manufacturing in Energy Chemical Process Ministry of Education, East China University of Science and Technology, Shanghai, CN(能源化工过程智能制造教育部重点实验室,东华大学,上海,中国);Department of Computer Science and Engineering, East China University of Science and Technology, Shanghai, CN(东华大学计算机科学与工程系,上海,中国);Department of Laboratory Medicine, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, CN(复旦大学附属瑞金医院检验医学科,上海,中国);School of Information Science and Technology, Shihezi University, Shihezi, CN(石河子大学信息科学与技术学院,石河子,中国)
MATTERIX: toward a digital twin for robotics-assisted chemistry laboratory automation
University of Toronto(多伦多大学);Acceleration Consortium(加速联盟);Vector Institute(向量研究所);University of Liverpool(利物浦大学);University of Salento(萨勒诺大学);NVIDIA Canadian Institute for Advanced Research(NVIDIA加拿大高级研究机构);Georgia Institute of Technology(佐治亚理工学院)
WetRobo: A Reproducible Robot Kit for Coding Agents in Biological Laboratories
The University of Tokyo(东京大学);RIKEN Pioneering Research Institute(日本理化学研究所先驱研究研究所);New York University(纽约大学);Google DeepMind(谷歌DeepMind);National Institute for Materials Science(日本国立材料科学研究所);RIKEN Center for Advanced Intelligence Project(日本理化学研究所先进智能项目中心)
Towards Robotic Laboratory Automation Plug & Play: Survey and Concept Proposal on Teaching-free Robot Integration with the LAPP Digital Twin
Takeda Manufacturing Austria AG(武田奥地利制造公司);Doctoral School of Applied Informatics and Applied Mathematics, Óbuda University(欧布达大学应用信息学与应用数学博士学院);Óbuda University(欧布达大学);Baxalta Innovations GmbH, a Takeda company(百特创新有限公司(武田旗下公司));Alba Regia Technical Faculty, Óbuda University(欧布达大学阿尔巴雷吉亚技术学院);Antal Bejczy Center for Intelligent Robotics, Óbuda University(欧布达大学安塔尔·贝伊奇智能机器人中心)
Accelerating Discovery in Natural Science Laboratories with AI and Robotics: Perspectives and Challenges from the 2024 IEEE ICRA Workshop, Yokohama, Japan
University of Liverpool(利物浦大学);SiLA Consortium(SiLA联盟);Acceleration Consortium(加速联盟);University of Toronto(多伦多大学);Technical University of Munich(慕尼黑工业大学);University of Tokyo(东京大学);University of British Columbia(不列颠哥伦比亚大学);Roche Diagnostics GmbH(罗氏诊断有限公司);Unchained Labs(Unchained实验室);University of Rostock(罗斯托克大学);Argonne National Laboratory(阿贡国家实验室);Takeda(武田药品工业);Óbuda University(欧布达大学);Institute of Science Tokyo(东京科学大学);Florida State University(佛罗里达州立大学)
The First WARA Robotics Mobile Manipulation Challenge -- Lessons Learned
Center for Applied Autonomous Sensor Systems (AASS)(应用自主传感系统中心);Politecnico di Milano(米兰理工学院);Division of Robotics, Perception and Learning, Royal Institute of Technology (KTH)(机器人、感知与学习 division,皇家理工学院);Lund Technical University (LTH)(卢恩技术大学);University of Basilicata(巴塞拉塔大学);Pharmaceutical Technology & Development, R&D, AstraZeneca(制药技术与开发,研发,阿斯利康);ABB Corporate Research(ABB企业研究)
本文总结首届 WARA Robotics 移动操作挑战赛的设置、产业动机、四类参赛方案与经验教训,并提出 2025 年第二届赛事的改进方向,以促进真实实验室中的自主移动操作机器人研发和产学协作。
Autonomous Elemental Characterization Enabled by a Low Cost Robotic Platform Built Upon a Generalized Software Architecture
Energy Geosciences Division, Lawrence Berkeley National Laboratory(能源地球科学部,劳伦斯伯克利国家实验室);Materials Science Division, Lawrence Berkeley National Laboratory(材料科学部,劳伦斯伯克利国家实验室)
Admittance-Based Motion Planning with Vision-Guided Initialization for Robotic Manipulators in Self-Driving Laboratories
Department of Electronics Systems, Aalborg University(电子系统系,奥尔堡大学);Department of Information Technology and Electrical Engineering, Università degli Studi di Napoli Federico II(信息科技与电气工程系,那不勒斯费迪里奇二世大学)
Compact Task-Aligned Imitation Learning for Laboratory Automation
Spatial Robotics Research Center, Fujitsu Limited.(富士通株式会社空间机器人研究中心);Faculty of Science and Engineering, Waseda University(早稻田大学理工学部);National Institute of Advanced Industrial Science and Technology(国家工业科学与技术研究院)
Exploring environment exploitation for self-reconfiguration in modular robotics
Department of Applied Mathematics University of Washington(应用数学系 华盛顿大学);Department of Mechanical Engineering and Applied Mechanics University of Pennsylvania(机械工程与应用力学系 费城大学);Department of Mechanical Engineering University of Nevada, Las Vegas(机械工程系 内华达大学拉斯维加斯分校)
Beyond Pixels: Learning Invariant Rewards for Real-World Robotics From a Few Demonstrations
School of Computing and Data Science, The University of Hong Kong(计算科学与数据科学学院,香港大学);LimX Dynamics Technology Co., Ltd(LimX动力技术有限公司);Southern University of Science and Technology(南方科技大学);Peking University(北京大学);Zhejiang University(浙江大学)
Perspective on Utilizing Foundation Models for Laboratory Automation in Materials Research
The University of Tokyo(东京大学);RIKEN Center for Biosystems Dynamics Research(理化学研究所生物系统动力学研究中心);Intellectual Moonshine;OMRON SINIC X Corporation(欧姆龙 SINIC X 公司);NexaScience;RIKEN Advanced General Intelligence for Science Program(理化学研究所科学高级通用智能项目);Institute of Science Tokyo(东京科学大学)
Robotic System for Chemical Experiment Automation with Dual Demonstration of End-effector and Jig Operations
Division of Information, Science Graduate School of Science and Technology(信息科学研究生院);Nara Institute of Science and Technology(奈良科学技术大学);Materials Informatics Initiative, RD technology and digital transformation center(材料信息学计划,RD技术与数字化转型中心);JSR Corporation(JSR公司);Fine Chemical Process Dept., JSR Corporation(JSR公司精细化学工艺部门);Department of Chemistry Faculty of Science and Technology(科学与技术学部化学系);Keio University(庆应义塾大学);Data Science Center(数据科学中心);Division of Materials Science, Graduate School of Science and Technology(材料科学系,研究生院)