Robotic assembly presents a long-standing challenge due to its requirement for precise, contact-rich manipulation. While simulation-based learning has enabled the development of robust assembly policies, their performance often degrades when deployed in real-world settings due to the sim-to-real gap. Conversely, real-world reinforcement learning (RL) methods avoid the sim-to-real gap, but rely heavily on human supervision and lack generalization ability to environmental changes. In this work, we propose a hybrid approach that combines a simulation-trained base policy with a real-world residual policy to efficiently adapt to real-world variations. The base policy, trained in simulation using low-level state observations and dense rewards, provides strong priors for initial behavior. The residual policy, learned in the real world using visual observations and sparse rewards, compensates for discrepancies in dynamics and sensor noise. Extensive real-world experiments demonstrate that our method, SPARR, achieves near-perfect success rates across diverse two-part assembly tasks. Compared to the state-of-the-art zero-shot sim-to-real methods, SPARR improves success rates by 38.4% while reducing cycle time by 29.7%. Moreover, SPARR requires no human expertise, in contrast to the state-of-the-art real-world RL approaches that depend heavily on human supervision.
IndustReal: Transferring Contact-Rich Assembly Tasks from Simulation to Reality
University of Southern California(南加州大学);Stanford University(斯坦福大学);NVIDIA Corporation(英伟达公司);University of Sydney(悉尼大学);University of Washington(华盛顿大学)
InsertAnything: Generalizable Contact-Rich Precision Insertion from Simulation to Reality
University of Chinese Academy of Sciences(中国科学院大学);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);PaXini AI Technology (Beijing) Co., Ltd.(北京他山科技有限公司)
The University of Hong Kong(香港大学);The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));ETH Zurich(苏黎世联邦理工学院);The Hong Kong University of Science and Technology(香港科技大学)
WireCraft: A Simulation Benchmark for Industrial DLO Manipulation
Department of Mechanical and Industrial Engineering, University of Toronto(多伦多大学机械与工业工程系);Department of Computer Science, University of Toronto(多伦多大学计算机科学系);CREFLE Inc.(CREFLE公司)
mimic-one: a Scalable Model Recipe for General Purpose Robot Dexterity
mimic robotics;ETH AI Center, ETH Zurich(ETH人工智能中心,ETH Zurich);Soft Robotics Lab, Dept. of Mechanical and Process Engineering, ETH Zurich(软机器人实验室,机械与过程工程系,ETH Zurich);Institute of Neuroinformatics, ETH Zurich and University of Zurich(神经信息研究所,ETH Zurich和苏黎世大学)
ForceRFT: Refining VLA Actions through Force-Guided Residual Reinforcement Learning
Eastern Institute of Technology(东方理工大学);Jiangnan University(江南大学);Shanghai Jiao Tong University(上海交通大学);The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));National University of Singapore(新加坡国立大学)
Information-Theoretic Graph Fusion with Vision-Language-Action Model for Policy Reasoning and Dual Robotic Control
Electrical and Computer Engineering Department, University of New Mexico, Albuquerque, United States, 87106(电气与计算机工程系,新墨西哥大学,阿尔伯克基,美国,87106);Dynamic Robot Systems Group, Oxford Robotics Institute, University of Oxford, United Kingdom, OX26NN(动态机器人系统组,牛津机器人研究所,牛津大学,英国,OX26NN);Mechanical and Aerospace Engineering Department, The George Washington University, DC, United States, 22202(机械与航空航天工程系,乔治华盛顿大学,华盛顿特区,美国,22202);Department of Computer Science, University of Alabama at Birmingham, Alabama, United States, 35294(计算机科学系,阿拉巴马大学伯明翰分校,阿拉巴马,美国,35294);The School of Computation, Information and Technology, Technical University of Munich, Germany, 85748(计算、信息与技术学院,慕尼黑技术大学,德国,85748);Department of Mechanical Engineering, Politecnico di Milano, Milan, Italy, 20156(机械工程系,米兰理工学院,米兰,意大利,20156)