Scenarios requiring humans to choose from multiple seemingly optimal actions are commonplace, however standard imitation learning often fails to capture this behavior. Instead, an over-reliance on replicating expert actions induces inflexible and unstable policies, leading to poor generalizability in an application. To address the problem, this paper presents the first imitation learning framework that incorporates Bayesian variational inference for learning flexible non-parametric multi-action policies, while simultaneously robustifying the policies against sources of error, by introducing and optimizing disturbances to create a richer demonstration dataset. This combinatorial approach forces the policy to adapt to challenging situations, enabling stable multi-action policies to be learned efficiently. The effectiveness of our proposed method is evaluated through simulations and real-robot experiments for a table-sweep task using the UR3 6-DOF robotic arm. Results show that, through improved flexibility and robustness, the learning performance and control safety are better than comparison methods.
Safety-Aware Imitation Learning via MPC-Guided Disturbance Injection
Department of Electrical Engineering, Tsinghua University（电子工程系，清华大学）；Department of Electrical and Computer Engineering, University of Southern California（电气与计算机工程系，南加州大学）；Department of Aeronautics and Astronautics, Stanford University（航空与航天工程系，斯坦福大学）
Bayesian Disturbance Injection: Robust Imitation Learning of Flexible Policies for Robot Manipulation
NAIST（奈良先端科学技术大学院大学）
人类在执行任务时表现出各种有趣的行为特征，例如在看似等效的最佳行动之间进行选择，在偏离最佳轨迹时执行恢复行动，或根据感知到的风险调整行动。然而，模仿学习试图通过观察人类的演示来教机器人执行这些相同的任务，但往往无法捕捉到这种行为。具体地，常用的学习算法体现了学习假设之间的固有矛盾（例如，单个最佳动作）和实际人类行为（例如，多个最佳动作），从而限制了机器人的通用性、适用性和演示可行性。为了解决这个问题，本文提出了设计模仿学习算法，重点是利用人类的行为特征，从而体现捕捉和利用实际的演示者的行为特征的原则。本文提出了第一个模仿学习框架，贝叶斯扰动注入（BDI），典型的人类行为特征，将模型的灵活性，鲁棒性和风险敏感性。贝叶斯推理用于学习灵活的非参数多行动策略，同时通过注入风险敏感的干扰来诱导人类恢复行动并确保演示可行性来增强策略。我们的方法是通过风险敏感的模拟和真实的机器人实验（例如，表扫描任务，轴达到任务和轴插入任务），使用UR 5e 6-DOF机械臂，以证明行为的改进表征。结果表明，通过提高灵活性，鲁棒性以及演示的可行性，任务性能显着改善。
Tube Diffusion Policy: Reactive Visual-Tactile Policy Learning for Contact-rich Manipulation
Meta Reality Labs Research（Meta现实实验室）；Idiap Research Institute（Idiap研究机构）；École Polytechnique Fédérale de Lausanne（瑞士联邦理工学院）；University of Toronto（多伦多大学）；Purdue University（普渡大学）
Where Small Mistakes Matter: Quantifying State Criticality for Imitation Learning
University of Southern California（南加州大学）；Stanford University（斯坦福大学）；Georgia Institute of Technology（佐治亚理工学院）；Inria Paris（巴黎信息与自动化研究所）；PSL University（巴黎文理研究大学）
Toward Autonomous Soft Robotic Endovascular Navigation via Imitation Learning
Johns Hopkins University（约翰霍普金斯大学）；Stanford University（斯坦福大学）；McGill University（麦吉尔大学）；University of Maryland（马里兰大学）；Swiss Federal Institute of Technology in Lausanne (EPFL)（日内瓦联邦理工学院（EPFL））；ETH Zurich（苏黎世联邦理工学院）
Imitation learning-based spacecraft rendezvous and docking method with Expert Demonstration
Department of Control Science and Engineering, Harbin Institute of Technology（控制科学与工程系，哈尔滨工业大学）；Department of Mechanical and Automation Engineering, The Chinese University of Hong Kong（机械与自动化工程系，香港中文大学）
Design and Experimental Validation of Sensorless 4-Channel Bilateral Teleoperation for Low-Cost Manipulators
Degree Programs in Intelligent and Mechanical Interaction Systems, University of Tsukuba（智能与机械交互系统专业，东京大学）；Faculty of Engineering, Information and Systems, University of Tsukuba（工程、信息与系统学部，东京大学）；Department of Electrical Engineering, Electronics, and Applied Physics, Saitama University（电子工程、电子学与应用物理系，埼玉大学）
A Whole-Body Motion Imitation Framework from Human Data for Full-Size Humanoid Robot
The State Key Laboratory of Industrial Control and Technology of Zhejiang University（浙江大学工业控制技术状态重点实验室）；Zhejiang Humanoid Robot Innovation Center（浙江人形机器人创新中心）；Department of Engineering Mechanics of Zhejiang University（浙江大学工程力学系）
OminiAdapt: Learning Cross-Task Invariance for Robust and Environment-Aware Robotic Manipulation
University of Science and Technology of China（中国科学技术大学）；Beijing Institute of Technology（北京理工大学）；Institute of Automation Chinese Academy of Sciences（中国科学院自动化研究所）
Aerobatic maneuvers in insect-scale flapping-wing aerial robots via deep-learned robust tube model predictive control
Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology（电气工程与计算机科学系，麻省理工学院）；Department of Aeronautics and Astronautics, Massachusetts Institute of Technology（航空宇航系，麻省理工学院）
RPG: Robust Policy Gating for Smooth Multi-Skill Transitions in Humanoid Fighting
Center for Artificial Intelligence and Robotics, Shenzhen International Graduate School, Tsinghua University（人工智能与机器人中心，深圳国际研究生院，清华大学）；Shanghai AI Laboratory（上海人工智能实验室）；Northwestern Polytechnical University（西北工业大学）；Shanghai Jiao Tong University（上海交通大学）
