As learning-based approaches progress towards automating robot controllers design, transferring learned policies to new domains with different dynamics (e.g. sim-to-real transfer) still demands manual effort. This paper introduces SimGAN, a framework to tackle domain adaptation by identifying a hybrid physics simulator to match the simulated trajectories to the ones from the target domain, using a learned discriminative loss to address the limitations associated with manual loss design. Our hybrid simulator combines neural networks and traditional physics simulation to balance expressiveness and generalizability, and alleviates the need for a carefully selected parameter set in System ID. Once the hybrid simulator is identified via adversarial reinforcement learning, it can be used to refine policies for the target domain, without the need to interleave data collection and policy refinement. We show that our approach outperforms multiple strong baselines on six robotic locomotion tasks for domain adaptation.
EIDA: Execution-Interface Dynamics Adaptation for Real-to-Sim-to-Real Robot Navigation
Eastern Institute of Technology（东方理工大学）；National University of Singapore（新加坡国立大学）；The Hong Kong Polytechnic University（香港理工大学）；University of Science and Technology of China（中国科学技术大学）
Human-Robot Shared Control for Surgical Robot Based on Context-Aware Sim-to-Real Adaptation
University of Bristol（布里斯托大学）；Bristol Robotics Lab（布里斯托机器人实验室）；Imperial College London（帝国理工学院）；Polytechnic University of Milan（米兰理工大学）；Shanghai Jiao Tong University（上海交通大学）；Worcester Polytechnic Institute（伍斯特理工学院）
VR-Robo: A Real-to-Sim-to-Real Framework for Visual Robot Navigation and Locomotion
Institute for Interdisciplinary Information Sciences, Tsinghua University（清华大学交叉信息研究院）；Shanghai Qi Zhi Institute（上海启智研究院）；Galaxea AI；School of Electronics, Information and Electrical Engineering, Shanghai Jiao Tong University（上海交通大学电子信息与电气工程学院）
EagleVision: A Multi-Task Benchmark for Cross-Domain Perception in High-Speed Autonomous Racing
Skolkovo Institute of Science, Intelligent Space Robotics Laboratory, Center for Engineering Systems and Sciences（斯科尔科沃科学技术学院，智能空间机器人实验室，工程系统与科学中心）；Khalifa University, KUCARS-KU Center for Autonomous Robotic Systems, Department of Computer Science（哈利法大学，KUCARS-KU自主机器人系统中心，计算机科学系）；Beijing Institute of Technology, Beijing Key Laboratory of UAV Autonomous Control（北京理工大学，北京无人机自主控制重点实验室）
MoSA: Motion-constrained Stress Adaptation for Mitigating Real-to-Sim Gap in Continuum Dynamics via Learning Residual Anisotropy
Hong Kong University of Science（香港科学大学）；MMLab, Chinese University of Hong Kong, Hong Kong SAR（香港中文大学MMLab, 香港特别行政区）；The University of Hong Kong, Hong Kong SAR（香港大学, 香港特别行政区）
Trust-Calibrated VLA Policy Post-Training with Evolving Imagination
Sun Yat-sen University（中山大学）；Shenzhen Loop Area Institute（深圳河套学院）；The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））；Shanghai AI Laboratory（上海人工智能实验室）；University of Oxford（牛津大学）；University of Science and Technology of China（中国科学技术大学）
