Legged robots can outperform wheeled machines for most navigation tasks across unknown and rough terrains. For such tasks, visual feedback is a fundamental asset to provide robots with terrain-awareness. However, robust dynamic locomotion on difficult terrains with real-time performance guarantees remains a challenge. We present here a real-time, dynamic foothold adaptation strategy based on visual feedback. Our method adjusts the landing position of the feet in a fully reactive manner, using only on-board computers and sensors. The correction is computed and executed continuously along the swing phase trajectory of each leg. To efficiently adapt the landing position, we implement a self-supervised foothold classifier based on a Convolutional Neural Network (CNN). Our method results in an up to 200 times faster computation with respect to the full-blown heuristics. Our goal is to react to visual stimuli from the environment, bridging the gap between blind reactive locomotion and purely vision-based planning strategies. We assess the performance of our method on the dynamic quadruped robot HyQ, executing static and dynamic gaits (at speeds up to 0.5 m/s) in both simulated and real scenarios; the benefit of safe foothold adaptation is clearly demonstrated by the overall robot behavior.
ViTAL: Vision-Based Terrain-Aware Locomotion for Legged Robots
Istituto Italiano di Tecnologia (IIT)（意大利理工学院）
这项工作是基于视觉的规划策略的足式机器人分离到立足点选择和姿势适应运动规划。目前的姿态自适应策略优化机器人的身体姿势相对于给定的立足点。如果未到达这些立足点，则机器人可能最终处于没有可到达的安全立足点的状态。因此，我们提出了一个基于视觉的地形感知运动（ViTAL）的战略，包括新的姿态适应和立足点选择算法。ViTAL在姿势适应中引入了一种不同的范例，该范例不优化相对于给定立足点的身体姿势，而是最大化腿部到达安全立足点的机会的身体姿势。ViTAL根据机器人的能力和地形感知能力来规划立足点和姿势。我们使用90 kg HyQ和140 kg HyQReal四足机器人来验证ViTAL，并表明它们能够以不同的速度和步态爬上各种障碍物，包括楼梯，间隙和粗糙的地形。我们比较ViTAL与基线策略，选择机器人姿势的基础上，给定选定的立足点，并显示ViTAL优于基线。
Istituto Italiano di Tecnologia（意大利技术研究院）；Università degli Studi di Genova（热那亚大学）；Swiss Federal Institute of Technology at Lausanne（洛桑联邦理工学院）；University of Texas（德克萨斯大学）；University of Naples Federico II（那不勒斯费德里科二世大学）
Walk the PLANC: Physics-Guided RL for Agile Humanoid Locomotion on Constrained Footholds
Department of Mechanical and Civil Engineering, California Institute of Technology（机械与土木工程系，加州理工学院）；Department of Computing and Mathematical Sciences, California Institute of Technology（计算与数学科学系，加州理工学院）
Watch Your Step: Learning Semantically-Guided Locomotion in Cluttered Environment
School of Electrical and Electronic Engineering, Nanyang Technological University（南洋理工大学电气与电子工程学院）；School of Mechanical and Mining Engineering, The University of Queensland（昆士兰大学机械与采矿工程学院）
BeamDojo: Learning Agile Humanoid Locomotion on Sparse Footholds
Shanghai AI Laboratory（上海人工智能实验室）；Shanghai Jiao Tong University（上海交通大学）；Zhejiang University（浙江大学）；The University of Hong Kong（香港大学）；The Chinese University of Hong Kong（香港中文大学）
Adaptive Step Duration for Accurate Foot Placement: Achieving Robust Bipedal Locomotion on Terrains with Restricted Footholds
Mechanical and Aerospace Engineering, The Ohio State University（机械与航空航天工程系，俄亥俄州立大学）；Electrical and Computer Engineering, The Ohio State University（电气与计算机工程系，俄亥俄州立大学）
START: Traversing Sparse Footholds with Terrain Reconstruction
Institute of Cyber-Systems and Control, Zhejiang University（浙江大学控制系统研究所）；MARMoT Lab, National University of Singapore（新加坡国立大学MARMoT实验室）；th Asian Games Hangzhou 2022 Organising Committee（杭州2022年第19届亚运会组委会）
TAGA: Terrain-aware Active Gaze Learning for Generalizable Agile Humanoid Locomotion
MarmotLab, National University of Singapore（马尔莫实验室，新加坡国立大学）；Center of X-Mechanics, Zhejiang University（浙大X力学中心）；South China University of Technology（华南理工大学）
CReF: Cross-modal and Recurrent Fusion for Depth-conditioned Humanoid Locomotion
Institute of Cyber-Systems and Control, Zhejiang University（浙江大学控制科学与工程学院）；School of Control Science and Engineering, Shandong University（山东大学控制科学与工程学院）
TACT-ful: Multi-Channel Terrain Affordance and Compliance Training for Payload-Robust Perceptive Humanoid Locomotion
VinRobotics, Vietnam（越南VinRobotics）；Center for AI Research, VinUniversity, Vietnam（越南VinUniversity人工智能研究中心）；Intelligent Autonomous Systems, TU Darmstadt, Germany（德国达姆施塔特工业大学智能自主系统）
Mind Your Steps: A General Learning Framework for Accurate Humanoid Foothold Tracking
Politecnico di Milano（米兰理工大学）；TU Darmstadt（达姆施塔特工业大学）；Max Planck Institute for Intelligent Systems（马克斯·普朗克智能系统研究所）；Italian Institute of Technology（意大利技术研究院）；University of Pisa（比萨大学）
SABER: Learning Attention-based Semantic Affordance for Legged Locomotion
Institute of Advanced Intelligence and Computing (IAIC), Agency for Science, Technology and Research (A*STAR)（先进智能与计算研究所，新加坡科技研究局）；Nanyang Technological University（南洋理工大学）
Load-bearing Assessment for Safe Locomotion of Quadruped Robots on Collapsing Terrain
Dynamic Legged Systems Lab, Istituto Italiano di Tecnologia (IIT)（动态腿系统实验室，意大利技术研究院）；Department of Mechanical Engineering, São Carlos School of Engineering (EESC), University of São Paulo (USP)（机械工程系，圣卡洛斯工程学院（EESC），圣保罗大学）
Zero-Shot Retargeting of Learned Quadruped Locomotion Policies Using Hybrid Kinodynamic Model Predictive Control
University of Notre Dame（圣母大学）；Robotics at Google（谷歌机器人部门）
强化学习（RL）在四足动物运动方面取得了长足的进步，并在策略的可靠仿真到现实迁移方面取得了持续的进展。然而，在另一个机器人上重用策略仍然是一个挑战，这可以节省重新训练的时间。在这项工作中，我们提出了一个框架zero-shot政策重定向，其中不同的运动技能可以在不同形状和大小的机器人之间转移。新的框架集中在一个规划和控制管道，系统地集成RL和模型预测控制（MPC）。规划阶段采用RL生成动态似然轨迹以及接触时间表，避免了接触序列优化的组合复杂性。然后，这些信息被用于播种MPC，以通过一个新的混合Kinodynamic（HKD）模型来稳定和鲁棒化政策推出，该模型隐式优化了立足点位置。硬件结果表明，能够将策略从A1和Laikago机器人传输到MIT Mini Cheetah机器人，而无需任何策略重新调整。
Agile and Generalized Legged Locomotion via Attention-Based Neural Map Encoding
Robotic Systems Lab, ETH Zurich, Switzerland（机器人系统实验室，苏黎世联邦理工学院，瑞士）；Secure, Reliable, and Intelligent Systems Lab, ETH Zurich, Switzerland（安全、可靠和智能系统实验室，苏黎世联邦理工学院，瑞士）；ETH AI Center, Switzerland（苏黎世联邦理工学院人工智能中心，瑞士）
Omnidirectional Humanoid Locomotion on Stairs via Unsafe Stepping Penalty and Sparse LiDAR Elevation Mapping
School of Artificial Intelligence and Automation, Huazhong University of Science and Technology（华中科技大学人工智能与自动化学院）；State Key Laboratory of Intelligent Manufacturing Equipment and Technology, Huazhong University of Science and Technology（华中科技大学智能制造装备技术国家重点实验室）；School of Mechanical Science and Engineering, Huazhong University of Science and Technology（华中科技大学机械科学与工程学院）
Terrain Traversal Without Motion Priors: Expert-Routed Humanoid Locomotion at Deployment Fidelity
VinRobotics（维诺机器人公司）；German Research Center for Artificial Intelligence (DFKI)（德国人工智能研究中心）；University of Stuttgart（斯图加特大学）；International Max Planck Research School for Intelligent Systems (IMPRS-IS)（国际马克斯·普朗克智能系统研究院）；Center for AI Research (CAIR), VinUniversity（VinUniversity人工智能研究中心）；Intelligent Autonomous Systems, TU Darmstadt（达姆施塔特工业大学智能自主系统研究所）
PointWalker: Mapless Quadruped Locomotion Directly From Raw LiDAR Point Clouds
College of Control Science and Engineering, Zhejiang University（浙江大学控制科学与工程学院）；Yunshenchu Technology Co., Ltd.（云深处科技有限公司）；Zhejiang Key Laboratory of Additive Manufacturing Technology and Equipment（浙江省增材制造技术与装备重点实验室）；Hangzhou Public Library（杭州图书馆）
Learning Quadrupedal Locomotion over Challenging Terrain
Robotic Systems Lab, ETH Zurich（苏黎世联邦理工学院机器人系统实验室）；Robotics and Artificial Intelligence Lab, KAIST（韩国科学技术院机器人与人工智能实验室）；Intelligent Systems Lab, Intel（英特尔智能系统实验室）
Feasible Region: an Actuation-Aware Extension of the Support Region
Istituto Italiano di Tecnologia（意大利理工学院）；ANYbotics AG（ANYbotics公司）；Laboratoire d’Informatique, de Robotique et de Microelectronique de Montpellier (LIRMM), CNRS-University of Montpellier（法国国家科学研究中心-蒙彼利埃大学蒙彼利埃计算机科学、机器人与微电子实验室）；Jet Propulsion Laboratory (JPL), NASA（美国国家航空航天局喷气推进实验室）
FootQuery: Future-Touchdown-Guided Retrieval from Depth History for Perceptive Humanoid Locomotion
Institute for AI Industry Research (AIR), Tsinghua University（清华大学人工智能产业研究院（AIR））；University of Science and Technology Beijing（北京科技大学）；Nanyang Technological University（南洋理工大学）
