We propose a hybrid model predictive control algorithm, consensus complementarity control (C3), for systems that make and break contact with their environment. Many state-of-the-art controllers for tasks which require initiating contact with the environment, such as locomotion and manipulation, require a priori mode schedules or are so computationally complex that they cannot run at real-time rates. We present a method, based on the alternating direction method of multipliers (ADMM), capable of highspeed reasoning over potential contact events. Via a consensus formulation, our approach enables parallelization of the contact scheduling problem. We validate our results on three numerical examples, including two frictional contact problems, and physical experimentation on an underactuated multi-contact system.
Improved Hierarchical ADMM for Nonconvex Cooperative Distributed Model Predictive Control
National University of Singapore（新加坡国立大学）；University of California, Berkeley（加州大学伯克利分校）；Temasek Laboratories, National University of Singapore（新加坡国立大学淡马锡实验室）；University of British Columbia（不列颠哥伦比亚大学）
Whole-Body Nonlinear Model Predictive Control Through Contacts for Quadrupeds
Agile & Dexterous Robotics Lab, ETH Zürich（敏捷灵巧机器人实验室，苏黎世联邦理工学院）；Robotic Systems Lab, ETH Zürich（机器人系统实验室，苏黎世联邦理工学院）；ETH Zürich（苏黎世联邦理工学院（注：此处ETH Zürich是机构名，与前面实验室同属苏黎世联邦理工学院））
Differentiable Compliant Contact Primitives for Estimation and Model Predictive Control
Fraunhofer IPK（弗劳恩霍夫生产设备与设计技术研究所）；DGIST（大邱庆北科学技术院）；Politecnico di Milano（米兰理工大学）；Istituto Dalle Molle di Studi sull’Intelligenza Artificiale (IDSIA)（达莱莫勒人工智能研究所（IDSIA））；Scuola Universitaria Professionale della Svizzera Italiana (SUPSI)（瑞士意大利语区应用科学大学（SUPSI））；Università della Svizzera Italiana (USI)（瑞士意大利语区大学（USI））
When to make a step? Tackling the timing problem in multi-contact locomotion by TOPP-MPC
Laboratoire d’Informatique, de Robotique et de Microélectronique de Montpellier (LIRMM), CNRS–University of Montpellier（蒙彼利埃计算机科学、机器人技术与微电子实验室（LIRMM），法国国家科学研究中心–蒙彼利埃大学）；School of Mechanical and Aerospace Engineering, Nanyang Technological University（南洋理工大学机械与航空航天工程学院）
Real-Time Nonlinear MPC via Sequential Quadratic Programming with Structure-Exploiting ADMM and Interior-Point Methods for Underactuated Double-Pendulum Swing-Up
Shield-Loco: Shielding Locomotion Policies with Predictive Safety Filtering
Robert Bosch Center for Cyber Physical Systems（罗伯特·博世网络物理系统中心）；Indian Institute of Science（印度科学研究院）；Munich Institute of Robotics and Machine Intelligence (MIRMI)（慕尼黑机器人与机器智能研究所（MIRMI））；Technical University of Munich（慕尼黑技术大学）；Department of Computer Science & Automation（计算机科学与自动化部门）；Department of Mechanical Engineering（机械工程系）；Carnegie Mellon University（卡内基梅隆大学）；Institute for Advanced Study（高级研究 institute）
Multi-contact Stochastic Predictive Control for Legged Robots with Contact Locations Uncertainty
Max Planck Institute for Intelligent Systems（马克斯·普朗克智能系统研究所）；Tandon School of Engineering, New York University（纽约大学坦登工程学院）；Munich Institute of Robotics and Machine Intelligence, Technical University of Munich（慕尼黑工业大学慕尼黑机器人与机器智能研究所）；University of Trento（特伦托大学）
Hybrid Data-Driven Predictive Control for Robust and Reactive Exoskeleton Locomotion Synthesis
Department of Computation and Neural Systems, Caltech（计算与神经系统系，加州理工学院）；Department of Mechanical and Civil Engineering, Caltech（机械与土木工程系，加州理工学院）；Wandercraft；Department of Computing and Mathematical Sciences, Caltech（计算与数学科学系，加州理工学院）；Department of Control and Dynamical Systems, Caltech（控制与动态系统系，加州理工学院）
Generative Models From and For Sampling-Based MPC: A Bootstrapped Approach For Adaptive Contact-Rich Manipulation
Oxford Robotics Institute, University of Oxford, UK（牛津大学机器人研究所）；Robotics and AI Institute (RAI), Boston, USA（机器人与人工智能研究所）；Cornell University, Ithaca, NY, USA（康奈尔大学）
WarpMPC: Large-Batch MPC on GPU via ADMM with Unrolled $LDL^\top$ Factorization
Institute for Data Science in Mechanical Engineering (DSME), RWTH Aachen University（亚琛工业大学机械工程数据科学研究所）；Department of Mechanical Engineering, Massachusetts Institute of Technology（麻省理工学院机械工程系）
Do My Samples Help? Compute Budget Allocation for Real-World Sampling-Based MPC in Contact-Rich Manipulation
TU Darmstadt（达姆施塔特工业大学）；German Research Center for AI (DFKI)（德国人工智能研究中心）；Robotics Institute Germany (RIG)（德国机器人研究所）；School of Computing, DePaul University（德保罗大学计算学院）
Centroidal Trajectory Generation and Stabilization based on Preview Control for Humanoid Multi-contact Motion
CNRS-AIST JRL (Joint Robotics Laboratory)（CNRS-AIST联合机器人实验室）；IRL（独立研究机构）；National Institute of Advanced Industrial Science and Technology (AIST)（国家先进工业科学与技术研究院）
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。
