Quadrupedal landing is a complex process involving large impacts, elaborate contact transitions, and is a crucial recovery behavior observed in many biological animals. This work presents a real-time, optimal landing controller that is free of pre-specified contact schedules. The controller determines optimal touchdown postures and reaction force profiles and is able to recover from a variety of falling configurations. The quadrupedal platform used, the MIT Mini Cheetah, recovered safely from drops of up to 8 m in simulation, as well as from a range of orientations and planar velocities. The controller is also tested on hardware, successfully recovering from drops of up to 2 m.
Decoupled Thrust-Axis Attitude Control Using Quaternions for Chandrayaan-3 Lunar Landing Mission
Controls and Digital Area, U R Rao Satellite Centre, Indian Space Research Organization (ISRO), Bangalore, India(控制与数字领域部,U R Rao卫星中心,印度空间研究组织(ISRO),班加罗尔,印度)
Massachusetts Institute of Technology(麻省理工学院);NSF AI Institute for Artificial Intelligence and Fundamental Interactions(美国国家科学基金会人工智能与基础交互研究所)
敏捷的机动,如短跑和高速转弯在野外是具有挑战性的足式机器人。我们提出了一个端到端的学习控制器,实现了MIT Mini Cheetah的创纪录敏捷性,维持速度高达3.9 m/s。该系统在草地、冰和砾石等自然地形上快速运行和转弯,并对干扰做出稳健的响应。我们的控制器是一个神经网络,通过强化学习在模拟中训练,并转移到现实世界。两个关键组成部分是(i)速度命令的自适应课程和(ii)利用先前的工作进行仿真到现实迁移的在线系统辨识策略。有关机器人行为的视频,请访问:https://agility.csail.mit.edu/
Momentum-Aware Trajectory Optimization and Control for Agile Quadrupedal Locomotion
Institute for Robotics and Intelligent Machines, Georgia Institute of Technology(佐治亚理工学院机器人与智能机器研究所);MIT Biomimetic Robotics Lab(麻省理工学院仿生机器人实验室);NAVER LABS(NAVER实验室)
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。
STL-SVPIO: Signal Temporal Logic guided Stein Variational Path Integral Optimization
Department of Electrical and Systems Engineering, University of Pennsylvania(宾夕法尼亚大学电气与系统工程系);Mechanical Engineering and Mechanics Department, Lehigh University(莱特大学机械工程与力学系)