让机器人用全身力量精准扔东西，比只用手更稳更准。
Whole-Body Dynamic Throwing with Legged Manipulators
- 用全身肌肉协同控制，通过强化学习优化投掷与平衡
- 仿真测试中投掷距离和精度提升，稳定性显著改善
- 可推广到任意3D目标，且成功移植到真实人形机器人
用腿式机械臂投掷物体需要精确协调物体操作与运动控制，对现实世界交互至关重要。现有研究多聚焦于操作或运动中的单一任务，极少关注二者结合的复杂行为。本文探索在腿式机械臂中利用全部电机（全身体）而非仅手臂进行投掷。将任务建模为深度强化学习目标，同时优化投掷准确性和机器人稳定性。在仿真中的人形机器人和带臂四足机器人上验证表明，全身体投掷通过利用身体动量、配重平衡和全身动力学，提升了投掷范围、精度和稳定性。我们提出一种优化的自适应课程训练策略，并设计专用强化学习环境，在稀疏奖励条件下实现高效学习。与以往工作不同，本方法可泛化至三维空间任意目标。最后，我们将学习到的控制器成功迁移至真实人形机器人平台。
原文摘要 · Abstract (English)
Throwing with a legged robot involves precise coordination of object manipulation and locomotion - crucial for advanced real-world interactions. Most research focuses on either manipulation or locomotion, with minimal exploration of tasks requiring both. This work investigates leveraging all available motors (full-body) over arm-only throwing in legged manipulators. We frame the task as a deep reinforcement learning (RL) objective, optimising throwing accuracy towards any user-commanded target destination and the robot's stability. Evaluations on a humanoid and an armed quadruped in simulation show that full-body throwing improves range, accuracy, and stability by exploiting body momentum, counter-balancing, and full-body dynamics. We introduce an optimised adaptive curriculum to balance throwing accuracy and stability, along with a tailored RL environment setup for efficient learning in sparse-reward conditions. Unlike prior work, our approach generalises to targets in 3D space. We transfer our learned controllers from simulation to a real humanoid platform.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。