外观
论文arXiv 2606.002525 个月前
HOIST框架融合VLA与RL优化人形机器人悬挂负载操作精度
本文由 AI 根据公开内容整理,可能有误,请以原文为准:HOIST: Humanoid Optimization with Imitation and Sample-efficient Tuning for Manipulating Suspended Loads(arXiv 2606.00252)
核心内容
针对人形机器人通过全身运动及间歇接触控制欠驱动振荡负载的挑战,研究提出HOIST框架以解决纯模仿学习无法优化终端放置、纯强化学习样本效率低且不安全的问题。该方法首先利用VR遥操作演示数据微调高层视觉-语言-动作(VLA)策略,并通过全身控制器执行指令;随后结合VLA回放轨迹与迭代批量强化学习(Batched RL),专门优化放置精度与停稳行为。
在仿真与真机实验中,HOIST性能显著优于纯模仿学习及增加演示数据的基线方案。量化数据显示,相比纯VLA推理,HOIST将平移放置误差降低19.9cm,原始角度误差减少3.56度。该技术路线证明了在人形机器人欠驱动物料搬运场景中,通过“人类示范初始化+RL精细化调优”的混合范式,可有效突破动态负载控制瓶颈并提升作业鲁棒性。
关键看点
- 混合训练架构:采用VR遥操作微调VLA策略作为安全初始行为,结合全身控制器执行,再利用批量RL进行样本高效调优
- 量化精度提升:真机与仿真实验证实,相比纯VLA基线,平移放置误差降低19.9cm,角度误差减少3.56度
- 欠驱动负载控制:解决人形机器人仅能通过全身运动和间歇接触影响悬挂振荡负载的动力学耦合难题
- 样本效率优化:通过VLA Rollouts引导RL探索,规避了真机从零训练强化学习的安全风险与低效问题
工程与产业观察
HOIST为具身智能处理复杂动力学任务提供了工程范式:利用VLA大模型提供语义理解与安全初值,再通过RL在物理约束下微调精细操作。这种分层优化思路有效平衡了泛化性与精度,对人形机器人在物流、建筑等非结构化场景下的柔性物体搬运具有直接参考价值。
英文摘要 (Abstract)
Manipulating suspended payloads with humanoid robots is challenging because the robot can only influence an underactuated, oscillatory load through whole-body motion and intermittent contact. Imitation learning provides safe initial behavior but does not directly optimize final placement, while reinforcement learning from scratch is unsafe and sample-inefficient on real humanoids. We present HOIST-Humanoid Optimized with Imitation and Sample-efficient Tuning for manipulating suspended loads. HOIST first finetunes a high-level vision-language-action (VLA) policy from virtual-reality (VR) teleoperation demonstrations and executes its commands through a whole-body controller. It then uses VLA rollouts and iterative batched RL to improve placement accuracy and stopping behavior. Experiments in simulation and on a real humanoid show that HOIST improves over imitation-only and additional-demonstration baselines; compared with pure VLA rollouts, HOIST reduces translational placement error by 19.9 cm and raw angular error by 3.56 degrees, demonstrating the potential of humanoids for underactuated material-handling tasks.
作者:Songyang Liu, Shunyu Yao, Dingyuan Huang, Shuai Li
来源
本页事实对应的公开页面,以原文为准。