论文arXiv 2611.240113 天前
WAM-Policy实现四足机器人语言指令操作推理加速14倍
本文由 AI 根据公开内容整理,可能有误,请以原文为准:WAM-Policy: World Action Models with On-Policy Distillation for Quadrupedal Loco-Manipulation(arXiv 2611.24011)
核心内容
针对无臂四足机器人缺乏语言条件操作策略及现有世界动作模型(WAM)推理慢的问题,研究提出WAM-Policy框架。该方法在QLM-Bench的10个仿真任务中,利用1000条非遥操作合成数据进行验证,最终策略达到85.0%成功率,并将跨形态迁移后的策略成功部署于Unitree Go2 Air真机,实现了低推理成本下的高性能四足移动操作。
技术架构分两阶段:首先通过跨形态适配与在线自蒸馏(OPSD),仅需1段视频和2步动作即可将机械臂WAM迁移至四足本体,使学生模型采样速度提升14倍且保持76.7%成功率;随后引入任务特定强化学习(RL)优化教师模型,并采用多教师在线蒸馏(M-OPD)合并专家策略,突破单纯模仿学习的性能上限,将任务成功率从76.7%进一步提升至85.0%。
关键看点
- 跨形态蒸馏:OPSD阶段仅需1段视频和2步动作完成WAM到四足适配,推理提速14倍且成功率保持76.7%
- 性能上限突破:结合任务特定RL与多教师在线蒸馏(M-OPD),将QLM-Bench仿真任务成功率从76.7%提升至85.0%
- 数据高效合成:基于1000条非遥操作合成演示数据训练,解决了四足移动操作领域高质量语言指令数据集稀缺难题
- 真机验证落地:加速后的学生策略已成功部署于Unitree Go2 Air,验证了低算力平台下语言驱动四足操作的工程可行性
工程与产业观察
该工作验证了将大模型语义理解能力向非标形态机器人迁移的工程路径。通过在线蒸馏解决WAM推理延迟痛点,结合RL弥补纯模仿学习短板,为四足机器人在低算力端侧实现复杂语言指令操作提供了可复用的算法范式,对具身智能通用化与轻量化部署具有重要参考价值。
英文摘要 (Abstract)
Arm-free quadrupeds can perform loco-manipulation, walking on three legs while manipulating with a front leg, yet no existing policy for it follows language instructions. World action models (WAMs) have emerged as effective language-conditioned manipulation policies, but are pretrained for robot arms and sample slowly. WAM-Policy turns one WAM into a quadrupedal loco-manipulation policy via two stages. The first, Cross-Morphology Policy Adaptation and Acceleration, adapts the arm-centric WAM to quadrupeds and distills the transferred policy into a student with one video and two action steps by on-policy self-distillation (OPSD). This student, however, is bounded by its teacher, which only imitates demonstrations. The second stage, Task-Specific Improvement and Merging, therefore tunes the teacher by reinforcement learning (RL) and merges the slow specialists into one few-step student by multi-teacher on-policy distillation (M-OPD). On QLM-Bench, ten simulated tasks with 1,000 demonstrations synthesized without teleoperation, the transferred policy reaches 77.0%, and the OPSD student keeps 76.7% while sampling 14 times faster. Task-specific RL and M-OPD lift it to 85.0%. We also deploy the OPSD student on a real Unitree Go2 Air. WAM-Policy thus achieves a stronger quadrupedal policy at a fraction of the inference cost.
作者:Junlin Long, Xu Deng, Yiran Wang, Yue Yang, Yang Zhao, Zeyu Zhang
外部来源
本页事实对应的公开页面,以原文为准。(点击将离开本站)