arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2606.00537cs.RO
收藏

PACE: 面向动作分块策略的相位感知分块执行方法

PACE: Phase-Aware Chunk Execution for Robot Policies with Action Chunking

Peking University(北京大学);JD Explore Academy(京东探索研究院)

Junnan Nie, Jiayi Li, Chenghao Liu, Junyi Lao, Jiachen Zhang, Tianle Zhang, Songfang Huang

AI总结:

提出PACE方法,通过在线预测动作分块中的低速过渡点作为重规划边界,自适应选择执行步长,无需重新训练即可提升机器人策略成功率。

摘要:

最近的视觉-语言-动作和基于扩散的机器人策略通常使用动作分块,其中每次策略查询预测一系列未来动作,机器人执行一个开环前缀后再重新查询。虽然这种接口改善了局部运动连续性,但部署时仍需选择执行步长:在获取新观测之前应执行每个预测分块的多少。然而,我们的实验表明,成功率强烈依赖于任务且相对于执行步长非单调,这使得单一恒定步长成为不可靠的部署规则。我们提出PACE(相位感知分块执行),一种无需训练的测试时执行方法,从预测分块本身在线选择执行步长。PACE通过识别预测速度剖面中的低速过渡点,利用操作轨迹的相位相关运动学结构,将其作为候选重规划边界。由于PACE仅使用预测的动作分块,因此即插即用,无需重新训练或访问策略内部。我们通过在仿真和真实机器人环境中的大规模评估验证了PACE。在50个RoboTwin2.0任务上,PACE将平均成功率从57.8%提升至64.2%。在双臂ALOHA和单臂Franka平台上的真实机器人实验中,PACE将平均任务得分从60.7提升至77.7,平均成功率从50.7%提升至70.4%。消融实验和轨迹级分析表明,PACE跨操作阶段自适应调整执行步长,在过渡附近缩短执行,同时在连贯运动中保持较长执行。

英文摘要:

Action-chunking robot policies predict sequences of future actions, but deployment requires choosing how many actions to execute before replanning. The execution horizon trades off feedback frequency against motion continuity. However, success varies non-monotonically with the execution horizon, and the best-performing horizon differs across tasks. These effects make a single fixed horizon unreliable as a default execution strategy across tasks. We propose PACE (Phase-Aware Chunk Execution), a training-free method that selects execution horizons online from predicted action chunks. PACE uses prominent low-speed valleys in the predicted speed profile as candidate replanning boundaries, exploiting the phase-dependent kinematics of manipulation trajectories. By default, PACE calibrates the valley threshold from each task's existing training demonstrations, without task-specific fixed-horizon tuning. On the 50-task RoboTwin2.0 benchmark, PACE improves average success from 57.8% to 64.2% over the strongest fixed-horizon baseline and outperforms the adaptive execution baselines AAC and AutoHorizon. With GR00T N1.7 on LIBERO, PACE achieves the highest overall success rate among the evaluated strategies, supporting its applicability across base policies, benchmarks, and action representations. On real robots, PACE improves the average task score from 60.7 to 77.7 and success rate from 50.7% to 70.4% over full-chunk execution. Code is available at https://niejunnan25.github.io/pace-project/.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑