跳到正文
返回论文列表
cs.RO提交于 已译

Grasp-TAGS:通过评论家蒸馏实现任务感知抓取选择,用于抓取式移动操作

Grasp-TAGS: Task-Aware Grasp Selection via Critic Distillation for Prehensile Loco-Manipulation

Xun Tu · Minghao Zou · Karthik Desingh

中文摘要

移动操作(loco-manipulation)要求移动机械臂在长时程中协调移动与操作,是服务机器人和仓储自动化中大物体重排的关键能力。由于性能依赖复杂的机器人-物体与物体-环境交互动力学,可靠执行此类任务仍具挑战。现有方法或依赖基于模型的方法(常需精确动力学模型),或采用基于学习的方法(降低模型需求但通常假设固定或预定义的初始抓取)。然而,不同几何稳定抓取会导致下游操作性能显著差异。为此,提出 Grasp-TAGS,一种面向目标条件大物体重排的任务感知抓取选择框架。在多种初始抓取配置上训练统一的基于 PPO 的强化学习策略,并将学习到的评论家(critic)蒸馏为轻量级抓取效用网络,仅利用候选抓取配置与目标信息对候选抓取排序。推理时,在接触前即选择预测下游效用最高的抓取,供下游策略完成任务。实验表明,Grasp-TAGS 相比固定单抓取策略和随机抓取选择提升了重排成功率,平均成功率达 89.84%,而最佳单抓取策略为 79.87%,提升 9.97 个百分点。Grasp-TAGS 还在所有评估物体上降低了最终朝向误差。更多定性结果见补充网页:https://rpm-lab-umn.github.io/Grasp-TAGS-web-page/

关键要点

  1. 01问题:现有移动操作方法多假设固定或预定义初始抓取,忽略不同几何稳定抓取对下游操作性能的显著影响。
  2. 02方法:在多种初始抓取配置上训练统一 PPO 策略,并将评论家蒸馏为轻量抓取效用网络,仅凭抓取配置与目标信息排序候选抓取。
  3. 03结果:平均重排成功率达 89.84%,较最佳单抓取策略的 79.87% 提升 9.97 个百分点。
  4. 04结果:在所有评估物体上均降低了最终朝向误差。
  5. 05局限:摘要未提及具体局限,仅说明依赖学习策略与蒸馏网络进行抓取选择。

解读

尚无解读。

原始英文摘要

arXiv:2611.09864v1 Announce Type: new Abstract: Loco-manipulation, which requires mobile manipulators to coordinate locomotion and manipulation over long horizons, is a key capability for large-object rearrangement in applications such as service robotics and warehouse automation. Despite recent progress, reliably executing these tasks remains challenging because performance depends on complex robot-object and object-environment interaction dynamics. Existing approaches address this challenge either through model-based methods, which often require accurate dynamics models, or through learning-based methods, which reduce this requirement but typically assume a fixed or predefined initial grasp. However, different geometrically stable grasps can lead to substantially different downstream manipulation performance. Therefore, we propose Grasp-TAGS, a task-aware grasp selection framework for goal-conditioned large-object rearrangement. We train a unified PPO-based reinforcement learning policy across multiple initial grasp configurations and distill the learned critic into a lightweight grasp utility network that ranks candidate grasps using only the candidate grasp configuration and goal information. At inference time, we select the grasp with the highest predicted downstream utility for the downstream policy to complete the task, before even making a contact. Experiments demonstrate that Grasp-TAGS improves rearrangement success over both fixed single-grasp policies and random grasp selection, achieving an average success rate of 89.84% compared with 79.87% for the best-performing single-grasp policies, an improvement of 9.97 percentage points. Grasp-TAGS also reduces final orientation errors across all evaluated objects. For more qualitative results, refer to the supplementary webpage: https://rpm-lab-umn.github.io/Grasp-TAGS-web-page/

同方向论文 · cs.RO

查看全部 →