TacBPM：一种用于灵巧重定向的触觉条件行为先验模型
TacBPM: A Tactile-conditioned Behavior Prior Model for Dexterous Reorientation
浏览论文内容
中文总结 AI 辅助
本文提出TacBPM，一种触觉条件行为先验模型，通过蒸馏多尺度球体专家到潜在控制器，利用残差潜在动作加速灵巧重定向策略训练，并在掌上和手臂-手部任务中实现稳定的仿真到现实迁移。
中文摘要 AI 辅助
灵巧的掌上操作需要策略通过间歇性、接触丰富的交互来协调高自由度的手部关节。除了目标姿态跟踪外，此类策略还必须发现手指步态，以在适应几何形状、各向异性、姿态、接触和传感变化的同时保持物体稳定性。我们提出了\method，一种用于灵巧重定向的触觉条件行为先验模型。\method将多尺度球体专家蒸馏到一个潜在控制器中，并让下游策略通过残差潜在动作重用固定的触觉先验，从而减少从原始关节命令进行的新探索。该先验以触觉-本体感觉历史为条件，使潜在行为反映当前的手-物体交互。我们评估了跨各向异性物体的任意姿态转移、指定轴旋转，以及一个手臂-手部抓取到任意姿态任务，在该任务中，机器人必须针对新颖的工具几何形状和泛化的放置位置抓取、提升、运输并达到目标姿态。大量实验表明，所提出的方法加速了训练，并在匹配的原始动作PPO接近失败的情况下实现了稳定策略，同时在掌上和手臂-手部任务中成功实现了仿真到现实的迁移。
英文摘要
Dexterous in-hand manipulation requires policies that coordinate high-DoF hand joints through intermittent, contact-rich interaction. Beyond target-orientation tracking, such policies must discover finger gaits that preserve object stability while adapting to geometry, anisotropy, pose, contact, and sensing changes. We propose \method, a tactile-conditioned behavior prior model for dexterous reorientation. \method distills multi-scale sphere specialists into a latent controller and lets downstream policies reuse the fixed tactile prior through residual latent actions, reducing renewed exploration from raw joint commands. The prior conditions on tactile-proprioceptive history so latent behavior reflects the current hand-object interaction. We evaluate arbitrary-pose transfer across anisotropic objects, commanded-axis rotation, and an arm-hand Grasp-to-AnyPose task in which the robot must grasp, lift, transport, and reach goal poses for novel tool geometries and generalized placements. Extensive experiments demonstrate that the proposed method accelerates training and enables stable policies where matched raw-action PPO remains near failure, with successful sim-to-real transfer in in-hand and arm-hand tasks.
发表机构
- Sharpa Robotics
机构由 AI 辅助整理，请以论文原文为准。
