论文/arXiv 2610.14926·18 天前

面向灵巧抓取分类的边缘原生时序-图双专家推理

针对多指接触不同步导致早接触手指过载，提出基于关节角和载荷反馈的DualExpert，并在离线、Flash部署和真机试验中验证。

AI 整理

• 硬件零部件
• 本体 SDK 与控制
• 数据与采集

原文链接：Edge-Native Temporal--Graph Dual-Expert Inference for Dexterous Grasp Classification（arXiv 2610.14926 · 发布日期：2026-10-10）

Edge-Native Temporal--Graph Dual-Expert Inference for Dexterous Grasp Classification

针对多指接触不同步导致早接触手指过载，提出基于关节角和载荷反馈的DualExpert，并在离线、Flash部署和真机试验中验证。

核心要点

• 多指手各指接触物体时刻不同，全局闭合命令可能使早接触手指过载。
• DualExpert用TCN建模短期接触演化并预测Stop，FGFN编码手指拓扑用于分类。
• 0.7/0.3概率融合在225次留出轨迹达93.78%分类准确率，Stop F1为95.25%。
• Flash部署保持精度并降低延迟；50次物理试验达96.00%识别和98.00%抓取成功。

验证范围与局限：验证包括五个物体实例的1,500条自动采集轨迹、225次留出试验、Flash部署和50次物理试验。

英文摘要 (Abstract)

Multi-finger hands contact objects at different times, so a global close command can overload early-contact fingers while others are still moving. Dense tactile arrays address this problem but add hardware, calibration, bandwidth, and computation. We present an edge-native pipeline that uses only built-in joint-angle and pre-use-calibrated actuator-load feedback to recognize the grasped object and predict an independent Stop event for each finger. Four samples and their first differences form a 4-by-20 contact-evolution window. At its core is a heterogeneous dual-expert architecture, termed DualExpert, that assigns complementary tasks to a Temporal Convolutional Network (TCN) and a Finger Graph Filter Network (FGFN). The TCN models short-term contact evolution and supplies all five Stop probabilities, while the FGFN encodes finger topology and contributes only to object classification. Fixed 0.7/0.3 class-probability fusion reaches 93.78% trajectory-level classification accuracy on 225 held-out trials, improving the standalone TCN and FGFN by 4.00 and 2.22 percentage points, respectively. The TCN Stop path attains 95.25% F1. The dataset contains 1,500 automatically collected trajectories from five object instances. In Flash deployment, mixed accelerator execution retains 93.78% classification accuracy and 94.84% Stop F1 with 2.890~ms mean inference time and 9.172~ms mean sample-to-stop latency. Across 50 physical trials, the system achieves 96.00% recognition and 98.00% grasp success while the complete embedded board draws approximately 0.80~W.

作者：Jiajian Zheng, Bin Zhang, Wen Fan, Yanan Liu

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.14926：Edge-Native Temporal--Graph Dual-Expert Inference for Dexterous Grasp Classification访问 2026-10-21