arXiv：cs.LG（机器学习，全量分类）· Hyeonbeen Lee, Min-Jae Jung, Tae-Kyeong Yeu, Jong-Boo Han, Daegil Park, Simon Stepputtis, Jin-Gyun Kim·· 15 小时前AI 评分32
FDN：面向振动丰富机器人接触的无传感器力/力矩估计频率感知分解网络
Frequency-aware decomposition learning for sensorless wrench estimation in vibration-rich robotic contact
AI 导读
研究者提出频率感知分解网络（FDN），仅凭机器人本体感知即可多步预测振动丰富的力/力矩，无需力/力矩传感器或已辨识的机器人模型。在6-DoF液压机械臂打磨数据上，FDN将高频振幅误差较基线最多降低47%，并在单CPU线程上11 ms内估计1,000 ms时域。消融实验支持其设计选择，从开源日常操作数据集迁移力/力矩动力学可降低低频误差8%，但高频动力学呈现领域特异性。
正文
Abstract:Force and torque (F/T) sensors enable contact-aware control by providing reactive feedback, but they are often fragile and expensive. To overcome these limitations, sensorless methods estimate F/T or wrench solely from robot proprioception, and have shown success in slow interaction tasks such as grasping. However, their low-pass characteristics limit the estimation of high-frequency signals, which are critical in rapid-contact tasks such as grinding. Communication delays can also make their estimates outdated during deployment, but few methods address this directly. To bridge these gaps, we propose a Frequency-aware Decomposition Network (FDN) to estimate vibration-rich wrench in a sensorless, multi-step-ahead manner. Considering higher-frequency stochasticity, FDN spectrally decomposes the wrench horizon into a low-frequency trend and a high-frequency residual, and estimates each by pointwise regression and a learned conditional distribution, respectively. The frequency-aware layers impose band decomposition priors on the outputs and adaptively enhance frequency amplitudes of the inputs. FDN requires neither an identified robot model nor an F/T sensor during estimation. On real-world grinding data from our 6-DoF hydraulic manipulator, FDN reduces high-frequency amplitude error by up to 47% over the baselines under assumed time delays and maintains competitive low-frequency pointwise accuracy, while the baselines fail to balance these two. We also find multi-step-ahead estimation feasible, with FDN estimating a 1,000 ms horizon within 11 ms on a single CPU thread. Ablation studies further support our design choices. In an exploratory study, transferring wrench dynamics learned from an open-source everyday manipulation dataset reduces low-frequency error by 8%, while high-frequency dynamics appear domain-specific.
| Comments: | revised. 27 pages, 10 figures, 10 tables. Code: this https URL Data: this https URL |
| Subjects: | Robotics (cs.RO); Machine Learning (cs.LG) |
| Cite as: | arXiv:2604.12905 [cs.RO] |
| (or arXiv:2604.12905v2 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2604.12905 arXiv-issued DOI via DataCite |
Submission history
From: Hyeonbeen Lee [view email]
[v1]
Tue, 14 Apr 2026 15:54:29 UTC (9,580 KB)
[v2]
Wed, 30 Sep 2026 22:38:23 UTC (14,376 KB)
来源：arXiv：cs.LG（机器学习，全量分类） · arxiv.org