[1]来若阳,殷俊清,陈 静,等.分层视触融合:机器人早期滑移检测[J].机械与电子,2026,44(06):89-95.
　LAI Ruoyang,YIN Junqing,CHEN Jing,et al.Layered Visual-tactile Fusion : Early Slip Detection Robots[J].Machinery & Electronics,2026,44(06):89-95.
点击复制

分层视触融合:机器人早期滑移检测()
分享到：

《机械与电子》[ISSN:1001-2257/CN:52-1052/TH]

卷:
44
期数:
2026年06期
页码:
89-95
栏目:
机器人技术
出版日期:
2026-06-27

文章信息/Info

Title:
Layered Visual-tactile Fusion : Early Slip Detection Robots
文章编号:
1001-2257 ( 2026 ) 06-0089-07
作者:
来若阳; 殷俊清; 陈 静; 李金龙
西安工程大学机电工程学院,陕西 西安 710048
Author(s):
LAI Ruoyang ;  YIN Junqing ;  CHEN Jing;  LI Jinlong
( School of Mechanicaland Electronic Engineering , Xi ’ an Polytechnic University , Xi ’ an 710048 , China )
关键词:
滑移检测; 视触觉融合; 稀疏光流; 机器人抓取
Keywords:
slip detection ;  fusion ;  sparse optical flow ;  robotic grasping
分类号:
TP242.6
文献标志码:
A
摘要:
针对机器人滑移检测中单一模态方法在面对复杂工况时难以兼顾高鲁棒性与低延迟的问题,提出一种基于三维触觉力感知与稀疏视觉光流融合的实时分层滑移检测框架。该框架协同“眼在手外”相机的全局视觉与三维触觉反馈来实现状态监测。一方面借助库仑摩擦模型推导切向力比值,将其作为预警信号以预判早期滑移;另一方面,利用 Lucas Kanade 算法提取稀疏光流特征,从而有效辨别目标物体发生的究竟是刚体位移还是弹性变型。动态抓取实验表明,该融合策略成功解决了纯视觉方法在软体如海绵形变时的误报问题,并有效克服了纯触觉方法的滞后性,定量分析证实其能在宏观滑移发生前平均 700 ms 提供关键预警时间窗。该系统显著提升了机器人在非结构化环境下的抓取鲁棒性,为复杂场景下的灵巧操作提供了有效方案。
Abstract:
To address the challenges of this proposes a real-time hierarchical slip detection framework three-dimensional tactile force sensing and sparse visual optical flow.Th framework global vision from an “ eye in hand out ” camera with three dimensional tactile feedback to achieve state monitoring.On one hand , the Coulomb friction model is used to derive the tangential force ratio , which serves as a warning signal to predict slip ; n the other hand , the Lucas Kanade algorithm is employed to extract sparse optical flow features , effectively distinguishing rigid displacement or elastic deformation.Dynamic grasping experiments demonstrate that this fusion strategy successfully the false vision methods soft materials sponges and effectively overcomes the latency limitations of tactile methods.Quantitative analysis confirms that provide a crucial warning window averaging 700 ms before macroscopic slip occurs.This significantly enhances the robustness of robotic grasping in unstructured environments and offers an effective solution for dexterous manipulation in complex scenarios.

参考文献/References:

[ 1 ] Liu Yuchen , Zhang Jintao , Lou Yuanxin , et al.Soft bionic gripper with tactile sensing and slip detection for damage-free grasping of fragile fruits and vegetables [ J ] . Computers and Electronics in Agriculture , 2024 , 220 : 108904.

[ 2 ] Jia Chun , Liu Yujie , Li Hiu.Single-frequency cycle slip detection and repair for a standalone GNSS receiver using a common-antenna-based dual-board design [ J ] .GPS Solutions , 2024 , 28 ( 4 ): 183.
[ 3 ] 王菊,管声启,李杭,等 . 基于改进 BP 神经网络的机械手抓取力 模 糊 PID 控 制[ J ] . 机 械 与 电 子, 2024 , 42( 12 ): 25-30.
[ 4 ] Wang Yancheng , Wu Xin , Mei Deqing , et al.Flexible tactile sensor array for distributed tactile sensing and slip detection in robotic hand grasping [ J ] .Sensors and Actuators A : Physical , 2019 , 297 : 111512.
[ 5 ] Jiang Chengpeng , Zhang Zhang , Pan Jing , et al.Finger skin-inspired flexible optical sensor for force sensing and slip detection in robotic grasping [ J ] .Advanced Materials Technologies , 2021 , 6 ( 10 ): 2100285.
[ 6 ] Hou Y C , Sahari K S M , How D N T.A review on modeling of flexible deformable object for dexterous robotic manipulation [ J ] .International Journal of Advanced Robotic Systems , 2019 , 16 ( 3 ): 196178717.
[ 7 ] Wu Guirong , Shi Junwei , Qiu Huangweiming , et al.Flexible multidimensional force sensors : a review of structural design , performance , and applications [ J / OL ] . Advanced Materials Technologies , 2025-10-31.https : ∥doi.org / 10.1002 / admt.202501575.
[ 8 ] Hu Xiaohai , Venkatesh A , Wan Yusen , et al.Learning to detect slip through tactile estimation of the contact force field and its entropy properties [ J ] .Mechatronics , 2024 , 104 : 103258.
[ 9 ] Fang Ke , Liu Zongzheng , Yang Haisen , et al.High speed grasping system of manipulator based on event camera and dynamic compressive sensing method [ C ] ∥ 2025 International Conference on Mechatronics , Robotics , and Artificial Intelligence ( MRAI ) .New York : IEEE , 2025 : 346-351.
[ 10 ] Sui Ruomin , Zhang Lunwei , Li Tiemin , et al.Incipient slip detection method for soft objects with vision based tactile sensor [ J ] . Measurement , 2022 , 203 : 111906.
[ 11 ] Florence P R , Manuelli L , Tedrake R.Dense object nets : Learning dense visual object descriptors by and for robotic manipulation [ PP / OL ] .V2.arXiv ( 2018-09-07 )[ 2026-01-02 ] .https : ∥doi.org / 10.48550 / arXiv.1806.08756.
[ 12 ] Keely M , Kim Y , Mehta S A , et al.Combining and decoupling rigid and soft grippers to enhance robotic manipulation [ J / OL ] .Soft Robotics , 2025 -03-10.https : ∥doi.org / 10.1089 / soro.2024.0062.
[ 13 ] Liu Chaowen , Wu Hao , Shen Jingjin.Efficient visual tactile transformer with token reorganization for robotic slip detection [ J ] .Journal of the Brazilian Society of Mechanical Sciences and Engineering , 2025 , 47 ( 6 ): 265.
[ 14 ] Murray R M , Li Zexing , Sastry S S.A mathematical introduction to robotic manipulation [ M ] .Boca Raton : CRC press , 2017.
[ 15 ] Howe R D , Cutkosky M R.Sensing skin acceleration for slip and texture perception [ C ] ∥ 1989 International Conference on Robotics and Automation.New York : IEEE , 1989 : 145-150.[ 16 ] Wriggers P , Krstulovic-Opara L.The moving friction cone approach for three-dimensional contact simulations [ J ] .International Journal of Computational Methods , 2004 , 1 ( 1 ): 105-119.
[ 17 ] Bicchi A.Hands for dexterous manipulation and robust grasping : a difficult road toward simplicity [ J ] . IEEE Transactions on Robotics and Automation , 2002 , 16 ( 6 ): 652-662.
[ 18 ] Tremblay M R , Cutkosky M R.Estimating friction using incipient slip sensing during a manipulation task [ C ] ∥Proceedings IEEE International Conference on Robotics and Automation.New York : IEEE , 1993 : 429-434.
[ 19 ] Kerr J , Roth B.Analysis of multifingered hands [ J ] . The International Journal of Robotics Research , 1986 , 4 ( 4 ): 3-17.
[ 20 ] Shi Jianbo , Tomasi.Good features to track [ C ] ∥1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.New York : IEEE , 1994 : 593-600.
[ 21 ] Dong Siyuan , Yuan Wenzhen , Adelson E H.Improved gelsight tactile sensor for measuring geometry and slip [ C ] ∥ 2017 IEEE / RSJ International Conference on Intelligent Robots and Systems ( IROS ) .New York : IEEE , 2017 : 137-144.

备注/Memo

备注/Memo:
收稿日期: 2026-03-08
作者简介:来若阳 ( 1999- ),男,陕西西安人,硕士研究生,研究方向为仿生机器人;殷俊清 ( 1981- ),男,江苏南通人,博士,副教授,研究方向为仿生机器人以及先进装配工艺与设备等。
更新日期/Last Update: 2026-08-26