A Gradient Grid Structure Enables a Flexible Tactile Sensor to Overcome the 'Impossible Triangle' in Touch Sensing
中科大团队用一层梯度网格，撬动触觉传感器的“不可能三角”| Science Advances
A team from the University of Science and Technology of China, in collaboration with several international universities, has developed a flexible tactile sensor called GriFIPS that addresses the 'impossible triangle' challenge in touch sensing. The sensor uses a gradient grid structure and a liquid metal composite electrode to achieve high sensitivity, wide linear range, and stable performance. It is capable of detecting pressures from 2.1 Pa to 360 kPa, covering the full range of human touch perception. The sensor is integrated into a dexterous robotic hand for tasks such as object recognition, texture classification, and force sensing, demonstrating its potential for real-world applications.
This research introduces a novel gradient grid structure that enables a flexible tactile sensor to overcome the 'impossible triangle' challenge in touch sensing by decoupling sensitivity, linearity, and range through a dual-scale mechanism. The design is validated with experimental results covering a wide range of tactile pressures and demonstrates robustness in real-world applications.
Source: RoboSpeak — WeChat · Read original article ↗
English translation unavailable. Choose the original article to read it.
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
test samples
1,400
View original evidence
在涵盖铁块、海绵、软硬木材、泡沫、织物、混凝土等14种不同材质物体识别实验中， 1400 组测试样本识别准确率达到99.86%。
Open source S14Reported accuracy
99.86%
View original evidence
在涵盖铁块、海绵、软硬木材、泡沫、织物、混凝土等14种不同材质物体识别实验中， 1400 组测试样本识别准确率达到99.86%。
Open source S14Reported accuracy
97.5%
Reported accuracy
99.72%
pressure range
200
View original evidence
150kPa压力下完成 6000 次循环加载，信号输出保持稳定；累计 2000 次 50% 幅度拉伸循环之后，200kPa 压力下的传感响应和初始状态基本保持一致；5个独立制备的器件，一致性表现良好。
Open source S13pressure stability
150
View original evidence
150kPa压力下完成 6000 次循环加载，信号输出保持稳定；累计 2000 次 50% 幅度拉伸循环之后，200kPa 压力下的传感响应和初始状态基本保持一致；5个独立制备的器件，一致性表现良好。
Open source S13
- paper: Available Artifact link Open source S21
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
What remains unknown
Not established in the collected evidence: Environment, Control, Data origin.
Reported performance applies to the described task. It does not establish general autonomy or deployment readiness.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
在涵盖铁块、海绵、软硬木材、泡沫、织物、混凝土等14种不同材质物体识别实验中， 1400 组测试样本识别准确率达到99.86%。
Open source S14
更换一颗全新传感器采集，不重新微调模型，识别准确率依旧保持97.5%，具备不错的跨器件泛化性。
Open source S15
结果显现， 依靠四路通道多维触觉信息，分类准确率 99.72%，显著优于76.81%-83.19%的单通道信号方案。
Open source S16
150kPa压力下完成 6000 次循环加载，信号输出保持稳定；累计 2000 次 50% 幅度拉伸循环之后，200kPa 压力下的传感响应和初始状态基本保持一致；5个独立制备的器件，一致性表现良好。
Open source S13
论文链接：https://www.science.org/doi/10.1126/sciadv.aej9537 END 工业机器人企业 埃斯顿自动化 | 埃夫特机器人 | 法奥机器人 | 越疆机器人 | 节卡机器人 | 松灵机器人 | 珞石机器人 | 阿童木机器人 | 极智嘉 | 海康机器人 | 翼菲科技 | 艾利特机器人 服务与特种机器人企业 亿嘉和 | 晶品特装 | 七腾机器人 | 史河机器人 | 普渡机器人 | 施罗德机器人 | 库犸科技MAMMOTION 人形机器人企业 优必选科技 | 宇树 | 云深处 | 星动纪元 | 伟景机器人 | 逐际动力 | 乐聚机器人 | 大象机器人 | 魔法原子 | 众擎机器人 | 帕西尼感知 | 赛博格机器人 | 数字华夏 | 傅利叶智能 | 天链机器人 | 开普勒人形机器人 | 灵宝CASBOT | 清宝机器人 | 浙江人形机器人创新中心 | 动易科技 | 智身科技 | PNDboti
Open source S21
Source：RoboSpeak — WeChat · mp.weixin.qq.com