磁致伸缩触觉传感器与织物表面纹理检测  

Magnetostrictive tactile sensor and fabric surface texture detection

作　　者：万丽丽[1,2] 王博文 袁进峰[3] WAN Lili;WANG Bowen;YUAN Jinfeng(State Key Laboratory of Reliability and Intelligence of Electrical Equipment,Hebei University of Technology,Tianjin 300130,China;School of Logistics Engineering,Tianjin Transportation Technical College,Tianjin 300110,China;Electronic Information and Automation Institute,Tianjin Light Industry Vocational Technical College,Tianjin 300350,China)

机构地区：[1]河北工业大学省部共建电工装备与智能化国家重点实验室,天津300130 [2]天津交通职业学院物流工程学院,天津300110 [3]天津轻工职业技术学院电子信息与自动化学院,天津300350

出　　处：《传感器与微系统》2023年第2期109-113,共5页Transducer and Microsystem Technologies

基　　金：国家自然科学基金资助项目(52077052);河北省自然科学基金资助项目(E2017202035)。

摘　　要：为了实现智能机器人操作的精细化,分析传感器与生物组织间的接触刺激关系,建立了振动信号作用下的磁致伸缩触觉传感器输出模型并进行输出特性分析。结果表明:磁致伸缩触觉传感器对频率为1~4 Hz的动态力的测量具有很好的测量精度。将设计的磁致伸缩触觉传感器检测系统检测织物纹理,提取输出信号的功率谱密度,应用反向传播(BP)神经网络算法进行分类,分类结果的平均准确率达93%。结果表明:研制的磁致伸缩触觉传感器检测系统能够有效地应用于织物表面纹理的分类识别。In order to realize the refinement of intelligent robot operation and analyze the contact stimulation relationship between sensor and biological tissue, the output model of magnetostrictive tactile sensor under the excitation of vibration is established and the output characteristics are analyzed.The results show that the magnetostrictive tactile sensor has good measurement precision for dynamic force with frequency of 1~4 Hz.The designed magnetostrictive tactile sensor detection system detects the fabric texture, extracts the power spectral density of the output signal, and applies back propagation(BP)neural network algorithm for classification.The average accuracy of classification results is 93 %.The results show that the developed magnetostrictive tactile sensor detection system can be effectively applied to the classification and recognition of fabric surface texture.

关 键 词：磁致伸缩 触觉传感器 振动模型 纹理检测 反向传播神经网络 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置] TH823[自动化与计算机技术—控制科学与工程]