Researching | Optical Fibre Bragg Based Sliding-tactile Sensing and Classification Training Method for Material Recognition

Acta Photonica Sinica, Volume. 53, Issue 2, 0206006(2024)

Optical Fibre Bragg Based Sliding-tactile Sensing and Classification Training Method for Material Recognition

Ruizhi PAN^1,2, Yan FENG^1,2、*, Hexiang LIU³, Haoxiang WANG^1,2, Hongpu ZHANG^1,2, Yinxiang ZHANG^1,2, and Hua ZHANG^1,2

Author Affiliations

¹School of Mechanical and Automotive Engineering，Shanghai University of Engineering Science，Shanghai 201620，China

²Shanghai Collaborative Innovation Center of Intelligent Manufacturing Robot Technology for Large Components，Shanghai 201620，China

³School of Advanced Manufacturing，Nanchang University，Nanchang 330038，China

show less

Abstract Get PDF(in Chinese)

Figures&Tables (15)

References (22)

Paper Information

Figures & Tables(15)

Fig. 1. FBG silicone rubber sensing unit

Download full size

View in Article

Fig. 2. Schematic of sliding mechanism

Download full size

View in Article

Fig. 3. Optical transport characteristics of the FBG

Download full size

View in Article

Fig. 4. Profile strain analysis

Download full size

View in Article

Fig. 5. Slider and surface material

Download full size

View in Article

Fig. 6. Three materials versus center wavelength

Download full size

View in Article

Fig. 7. Schematic of feature mapping process

Download full size

View in Article

Fig. 8. Confusion matrix

Download full size

View in Article

Fig. 9. Three-dimensional characteristic values of each material

Download full size

View in Article

Fig. 10. Number of iterations versus minimum classification error

Download full size

View in Article

Fig. 11. Material recognition step-by-step

Download full size

View in Article

Fig. 12. Photo of the system hardware

Download full size

View in Article

Fig. 13. Host computer interface

Download full size

View in Article

Table 1. Accuracy of SVM classification models for each data feature three times

View in Article

Table 1. Accuracy of SVM classification models for each data feature three times

Data sources Method	Direct classify λ_B1，λ_B1，…	General classify（Ave/Mid）	3-features classify（Max_diff/Range/Std）
15 cm/s slip	96.2%	79.5%	94.9%
10 cm/s slip	98.7%	97.4%	97.4%
5 cm/s slip	96.2%	88.5%	97.4%
Mixing of three speeds	55.1%	82.9%	93.6%

Table 2. Results of the tests
View table
View in Article
Table 2. Results of the tests
Material category Correct predictions/tests
Rough 12/12
PLA 12/12
800-grit sandpaper 10/12
Total 34/36

Tools

Get Citation

Copy Citation Text

Ruizhi PAN, Yan FENG, Hexiang LIU, Haoxiang WANG, Hongpu ZHANG, Yinxiang ZHANG, Hua ZHANG. Optical Fibre Bragg Based Sliding-tactile Sensing and Classification Training Method for Material Recognition[J]. Acta Photonica Sinica, 2024, 53(2): 0206006

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Fiber Optics and Optical Communications

Received: Aug. 1, 2023

Accepted: Sep. 14, 2023

Published Online: Mar. 28, 2024

The Author Email: FENG Yan (confirmfyan@163.com)

DOI:10.3788/gzxb20245302.0206006

Topics

laser devices and laser physics

Lasers and Laser Optics

laser manufacturing

Instrumentation, Measurement and Metrology