TY - JOUR
T1 - A Small-Scale, Rat-Inspired Whisker Sensor for the Perception of a Biomimetic Robot
T2 - Design, Fabrication, Modeling, and Experimental Characterization
AU - Zhang, Yulai
AU - Yan, Shurui
AU - Wei, Zihou
AU - Chen, Xuechao
AU - Fukuda, Toshio
AU - Shi, Qing
N1 - Publisher Copyright:
© 1994-2011 IEEE.
PY - 2022/12/1
Y1 - 2022/12/1
N2 - In this article, we present a rat-inspired whisker sensor for a biomimetic robotic rat and demonstrate its superior tactile perception performance. In particular, the design, fabrication, modeling, and experimental characterization are presented. The sensor is fabricated using silicon-on-insulator (SOI) technology. Moreover, by modeling the whisker as a Euler-Bernoulli beam (EBB) using a support vector machine (SVM) algorithm, we develop the sensor's capability to identify object features. An experimental evaluation demonstrates the sensor's outstanding texture discrimination ability (with an accuracy of 88.3%) and excellent performance on contour reconstruction [with a 97.14% goodness of fit (GOF)]. In addition, by realizing the robotic rat's wall-following motion using the whisker sensor, many more features of a wall can be obtained with high accuracy. As a result, this work is, to the best of our knowledge, the first to achieve experimental progress in using a whisker sensor to control a small-scale robot on the order of a few tens of centimeters. Moreover, the whisker sensor extends this off-the-shelf SOI technology from micro to macro environments, giving it the potential for use in other miniature robots for tactile perception.
AB - In this article, we present a rat-inspired whisker sensor for a biomimetic robotic rat and demonstrate its superior tactile perception performance. In particular, the design, fabrication, modeling, and experimental characterization are presented. The sensor is fabricated using silicon-on-insulator (SOI) technology. Moreover, by modeling the whisker as a Euler-Bernoulli beam (EBB) using a support vector machine (SVM) algorithm, we develop the sensor's capability to identify object features. An experimental evaluation demonstrates the sensor's outstanding texture discrimination ability (with an accuracy of 88.3%) and excellent performance on contour reconstruction [with a 97.14% goodness of fit (GOF)]. In addition, by realizing the robotic rat's wall-following motion using the whisker sensor, many more features of a wall can be obtained with high accuracy. As a result, this work is, to the best of our knowledge, the first to achieve experimental progress in using a whisker sensor to control a small-scale robot on the order of a few tens of centimeters. Moreover, the whisker sensor extends this off-the-shelf SOI technology from micro to macro environments, giving it the potential for use in other miniature robots for tactile perception.
UR - http://www.scopus.com/inward/record.url?scp=85134202186&partnerID=8YFLogxK
U2 - 10.1109/MRA.2022.3182870
DO - 10.1109/MRA.2022.3182870
M3 - Article
AN - SCOPUS:85134202186
SN - 1070-9932
VL - 29
SP - 115
EP - 126
JO - IEEE Robotics and Automation Magazine
JF - IEEE Robotics and Automation Magazine
IS - 4
ER -