Abstract —In this paper, a novel wearable tactile sensor array
for fingertip motion recognition is proposed. The design of the
sensor is presented that the tactile sensor array using conductive
thread electrodes. The sensing points distributed in ring
structure, the main sensing point of which is in the center and 24
secondary sensing points in the outer ring. It is more suitable for
fingertip state recognition than the traditional rectangular array
method. The circuit is designed with flexible printed circuit
board, which is “T-type” and easy to assemble at the fingertips.
We classify all fingertip motion into eight categories and it is
validated that the properties of the sensor in an application for
fingertip motion recognition is well. We represent the array if
tactile information as a time series of features and use this as the
input for classifier. It is demonstrated that the tactile sensor is
high accuracy in identifying the categories of fingertip motion.
The result is proved that this wearable tactile sensor array is
stable performance and excellent flexibility. Furthermore, due to
its good performance, it can be easily integrated onto a curved
surface, such as robotic and prosthetic hands.
Keywords—Tactile; sensor array; wearable; fingertip
I. INTRODUCTION
Humans have a sensitive sense of touch, which allows them
to perceive some external environment or to respond to the
environment in a timely manner, such as the slippage of
manipulated object. As for robot, the tactile sensors in the
robotic fingertips are crucial to the humanoid operation.
Therefore, tactile sensor is one of the essential sensors for robot,
due to its various purposes such as determining object pose [1],
detecting slipping [2], recognizing gesture [3]. As for tactile
sensor, there are many kinds of construction, namely,
piezoresistance, capacitive sensor, piezoelectric sensor, visual
device, magnetic etc [4]. Piezoresistance or capacitive sensor
is a contact based device that depends on physical interaction
with the users. For example, [5] introduces a piezoresistive
sensor, constituted by a polymeric composite with
nanostructured spiky particles as filler. The planar sensor can
thus be modeled as a two-dimensional array of resistors whose
value decreases by increasing the applied pressure. Reference
[6]describe a novel solution of capacitive sensor in which the
dielectric is made using a thin layer of 3D fabric which is glued
to conductive and protective layers using techniques adopted
in the clothing industry. The FPCB of sensor is shaped as a
triangle hosting 12 sensors. It shows small hysteresis
integrated into the iCub. Vision based devices capture the
video streams for analysis to determine the hand motion. The
sensor reported in [7] consists of a CCD camera, LED lights, a
transparent acrylic plate and a deformable touchpad. This
paper clarify the mechanism of slippage between a fluid-type
1
Zhudong Huang is with the Institute of Robotics and Intelligent Systems,
Wuhan University of Science and Technology, Wuhan, 100084, China.
*Huasong Min is the corresponding author of this paper, and with the
Institute of Robotics and Intelligent Systems, Wuhan University of Science
and Technology, China. His research interests include embedded system and
intelligent robotics.
(corresponding author to provide e-mail: mhuasong@wust.edu.cn )
touchpad and contacted object. our sensor has demonstrated
that it can simultaneously obtain multiple types of tactile
information, including the contact force, moment, slippage
degree, shape, contact region, position and orientation of an
object in contact with the fluid-type touchpad. Micro-
fabricated tactile sensors [8] is a piezoresistance sensor gaining
importance for their application in bio-robotics. It is designed
as
array into
mm
3
. The obtained results
show that the sensory system has good metrological features,
in terms of linearity and sensitivity, which well match the
application in robotics. However, piezoresistance sensor is
more susceptible to the voltages added on them, while wiring
is more complicated. Capacitive sensor have some advantages,
as like simple structure, high sensitivity.
Tactile classification also has many mature algorithms at
home and abroad. For example, D.S. [9] introduces a texture
classification algorithm utilizing support vector machine
(SVM) classifier by using a three axis tactile sensor. Reference
[10] presents a novel tactile-array sensor for use in robotic
grippers based on flexible piezoelectric rubber. In this paper, a
method for classifying rigid and vision deformable objects
based on time series of features recorded from a palpation
procedure is proposed. They use dynamic time warping to
compare the similarity in the signal based on basic features
extracted from a series if tactile images with a k-nearest
neighbor classifier. Heyneman et al. [11] developed a slip
classification that two features are extracted from dynamic
tactile array sensors to discriminate between hand/object and
object/world slips. The advantage of this way is that it avoids
to get extensive traning with particular combinations of objects
and skin. Reference [12] introduces a classification method of
tactile patterns. The method is a two-stage algorithm, which
consists of the codebook generation and classification stages,
respectively. First, local neighbourhood structures are
extracted as the bases of the first layer of the codebook. Next,
the pressure distributions over the local neighbourhood
features are clustered for vector quantisation to create the sub-
layer codewords. The Support Vector Machine with the
intersection kernel is used for classifying the histogram-based
features based on the codebook.
The main purpose of this paper is to present a novel
wearable tactile sensor for fingertip motion recognition. In
order to obtain stable haptical information, a particular multi-
array tactile sensor is designed. The sensor uses professional
equipment for the calibration of force and direction. According
to the orientation and force size of each sensor of the tactile
sensor, it is easy to obtain the movement trend of fingertips.
2
Bin Fang, Fuchun Sun and Wenliang Zhang are all with the
Tsinghua National Laboratory for Information Science and Technology,
Department of Computer Science and Technology, Tsinghua University,
Beijing, 100084, China.
3
Xiaohui Hu is with the Institute of Semiconductors, Academy of Sciences,
Beijing 100083, China.
A Novel Wearable Tactile Sensor Array Designed for Fingertip
Motion Recognition
Zhudong Huang
1
, Fuchun Sun
2
, Huasong Min*
1
, Bin Fang
2
, Wenliang Zhang
2
and Xiaohui Hu
3
978-1-5386-3741-8/17/$31.00 © 2017 IEEE
Proceedings of the 2017 IEEE
International Conference on Robotics and Biomimetics
December 5-8, 2017, Macau SAR, China