Instrumentation & Measurement Magazine 24-5 - 28

Optical Force/Tactile Sensors for
Robotic Applications
Marco Costanzo and Salvatore Pirozzi
N
owadays, robotic systems use tactile sensing as a
key enabling technology to implement complex
tasks. For example, manipulation and grasping
problems strongly depend on the physical and geometrical
characteristics of the objects; in fact, objects may be deformable
or change their shape when in contact with the robot or
the environment. For this reason, often, robots end effectors
are equipped with sensorized fingers which can estimate the
objects' features, forces, and contact locations. This is useful in
a safe and efficient physical Human-Robot Interaction (pHRI)
to perform cooperation and co-manipulation tasks while limiting
damage from accidental impacts.
Currently, the manipulation abilities of modern robots are
far from human dexterous manipulation skills. A grasping device
should be able to firmly grasp any kind of object without
damaging it. This can be done by controlling the grasping force
which again requires accurate estimates of contact forces and
moments. Tactile sensors are of paramount importance in dexterous
robotic manipulation, and they should be able to detect
gross and incipient slippage and measure friction on contact
with an object. Research in [1] surveys the field of tactile sensing
from this point of view.
The use of tactile sensors in robotics is not limited to manipulation
tasks. In [2], the authors use tactile sensing to recognize
the nature of unknown surfaces by exploiting bimorph piezoceramic
actuators and sensors to stimulate a surface and
measure the response. Research in [3] considers the capacitive
variations in commercial piezoresistive transducers to sense
applied pressure for hand exoskeletons. Instead, [4] proposes
tactile feedback for human-computer interaction with an emphasis
on foot-based interfaces. Tactile data are also used to
solve classification problems such as object recognition; as an
example, [5] investigates tactile data as time sequences to solve
object recognition problems.
This paper reviews the evolution of the tactile sensing technology
developed at the Robotics Lab of the University of
Campania Luigi Vanvitelli. The paper first describes the hardware
technology and its evolution in the last decade. Then, the
work shows how, by means of the tactile measurements, it is
possible to reconstruct the contact wrench through an Artificial
Neural Network (ANN) and/or to estimate the shape of an
object. Finally, some advanced robotic applications are shown.
By using force/torque estimation it is possible to both automatically
choose the grasp force that avoids the slippage and
control the in-hand sliding motion of the object between the
fingers to change the relative pose between the fingers and the
grasped object. Instead, by using the shape recognition ability,
it is possible to estimate the current shape of a grasped cable to
correctly align the cable tip to the hole of a switchgear in an autonomous
assembly task.
The Sensing Technology Evolution
The idea of designing and developing, in our laboratories,
a tactile sensor based on optoelectronic technology dates to
about a decade ago, within the European research project DEXMART.
During recent years, the evolution of optoelectronic
devices and our experience in the field allowed us to optimize
our prototypes, achieving in the latest versions a high measurement
performance and a high mechatronic integration
level. The working principle (reported the first time in [6]) is
based on the design of a deformable layer to be suitably assembled
with a discrete number of optoelectronic sensing devices,
with the objective to transduce the external contacts into deformations
measured by optical sensible points (typically called
" taxels " in literature). The sensing points, positioned below
the deformable layer, provide a " tactile map " corresponding to
spatially distributed information about the contact. Based on
application task, the tactile map can be used to reconstruct contact
properties, e.g., contact force, contact torque, object shape.
The first version [6] comprised only two layers: an optoelectronic
layer (i.e., a Printed Circuit Board-PCB) and a
deformable one. Each sensing point was realized by using a
This work was supported by the European Commission within H2020 REFILLS
Project (no. 731590) and H2020 REMODEL Project (no. 870133).
28
IEEE Instrumentation & Measurement Magazine
1094-6969/21/$25.00©2021IEEE
August 2021
Instrumentation & Measurement Magazine 24-5

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 24-5
