Tech Briefs Magazine - October 2023 - MD-24

Without knowledge about the interaction
forces and the shape of the handled
object, the robot fingers would
not have any " feel of touch, " and objects
could easily slip out of the robot
hand's fingers or even be crushed if
they are fragile.
The study, led by Professor Kaspar
Althoefer of Queen Mary University
of London, presents the new L3
F-TOUCH - high-resolution fingertip
sensor - where L3 stands for
Light weight, Low-cost, wireLess communication.
The sensor can measure
an object's geometry and determine
the forces to interact with it.
Unlike other sensors that estimate
interaction forces via tactile information
acquired by camera images, the
L3 F-TOUCH measures interaction
forces directly, achieving higher measurement
accuracy.
" In contrast to its competitors that
estimate experienced interaction
forces through reconstruction from
camera images of the deformation of
their soft elastomer, the L-3 F-TOUCH
measures
interaction forces directly
through an integrated mechanical suspension
structure with a mirror system
achieving higher measurement accuracy
and wider measurement range.
The sensor is physically designed to
decouple force measurements from
geometry information. Therefore,
the sensed three-axis force is immune
from contact geometry compared to
its competitors. Through embedded
wireless communications, the sensor
also
outperforms
competitors
with
regards to integrability with robot
hands. " said Professor Althoefer.
When the sensor touches the surface,
a compact suspension structure
enables the elastomer - a rubber-like
material that deforms to measure
high-resolution contact geometry exposed
to an external force - to displace
upon contact. To make sense of
this data, the elastomer's displacement
is tracked by detecting the movement
of a special marker, a so-called ARTag,
allowing us to measure contact forces
along the three major axes (x, y, and
z) via a calibration process.
" We will focus our future work on
extending the sensor's capabilities to
measure not only force along the
three major axes but also rotational
forces such as twist, which could be
experienced during screw fastening
while remaining accurate and compact.
These advancements can enable
the sense of touch for more dynamic
and agile robots in manipulation
tasks, even in human-robot interaction
settings, like for patient rehabilitation
or physical support of the elderly. "
added Professor Althoefer.
This breakthrough could pave the
way for more advanced and reliable
robotics in the future, as with the L3
F-TOUCH sensor, robots can have a
sense of touch, making them more
capable of handling objects and performing
complex manipulation tasks.
For more information, contact Melissa
Bradshaw at m.bradshaw@qmul.ac.uk.
Available on Demand!
Webinar
Webinar Series: How to Control Robots with Gestures
This two-part Webinar series explores how to train and deploy deep learning models and integrate
them into the Robot Operating System (ROS2) with cutting-edge, cost-optimized programmable
logic and system-on-a-chip (SoC) solutions.
Each 60-minute Webinar uses a fun robotics application to demonstrate the development steps required
to create the foundation designs, as well as how to augment these designs with hardware accelerators.
Part I: Training and Deploying a Classification Model
Part II: Integrating a Model into ROS2
Speakers:
Mario Bergeron
Machine Learning Specialist,
Avnet
Bryan Fletcher
Product Line Manager,
AMD
Please visit www.techbriefs.com/webinar370
24
MD Webinar Ad 0823.indd 1
Motion Design, October 2023
7/11/23 10:31 AM

http://www.techbriefs.com/webinar370

Tech Briefs Magazine - October 2023

Table of Contents for the Digital Edition of Tech Briefs Magazine - October 2023