IEEE Robotics & Automation Magazine - December 2020 - 11

performance evaluation of a design
candidate is lacking.
3) Reliable and robust optimization
algorithms are yet to be developed.
It also highlights four areas where future
work may be done by modeling material and robot dynamics, simulation, and
optimization algorithms.
In the third article, Terryn et al. present a healable soft gripper and soft hand
whose flexible membranes are composed of Diels-Alder polymer networks. The authors argue that soft
robotic actuators based on their
approach are able to recover their performance after severe damage at room
temperature without the need for an
externally applied stimulus. They also
argue that this healing ability can help
reduce the over dimensioning of systems and optimize designs based on
function instead of on potential damaging conditions. The healing process can
take seconds or up to a week, depending on the location and extension of the
damage. As an extreme case, after cutting the actuator in two parts, it took
seven days to heal without the need for
any external heat stimulus, and the
actuator performance recovered, too.
Next, Pagoli et al. present the design
optimization of a robotic finger with a
sliding, rotating, and soft bending
mechanism. The novelty of the mechanism is its potential larger workspace
compared to prior robots with soft
appendages. The finger can rotate 300°
about its axis, and the mechanism consists of three actuators: two stepper
motors for sliding and rotating motion
along the longitudinal axis and an air
pump for bending. The design optimization relies on the nondominated sorted genetic algorithm II to find optimal
values for seven geometric parameters
of the soft joint and thus maximize the
bending angle and minimize joint
dimensions (i.e., length and diameter)
under a given pressure. A prototype
fabricated after the optimal design is
evaluated for different bending angles
and tip forces.
Finally, Lee at al. propose the design of
pouch motors with geometric constraints
that result in versatile and repeatable patterns of bending deformations with a

controllable force. The pouch motors rely
on the inflation of thin hollow pockets
instead of material stretching. The
authors describe a system of four pouch
motors with geometric constraints
allowing the actuator to produce a programmable bending deformation. They
suggest that the proposed pouch pattern
can be adapted to both soft robotic fingers and soft robotic joints. To demonstrate the efficacy of their approach,
they assembled a soft robotic arm with
three degrees of freedom and show that
it can perform pick-and-place operations and change its stiffness for safe
operation around humans. They also
build a larger soft robotic gripper for
grabbing larger objects.
We hope that the articles presented
in this special issue provide new or better insights in the field of soft robot
design and inspire the design of a new
generation of adaptive, compliant, and
reliable soft robots.

Surya G. Nurzaman, Monash University,
Malaysia. Email: surya.nurzaman@
monash.edu.
Liyu Wang, University of California
Berkeley. Email: liyu.wang@wadh.oxon
.org.
Fumiya Iida, University of Cambridge,
United Kingdom. Email: fi224@cam
.ac.uk.
Jeffrey Lipton, University of Washington,
Seattle. Email: jilipton@uw.edu.
Dario Floreano, Swiss Federal Institute
of Technology, Lausanne. Email: dario
.floreano@epfl.ch.
Daniela Rus, Massachusetts Institute of
Technology, Cambridge. Email: rus@
csail.mit.edu.

DECEMBER 2020

IEEE ROBOTICS & AUTOMATION MAGAZINE

http://butterflyhaptics.com

IEEE Robotics & Automation Magazine - December 2020

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - December 2020