Page 81 - Read Online

Page 2 of 35 Kulkarni et al. Soft Sci. 2025, 5, 12 https://dx.doi.org/10.20517/ss.2023.51

INTRODUCTION
Robotic systems are widely used in industrial settings such as manufacturing , construction , retail ,
[2]
[3]
[1]
[5]
[4]
agriculture , and healthcare . Robots can improve product quality, reduce labor costs, and, most
importantly, increase human safety in hazardous industrial settings . Robotic systems can carry out
[6]
inspection and monitoring on job sites as they allow for terrain navigation, automated detection of defects,
and data acquisition . These systems reduce the risk of human injury caused by heavy equipment or
[7]
[8]
hazardous materials, including chemicals or industrial waste . Robotic devices can improve healthcare via
[11]
[9]
[10]
efficient drug delivery , minimally invasive surgeries , and rehabilitation wearables . Robots can reduce
risks to humans and improve performance in complex or otherwise unsafe environments. In this paper, we
explore the advantages of considering the material composition of robots, specifically soft materials, when
designing for tasks in extreme environments.
The field of soft robotics research has rapidly evolved in the last decades after pioneering works in the 1990s
through the early 2010s . However, pneumatic soft actuators were first introduced in 1950 when
[12]
[14]
McKibben developed a soft braided actuator for an orthotic appliance . McKibbens muscles , otherwise
[13]
called pneumatic artificial muscle (PAM) actuators, consist of a mesh-constrained elastomeric bladder that
increases longitudinal stiffness to expand radially and contract linearly . More generally, soft robots are
[15]
[16]
composed of low-modulus materials including polymers, elastomers, or gels . These materials provide
flexibility , resistance to brittle fracture , biostability , and self-healing capabilities . These properties
[19]
[17]
[18]
[19]
allow soft robot structures to withstand highly variable conditions and may fill performance gaps that
traditional machines cannot accommodate.
In this paper, we aim to compare the utility of soft robots in specific environments to more traditional, rigid
robots (Section “COMPARING RIGID AND SOFT ROBOT PROPERTIES”). This review will outline
current state-of-the-art soft robotic actuators, accompanying sensors, and control methods for operation
specifically within challenging environments [Figure 1]. These environments include the human body
(Section “SOFT ROBOTS INSIDE THE HUMAN BODY”), marine environments (Section “SOFT
ROBOTS FOR MARINE ENVIRONMENTS”), space exploration (Section “SOFT ROBOTS FOR SPACE
EXPLORATION”), and search and rescue sites including confined spaces (Section “SOFT ROBOTS FOR
SEARCH, RESCUE, AND CONFINED SPACES”). Design methods using topology optimization, control
system strategies, and considerations for end users of potential commercial and educational products
represent areas of growth in soft robotics research (Section “EXPANDING THE USE OF SOFT ROBOTS
IN EXTREME ENVIRONMENTS”). Altogether, this review presents strategies for developing robotic
devices in extreme environments and opportunities for the future of this field.

COMPARING RIGID AND SOFT ROBOT PROPERTIES
Rigid and soft robots have properties that make them effective in different environments. Traditional
robotic systems have decades of development to support their application in many fields. Control systems
for rigid robots can be programmed to perform multiple actions and can be accurately modeled through
[24]
kinematic and dynamic models . Soft robots have been developed for impact resistance which makes them
advantageous for more unpredictable environments . However, most soft robotic actuators require
[25]
changes to the structure of the actuator to perform different actions . This is because soft actuators usually
[26]
perform a single action and need to be constructed with multiple parts to perform more than one type of
[26]
motion . The control and modeling of soft robots are challenging as these structures consist of nonlinear
[27]
systems that are challenging to represent mathematically due to material nonlinearities . In terms of
physical strength, soft robots are prone to buckling due to their soft material composition which may
[26]
prevent them from carrying heavy loads . Thus, rigid robots may be more advantageous when precise