Exploring Actuation Strategies in Soft Robotics

Keying Qu

Art And Technology, Central Academy of Fine Arts, Beijing, China

Keywords: Soft Robotics, Flexible Materials, Robot Control System.

Abstract: Soft robots are increasingly attracting attention due to their flexibility, safety, and high adaptability in complex

environments. Since drive technology plays an extremely important core role in determining the performance

and functionality of soft robots, conducting a systematic review of existing drive technologies is crucial for

advancing this field. This paper provides an overview of the four major categories of soft robot drive

technologies: pneumatic-hydraulic drives, electromagnetic drives, chemical drives, and hybrid drive systems.

For each category, this paper analyzes representative studies to summarize their working principles,

application scenarios, and performance characteristics. By comparing these drive technologies across metrics

such as energy efficiency, controllability, and integration potential, the paper identifies their respective

advantages and limitations. This paper aims to provide engineers and researchers developing high-

performance soft robots with a comprehensive reference on drive technologies and highlights potential future

research directions in areas such as material innovation, multimodal control, and system integration.

1 INTRODUCTION

In recent years, with the rapid development of bionic

engineering, flexible materials science, and

intelligent control technologies, soft robots have

emerged as a prominent research focus in the field of

robotics. Compared with traditional rigid robots, soft

robots are primarily inspired by nature and are

characterized by functional diversity, high

adaptability, and the ability to perform multiple

complex tasks simultaneously (Yasa et al., 2023).

They exhibit broad application prospects in areas

such as biomedicine, wearable devices, and

operations in complex environments.

The initial development of soft robots largely

relied on traditional actuation methods such as

pneumatic and hydraulic systems, with structural

designs often inspired by biological organisms like

octopuses and worms (Kim, Laschi, & Trimmer,

2013). In recent years, researchers have begun

exploring hybrid actuation systems that integrate

magnetic, electrical, and chemical stimulus-

responsive mechanisms, aiming to achieve higher

degrees of freedom and more precisely controllable

soft actuation (Ebrahimi et al., 2021). Constructed

primarily from flexible materials that mimic the

https://orcid.org/0009-0006-6236-3140

properties of soft-bodied organisms, soft robots are

capable of navigating freely through dynamic,

confined, and irregular environments (Wang et al.,

2024). As one of the core technologies enabling the

functionality of soft robots, actuation methods

directly determine their locomotion patterns, load-

bearing capacity, operational precision, and response

speed, and thus play a critical role in enhancing

overall robotic performance. The effectiveness of an

actuation strategy not only influences the

fundamental capabilities of soft robots but also

directly impacts their applicability in the real world.

For instance, in medical catheters or minimally

invasive surgical instruments, actuation systems must

be compact, highly controllable, and durable

(Cianchetti et al., 2018). Current mainstream

actuation methods include pneumatic, hydraulic, and

magnetic actuation. However, they still face several

challenges in terms of controllability, energy

efficiency, and system stability. These limitations

become even more pronounced in the face of

emerging requirements such as actuation within

highly flexible materials, power delivery for wearable

systems, and the miniaturization of soft robots. As a

result, the development of diverse and high-

performance actuation strategies has become a focal

point of research in the field (Walker et al., 2020).

Qu, K.

Exploring Actuation Strategies in Soft Robotics.

DOI: 10.5220/0014362000004718

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 2nd International Conference on Engineering Management, Information Technology and Intelligence (EMITI 2025), pages 511-516

ISBN: 978-989-758-792-4

511

Although several recent reviews have focused on

specific actuation technologies within the field, there

remains a lack of systematic cross-comparative and

integrative analyses of diverse actuation principles.

This paper aims to provide a comprehensive review

of current mainstream actuation methods for soft

robots from the following perspectives: gas/liquid-

based actuation, electromagnetic/magnetic actuation,

chemically driven actuation, and the development of

emerging actuation techniques. By systematically

outlining the research landscape and identifying the

current challenges in soft robotic actuation, this

review seeks to offer theoretical guidance for future

research and development in soft robotic actuation

systems.

2 PNEUMATIC AND

HYDRAULIC ACTUATION IN

SOFT ROBOTICS

Pneumatic and hydraulic actuation are among the

earliest and most widely applied methods in soft

robotic systems. These approaches rely on the

injection or withdrawal of gas or liquid into a sealed

flexible chamber to induce internal pressure changes,

which in turn lead to structural deformation and

enable specific movement functions. This actuation

mechanism closely resembles the behavior of soft-

bodied organisms in nature, such as octopuses and

worms, and is therefore commonly used in the design

of bioinspired soft structures.

2.1 Pneumatic Actuation Application

One representative example is the McKibben

pneumatic actuator (MPA), a type of soft artificial

muscle composed of a rubber bladder encased in a

braided mesh sleeve. When compressed air is pumped

into the bladder, it contracts in length and expands in

diameter, mimicking the contraction behavior of

biological muscles. Researchers have enhanced the

functionality of MPAs by applying elastic adhesive

coatings to enable bending motion, and based on this

design, they have developed multi-segmented soft

robotic arms capable of performing tasks such as

object grasping and fine manipulation (Kan et al.,

2025).

Bertrand Tondu et al. developed a static model of

the basic McKibben muscle based on three key

parameters: the initial braid angle, the initial muscle

length, and the initial muscle radius. The model also

incorporates a three-parameter friction model for the

muscle fiber itself, and has been shown to perform

effectively under both isometric and isotonic

contraction conditions (Tondu & Lopez, 2002).

This type of actuation features a simple structure

that is easy to manufacture and maintain. By

mimicking the contraction behavior of biological

muscles, it offers good compliance and biomimetic

performance, along with a high actuation force

capable of supporting heavy loads. However, it

suffers from limited control precision due to

sensitivity to pressure fluctuations. Additionally, both

inflation and deflation processes require time,

resulting in relatively slow response speeds and

making it difficult to achieve high-frequency

motions.

2.2 Hydraulic Actuation Application

Hydraulic actuation, which operates on principles

similar to pneumatic systems, utilizes incompressible

fluids to deliver higher output forces, making it

suitable for applications requiring strong load-

bearing capabilities. Qinlin Tan et al. proposed the

use of rigid structural components to strategically

reinforce otherwise omnidirectionally flexible soft

actuators, significantly improving their load capacity

and actuation precision. Their hybrid soft-rigid multi-

joint leg design features quasi-linear motion and force

characteristics while preserving excellent passive

impact compliance through the inherent flexibility of

soft actuators. The team also developed a novel

valveless hydraulic actuation system incorporating a

peristaltic pump, enabling a compact, lightweight,

and untethered underwater crawling robot prototype.

The robot demonstrated a payload-to-weight ratio of

5:1 and supported multiple gait modes (Tan et al.,

2021).

This actuation structure provides a strong driving

force, making it suitable for complex terrains and

heavy-load scenarios. It enables smoother and more

stable motion and offers higher precision in

propulsion control. The system also exhibits overall

good compliance and low material cost. However,

hydraulic systems are mechanically complex,

requiring precise pump control and effective sealing,

and they are prone to fluid leakage, demanding high

standards of system stability and reliability.

2.3 Analysis of Advantages and

Disadvantages

Due to their high actuation force and flexibility,

pneumatic and hydraulic systems have played a

crucial role in the field of soft robotics, particularly in

EMITI 2025 - International Conference on Engineering Management, Information Technology and Intelligence

512

task-oriented systems such as soft arms, grippers, and

mobile platforms. Nevertheless, these actuation

methods typically rely on external equipment (e.g.,

compressors or air pumps), resulting in high system

complexity and slow response speeds. Such

limitations hinder their applicability in future-

oriented scenarios that require portability,

wearability, or miniaturization. As a result, although

fluidic actuators currently dominate applications in

soft robotics, their development is gradually shifting

toward system integration, intelligent control, and

material optimization.

3 ELECTROMAGNETIC

ACTUATION IN SOFT

ROBOTICS

Electromagnetic actuation refers to the use of electric

fields, magnetic fields, or electro-thermal energy

conversion to induce physical deformation in actuator

materials, thereby enabling motion in soft robots.

This category of actuators typically does not require

bulky external pump systems, offering advantages

such as compact structure and fast response speed. As

a result, electromagnetic actuation has attracted

significant attention in applications involving

miniaturization, wearability, and bioinspired

robotics.

Dielectric elastomers, which are electrostatically

actuated by applying voltage across compliant

electrodes and dielectric polymers, represent a

prominent class of actuators in this domain. Owing to

their lightweight, geometric flexibility, cost-

effectiveness, and rapid response, dielectric

elastomer actuators (DEAs) can be configured into a

variety of shapes, making them highly promising for

artificial muscle-like actuation. To date, soft actuators

such as electroactive polymer actuators, shape

memory alloy actuators, shape memory polymer

actuators, and fluidic actuators (hydraulic or

pneumatic) have opened up numerous possibilities in

areas such as artificial muscles, biomimetic robotics,

and human–machine interfaces (Youn et al., 2020).

3.1 Application

In 2020, Christoph Keplinger proposed a novel circuit

design that eliminates the need for any high-voltage

sensing components, thereby enabling the use of off-

the-shelf components to create simple and cost-

effective circuits. This design enabled the

synchronized sensing and actuation of a range of

electrostatic transducers, achieving precise

displacement estimation with an error margin of less

than 4%. Furthermore, the circuit was developed into

a compact and portable system that integrates high-

voltage actuation, sensing, and computation, serving

as a prototype for wireless, multifunctional soft

robotic systems (Ly et al., 2021).

Magnetic-responsive actuation typically involves

embedding trace amounts of magnetic particles into

elastomeric materials, allowing the structure to

deform or move under the influence of an external

magnetic field. Dan Liu et al. developed a

magnetically actuated soft continuum microrobot for

intravascular microsurgery, which demonstrated

capabilities in both steering and locomotion. The soft

continuum microrobot was fabricated from a

composite of neodymium–iron–boron (NdFeB)

particles and polydimethylsiloxane (PDMS), with

diameters as small as 200 μm. Moreover, the robot’s

surface was coated with a hydrogel layer that not only

mitigated the adhesive forces between the miniature

components and the soft tip, but also reduced friction

between the robot and the substrate. The resulting

system is wirelessly operated and remotely

controllable, making it well-suited for tasks in

complex or confined environments, such as in

medical applications (Liu et al., 2022).

3.2 Analysis of Advantages and

Disadvantages

Overall, electromagnetically actuated soft robots

offer the advantage of remote control, enhancing

system deployment flexibility. Their structural

simplicity and actuation precision—determined by

magnetic field strength and particle distribution—

make them well-suited for miniature soft robotic

designs, particularly in applications requiring

compactness, high precision, or wearability.

Compared to pneumatic or hydraulic systems, they do

not require external pump sources, facilitating

integration and lightweight system design. However,

such robots typically generate relatively low

actuation forces, which limits their applicability.

Furthermore, the need for uniform magnetic materials

and highly precise magnetic field control systems

often results in high manufacturing costs.

4 CHEMICAL ACTUATION IN

SOFT ROBOTICS

Chemical actuation induces deformation or motion in

soft structures by releasing gases, heat, or other forms

Exploring Actuation Strategies in Soft Robotics

513

of physical energy through chemical reactions.

Without relying on external power sources or pump

systems, this type of actuation can achieve a certain

degree of "autonomous response." Kousuke

Moriyama et al. employed an enzymatic reaction as

the power source for soft robots, utilizing oxygen (O₂)

gas generated from the catalyzed decomposition of

hydrogen peroxide (H₂O₂) by the widely available

enzyme catalase to drive a pneumatic soft robot. The

generation of O₂ gas was influenced by the

concentration of catalase, the concentration of H₂O₂,

and the supply rate of H₂O₂. The study demonstrated

that catalase-catalyzed reactions can serve as an

effective power source for soft robotic systems,

highlighting a novel application prospect for

enzymatic processes. Moreover, enzymatic reactions

occur under mild conditions, reducing the risk of

overheating or damage to robotic components and

materials. These reactions also offer high

biocompatibility, and their byproducts are typically

non-toxic, enhancing safety during both operation

and disposal (Moriyama et al., 2024).

4.1 Application

In 2013, Dr. Robert F. Shepherd and colleagues

achieved rapid actuation of a soft robot using a high-

temperature chemical reaction. A computer-

controlled electric spark ignited a premixed

combination of methane and oxygen within the robot,

triggering combustion that pressurized the robot’s

pneumatic channels and enabled it to jump. The heat

generated by the explosion dissipated quickly,

allowing repeated jumps without damaging the robot.

This actuation approach offers self-contained

mobility without the need for external power sources

or controllers. Due to the high efficiency of chemical

reactions and their large instantaneous energy release,

this method demonstrates strong potential for high-

power, untethered soft robotic systems (Shepherd et

al., 2013).

4.2 Analysis of Advantages and

Disadvantages

Chemical actuation methods, characterized by the

absence of external equipment, structural simplicity,

and ease of deployment, are particularly well-suited

for single-use tasks in field environments, disaster

zones, and microscale settings. However, their

limited controllability, irreversibility of actuation

reactions, low repeatability, and high sensitivity to

environmental conditions pose significant challenges.

These factors hinder their broader adoption in soft

robotic systems that require high stability and

precision.

5 EMERGING HYBRID

ACTUATION STRATEGIES IN

SOFT ROBOTICS

As the limitations of traditional actuation methods—

such as pneumatic, electromagnetic, and chemical

systems—in terms of controllability, miniaturization,

and environmental adaptability become increasingly

apparent, researchers have begun exploring hybrid

actuation strategies that couple two or more

mechanisms. These emerging hybrid systems often

integrate different forms of energy input to achieve

multi-degree-of-freedom motion, high

responsiveness, and improved energy efficiency in

soft robotic actuation.

5.1 Application

Seyed M. Mirvakili et al. developed a simple

mechanism and design for actuating pneumatic

artificial muscles and soft robotic grippers without

relying on compressors, valves, or pressurized gas

tanks. Their actuation approach leverages

magnetically susceptible fluids undergoing a liquid–

gas phase transition, which generates internal

pressure within the artificial muscle. The volume

expansion during the phase transition creates

sufficient force to perform mechanical operations.

This actuation mechanism was integrated into both

McKibben-type artificial muscles and soft robotic

arms. The untethered McKibben artificial muscle

achieved up to 20% actuation strain within 10

seconds, with an energy density of 40 kJ/m³—

exceeding the peak strain and energy density of

skeletal muscle. The untethered soft robotic arm,

powered solely by two lithium-ion batteries,

demonstrated the capability to lift objects effectively

(Mirvakili et al., 2020).

5.2 Analysis of Advantages and

Disadvantages

Biohybrid soft robots integrate living cells or

biological tissues with artificial flexible structures,

achieving actuation and motion through spontaneous

cellular contractions or electrochemical stimulation.

These robotic systems exhibit a high degree of

biomimicry, possessing capabilities such as self-

healing, environmental adaptability, and a certain

EMITI 2025 - International Conference on Engineering Management, Information Technology and Intelligence

514

level of autonomy. While most conventional

actuators rely on external power sources, biohybrid

actuators can harvest energy directly from their

surroundings, enabling the design of untethered

systems without the need for bulky battery packs.

The ability of biohybrid actuators to adapt to

mechanical loads and self-repair makes robotic

devices more resilient to damage and capable of

regaining function after failure. Furthermore, the

capacity to recruit additional muscle fibers to

modulate actuator strength allows for force control

within compact system architectures. By leveraging

these capabilities, biohybrid robots at the macroscale

will be able to safely interact with various biological

organisms, adapt to mechanical stresses and

environmental conditions, and sustain themselves

through energy harvested directly from their

environment (Won et al., 2020).

6 NOVEL HYBRID ACTUATION

Novel hybrid actuation strategies overcome the

physical limitations of conventional mechanisms by

integrating multimodal energy conversion processes,

demonstrating significant advantages in biomimicry,

energy efficiency, and structural compactness. These

systems show great promise in cutting-edge

applications requiring untethered operation, remote

control, and adaptability to microscale environments.

However, their practical deployment still faces

considerable challenges. Current hybrid actuators

generally exhibit low technological maturity and high

manufacturing costs, while issues such as control

precision, response stability, and system longevity

remain in need of substantial improvement.

Therefore, despite their considerable potential, hybrid

actuation systems are presently focused on

fundamental research and prototype validation. Their

transition toward practical implementation will

require advances in novel materials, interdisciplinary

collaboration, and systematic optimization of

integration strategies.

7 RESEARCH BOTTLENECKS

AND FUTURE DEVELOPMENT

TRENDS

With the continued advancement of soft robotics

research, the diversity and functional complexity of

actuation systems have significantly increased.

However, transitioning from laboratory prototypes to

real-world applications still faces a series of common

technical bottlenecks. At the same time, emerging

technological innovations provide directional insights

for future developments in actuation systems.

7.1 Common Bottlenecks

First is response speed and latency issues. Certain

actuation methods, such as pneumatic/hydraulic

systems, shape memory alloys (SMAs), and chemical

actuators, suffer from slow response times and low

cycle efficiency, making them unsuitable for high-

frequency control or rapid movements. This

limitation is particularly critical in miniaturized

devices and medical scenarios, where actuation speed

is a key determinant of usability.

Second is high integration complexity. Most

current actuators still rely on external gas sources,

hydraulic pumps, power supplies, or magnetic coils,

resulting in bulky system designs and complex

control architectures. In multimodal actuation

systems, the diversity of actuation mechanisms,

energy requirements, and control logic across

different modules leads to significant integration

challenges. Achieving centralized power supply and

unified control of multiple modes remains a major

engineering obstacle in actuation system

development.

7.2 Future Perspectives

Future advances are expected in the development of

novel actuation materials that exhibit high flexibility,

strength, and self-healing capabilities. Breakthroughs

in deformable metals, conductive elastomers, and

liquid metals may provide essential material

foundations for constructing high-performance, high-

strength soft actuation systems. The integration of

artificial intelligence for intelligent control of

multimodal actuation systems will be crucial for

managing complex tasks. Through machine learning

and data fusion algorithms, real-time feedback,

predictive control, and energy management across

multiple actuators can be realized, improving the

overall system's responsiveness and coordination.

In addition, advanced manufacturing techniques

will play an important role in microstructural

fabrication and functional material printing for

actuators. Specifically, multi-material 3D printing

and rapid photopolymerization techniques will

support the construction of highly integrated and

complex actuation components, enabling a transition

from manual assembly to functionally integrated

structural manufacturing in soft robotics.

Exploring Actuation Strategies in Soft Robotics

515

8 CONCLUSIONS

This paper systematically explores the drive

mechanisms in the field of soft robotics from four

perspectives: pneumatic-hydraulic drive,

electromagnetic drive, chemical drive, and emerging

hybrid drive mechanisms. By analyzing

representative studies within each category, this paper

compares and contrasts the intrinsic implementation

mechanisms, adaptability, and performance of these

driving mechanisms.

Pneumatic-hydraulic drive systems excel in high-

stress output performance, enabling them to carry

heavier loads. However, they still suffer from the

drawbacks of bulky external components and limited

operational speed. Electromagnetic drive systems

demonstrate faster response times and greater

integration potential but are constrained by output

strength and energy requirements. Chemical drive

systems offer advantages such as integration,

compact size, and self-sufficiency but struggle with

precise control and lack repeatability. Hybrid drive

systems can combine multiple drive systems to

achieve multifunctionality, wireless control, and

adaptive soft robotics, but most hybrid soft robots are

still in the experimental stage and have not yet been

mass-produced.

This paper highlights the key challenges in the

development of soft robotics through classification

and comparative analysis, including energy

efficiency, system integration challenges, and control

precision. This review emphasizes the importance of

interdisciplinary collaboration, encompassing

materials science, control engineering, and

biomimetic design, to drive the transition of soft

robots from laboratory prototypes to practical

applications.

This review provides a theoretical foundation and

reference for future research on high-performance,

intelligent soft robot systems and may guide the

development of soft robots for biomedical

applications, wearable devices, and complex

environments.

REFERENCES

Cianchetti, M., Laschi, C., Menciassi, A., & Dario, P.

(2018). Biomedical applications of soft robotics. Nature

Reviews Materials, 3(6), 143–153.

Ebrahimi, N., Bi, C., Cappelleri, D. J., Ciuti, G., Conn, A.

T., Faivre, D., ... & Jafari, A. (2021). Magnetic

actuation methods in bio/soft robotics. Advanced

Functional Materials, 31(11), 2005137.

Kan, K., Goto, T., Naniwa, K., Nakanishi, D., Osuka, K., &

Sugimoto, Y. (2025). Realizing the bending motion of

a McKibben pneumatic actuator via elastic adhesive

coating. Journal of Robotics and Mechatronics, 37(1),

13–22.

Kim, S., Laschi, C., & Trimmer, B. (2013). Soft robotics:

A bioinspired evolution in robotics. Trends in

Biotechnology, 31(5), 287–294.

Liu, D., Liu, X., Chen, Z., Zuo, Z., Tang, X., Huang, Q., &

Arai, T. (2022). Magnetically driven soft continuum

microrobot for intravascular operations in microscale.

Cyborg and Bionic Systems.

Ly, K., Kellaris, N., McMorris, D., Johnson, B. K., Acome,

E., Sundaram, V., ... & Correll, N. (2021). Miniaturized

circuitry for capacitive self-sensing and closed-loop

control of soft electrostatic transducers. Soft Robotics,

8(6), 673–686.

Mirvakili, S. M., Sim, D., Hunter, I. W., & Langer, R.

(2020). Actuation of untethered pneumatic artificial

muscles and soft robots using magnetically induced

liquid-to-gas phase transitions. Science Robotics,

5(41), eaaz4239.

Moriyama, K., Nakao, S., Tsuji, M., Nakagawa, N., Satake,

T., & Johno, Y. (2024). Enzyme-powered soft robots:

Harnessing biochemical reaction for locomotion.

Biochemical Engineering Journal, 208, 109338.

Shepherd, R. F., Stokes, A. A., Freake, J., Barber, J.,

Snyder, P. W., Mazzeo, A. D., ... & Whitesides, G. M.

(2013). Using explosions to power a soft robot.

Angewandte Chemie International Edition, 52(10),

2892–2896.

Tan, Q., Chen, Y., Liu, J., Zou, K., Yi, J., Liu, S., & Wang,

Z. (2021). Underwater crawling robot with hydraulic

soft actuators. Frontiers in Robotics and AI, 8, 688697.

Tondu, B., & Lopez, P. (2002). Modeling and control of

McKibben artificial muscle robot actuators. IEEE

Control Systems Magazine, 20(2), 15–38.

Walker, J., Zidek, T., Harbel, C., Yoon, S., Strickland, F.

S., Kumar, S., & Shin, M. (2020, January). Soft

robotics: A review of recent developments of

pneumatic soft actuators. In Actuators, 9(1), 3. MDPI.

Wang, Y., Wang, Y., Mushtaq, R. T., & Wei, Q. (2024).

Advancements in soft robotics: A comprehensive

review on actuation methods, materials, and

applications. Polymers, 16(8), 1087.

Won, P., Ko, S. H., Majidi, C., Feinberg, A. W., & Webster-

Wood, V. A. (2020, September). Biohybrid actuators

for soft robotics: Challenges in scaling up. In Actuators,

9(4), 96. MDPI.

Yasa, O., Toshimitsu, Y., Michelis, M. Y., Jones, L. S.,

Filippi, M., Buchner, T., & Katzschmann, R. K. (2023).

An overview of soft robotics. Annual Review of

Control, Robotics, and Autonomous Systems, 6(1), 1–

29.

Youn, J. H., Jeong, S. M., Hwang, G., Kim, H., Hyeon, K.,

Park, J., & Kyung, K. U. (2020). Dielectric elastomer

actuator for soft robotics applications and challenges.

Applied Sciences, 10(2), 640.

EMITI 2025 - International Conference on Engineering Management, Information Technology and Intelligence

516