TABLE 3.

Comparison between different actuator structures.

Structure	Speed	Driven period	Feature	Advantage/limitation	Driving force	Test environment	Reference
BCF locomotion	850 mm/s (2.02 body length/s) (max)	5.46 Hz	Using the continuous rotation of a DC motor to pull the cables connected to both sides of the active tail segment	Using active tail segment and passive tail segment/exceed the predictions from a tailbeat frequency of approximately 2.5 Hz	Pulling cables	Laboratory	Berg et al. (2022)
BCF locomotion	217 mm/s (0.5 body length/s at depths of 0–18 m) (avg.)	0.9–1.4 Hz	The on-board capabilities of an untethered mobile underwater observatory	Tested at coral reefs 18 m depths/low swimming efficiency, limited depth range	Hydraulic actuation	Underwater (18 m)	Katzschmann et al. (2018); Katzschmann et al. (2016)
BCF locomotion	150 mm/s (0.44 body length/s) (avg.)	1.67 Hz	Using gas pump to reserve energy and having escape ability	Having the escape response/the gas delivery system is responsible for considerable resistive energy losses	Pneumatic actuator	Laboratory	Marchese et al. (2014)
BCF locomotion	37.2 mm/s (0.25 body length/s) (max)	0.75 Hz/5 kV	Only 4.4 g of weight and using DEAs as a main driver	The robot resembles real fish and displays a Strouhal number very close to that of living fish/the feature of the structure may change by the oxidation	DE actuator	Laboratory	Shintake et al. (2018)
BCF locomotion	300 mm/s (2 body length/s) (max)	14 Hz	Segmented caudal fin, achieved 70 Hz operating frequency	Peak frequency could be selected using fin design/external power supply and control electronics via tether	Solenoid actuator	Laboratory	Zhang et al. (2018)
BCF locomotion	1,020 mm/s (4 body length/s) (max)	15 Hz	Swims at 0.4 m/s and has a range of 9.1 km, swims at 1.0 m/s and has a range of 4.2 km (assuming a 10 Wh battery pack)	Low cost of transport/external control signal via tether and lacks control surfaces to adjust turns in yaw or changes in pitch	DC motor trans to lateral linear movement	Laboratory	Zhu et al. (2019)
MPF locomotion	51.9 mm/s (0.45 body length/s)	1 Hz/8 kV	Lightweight DE actuators, snailfish-like	Decentralized electronics and DE-driven flapping fins/needs high-voltage amplifier	DE actuator	Underwater (10,900 m)	Li et al. (2021)
MPF locomotion	64 mm/s (0.69 body length/s)	5 Hz/8 kV	Self-powered soft robots with high mobility, environmental tolerance, and long endurance	Untethered electronic fish, excellent environmental adaptability and disguising performance/need high-voltage amplifier	DE actuator	Laboratory	Li et al. (2017)
JetDrive	10 mm/s (max)	1.6 Hz/9 kV	Jellyfish-like	Low cost and easy assemble/low anti-interference performance	DE actuator	Laboratory	Cheng et al. (2019)
JetDrive	Untethered: 3.2 mm/s (avg.),7.1 mm/s (max)	0.2 Hz/7 kV	Jellyfish-like, low cost of transport (30), 2.7 h actuation (180 mAh)	Swim bladder equipped to provide buoyancy control/incapable of providing large forces	DE actuator	Laboratory	Christianson et al. (2019)
JetDrive	/	0.8 Hz (half-stroke)	Best performance, tentacle actuator-flap material Shore hardness composition of 30–30	Squeezes through narrow conduits, swims directionally by temporally offsetting tentacle actuation strokes on opposing sides of the robot/the horizontal motion of the jellyfish was not directly controlled	Hydraulic/pneumatic actuators	Underwater (5 m)	Frame et al. (2018)
JetDrive	Rigid, tethered	Rigid, tethered	Rigid bell with the dielectric elastomer actuator	Fast response and high capacity of payload external power supply and control electronics via tether/was not directly controlled	DE actuator	Laboratory	Godaba et al. (2016)
JetDrive	295 mm per release	15 s (inflate), 0.5 s (eject period)	Change the periodic conversion of slowly charged elastic potential energy into fluid kinetic energy, cephalopods’ propulsion	Short range, highly maneuverable/ significant viscous losses due to the sharp corners within the nozzle conduit	Hydraulic actuation	Laboratory	Wang et al. (2019c)
JetDrive	5.4 mm/s (avg.)	2 Hz/5 kV	Using hybrid silver nanowire networks, has large stretchability, low stiffness, high transmittance, and excellent conductivity	Transparent DEA and can achieve maximum area strain of 146% with only 3% hysteresis loss/jamming by the dragging force, tethered	DE actuator	Laboratory	Wang et al. (2022c)
JetDrive	Rigid, tethered	Rigid, tethered	A novel central flow-regulative duct encircled by three circumferential siphon actuation muscles	Bending range of over 180° and flow-restricting capability of up to 100%/tether was not directly controlled	Hydraulic actuation	Laboratory	Zhang et al. (2021)
Crawling	15 mm/s (still water)	/	An underwater legged robot with soft legs and a soft inflatable morphing body	Able to change body shape against hydraulic flow/lack resistance of the drag on the hydraulic tether	Hydraulic actuation	Laboratory	Ishida et al. (2019)
Crawling	10 mm/s (0.04 bldy length/s) (avg.)	2.52 s (avg.)	Five flexible legs actuated by 20 shape memory alloy (SMA) wires	A wide variety of possible motions/every actuator is a nonlinear dynamical system with a large amount of hysteresis	SMA actuation	Laboratory	Patterson et al. (2020)
Crawling	/	/	Rigid–soft hybrid multi-joint leg with quasi-linear motion range and force exertion	Using rigid structural components to reinforce the flexible soft actuators/complicated production	Hydraulic actuation	Laboratory	Tan et al. (2021)
Floating	2.83 mm/s	/	Untethered omnidirectional star-shaped swimming soft robot	Capable of moving with a variety of swimming gaits/disturb by flow	SMA actuation	Laboratory	Huang et al. (2021)
Floating	48 mm/s (avg.)	1.4 Hz	An untethered aquatic soft robot that performs frog-like rowing behaviors	Limbs can be replaced within seconds/lack of data on manufacturing accuracy	SMA actuation	Laboratory	Huang et al. (2022)
Crawling	5 cm/s (max)	1.4 s (avg.)	The complete absence of rigid parts makes it possible to replicate the high compliance and flexibility of the octopus arm	Pass through confined and unstructured spaces/the control of soft robots through distributed sensors is difficult	SMA springs or cables	Laboratory	Krieg et al. (2015)