The deep sea remains the largest unknown territory on Earth because it is so difficult to explore1,2,3,4. Owing to the extremely high pressure in the deep sea, rigid vessels5,6,7 and pressure-compensation systems8,9,10 are typically required to protect mechatronic systems. However, deep-sea creatures that lack bulky or heavy pressure-tolerant systems can thrive at extreme depths11,12,13,14,15,16,17. Here, inspired by the structure of a deep-sea snailfish15, we develop an untethered soft robot for deep-sea exploration, with onboard power, control and actuation protected from pressure by integrating electronics in a silicone matrix. This self-powered robot eliminates the requirement for any rigid vessel. To reduce shear stress at the interfaces between electronic components, we decentralize the electronics by increasing the distance between components or separating them from the printed circuit board. Careful design of the dielectric elastomer material used for the robot’s flapping fins allowed the robot to be actuated successfully in a field test in the Mariana Trench down to a depth of 10,900 metres and to swim freely in the South China Sea at a depth of 3,224 metres. We validate the pressure resilience of the electronic components and soft actuators through systematic experiments and theoretical analyses. Our work highlights the potential of designing soft, lightweight devices for use in extreme conditions.
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
All data are available from the corresponding authors upon reasonable request.
Finite element analysis was performed using ABAQUS.
Danovaro, R., Corinaldesi, C., Dell’Anno, A. & Snelgrove, P. V. R. The deep-sea under global change. Curr. Biol. 27, R461–R465 (2017).
Clarke, T. Robots in the deep. Nature 421, 468–470 (2003).
Yoerger, D. R., Jakuba, M., Bradley, A. M. & Bingham, B. Techniques for deep sea near bottom survey using an autonomous underwater vehicle. Int. J. Robot. Res. 26, 41–54 (2007).
Petersen, S. et al. News from the seabed—geological characteristics and resource potential of deep-sea mineral resources. Mar. Policy 70, 175–187 (2016).
Stachiw, J. D., Peters, D. & McDonald, G. Ceramic external pressure housings for deep sea vehicles. In IEEE Oceans Conf. https://doi.org/10.1109/OCEANS.2006.306971 (IEEE, 2006).
Błachut, J. & Smith, P. Buckling of multi-segment underwater pressure hull. Ocean Eng. 35, 247–260 (2008).
Kampmann, P., Lemburg, J., Hanff, H. & Kirchner, F. Hybrid pressure-tolerant electronics. In IEEE Oceans Conf. https://doi.org/10.1109/OCEANS.2012.6404828 (IEEE, 2012).
McPhail, S. D. Autosub6000: a deep diving long range AUV. J. Bionics Eng. 6, 55–62 (2009).
Umapathy, A. et al. Influence of deep-sea ambient conditions in the performance of pressure-compensated induction motors. Mar. Technol. Soc. J. 53, 67–73 (2019).
Bingham, N. Designing pressure-tolerant electronic systems. Unmanned Underwater Technology https://www.uutech.com/ptepaper/ (2013).
Kunzig, R. Perceptions of science. Deep-sea biology: living with the endless frontier. Science 302, 991 (2003).
Danovaro, R. et al. Deep-sea biodiversity in the Mediterranean Sea: the known, the unknown, and the unknowable. PLoS ONE 5, e11832 (2010).
Linley, T. D. et al. Fishes of the hadal zone including new species, in situ observations and depth records of Liparidae. Deep Sea Res. Part I 114, 99–110 (2016).
Yancey, P. H., Gerringer, M. E., Drazen, J. C., Rowden, A. A. & Jamieson, A. Marine fish may be biochemically constrained from inhabiting the deepest ocean depths. Proc. Natl Acad. Sci. USA 111, 4461–4465 (2014).
Wang, K. et al. Morphology and genome of a snailfish from the Mariana Trench provide insights into deep-sea adaptation. Nat. Ecol. Evol. 3, 823–833 (2019).
Hoving, H. J. T. & Haddock, S. H. D. The giant deep-sea octopus Haliphron atlanticus forages on gelatinous fauna. Sci. Rep. 7, 44952 (2017).
Kawabata, T. et al. Evaluation of the bioactivities of water-soluble extracts from twelve deep-sea jellyfish species. Fish. Sci. 79, 487–494 (2013).
Picardi, G. et al. Bioinspired underwater legged robot for seabed exploration with low environmental disturbance. Sci. Robot. 5, eaaz1012 (2020).
Stuart, H. S., Wang, S., Khatib, O. & Cutkosky, M. R. The Ocean One hands: an adaptive design for robust marine manipulation. Int. J. Robot. Res. 36, 150–166 (2017).
Stuart, H. S., Wang, S. & Cutkosky, M. R. Tunable contact conditions and grasp hydrodynamics using gentle fingertip suction. IEEE Trans. Robot. 35, 295–306 (2019).
Renda, F., Giorelli, M., Calisti, M., Cianchetti, M. & Laschi, C. Dynamic model of a multibending soft robot arm driven by cables. IEEE Trans. Robot. 30, 1109–1122 (2014).
Lum, G. Z. et al. Shape-programmable magnetic soft matter. Proc. Natl Acad. Sci. USA 113, E6007–E6015 (2016).
Laschi, C. Soft robot arm inspired by the octopus. Adv. Robot. 26, 709–727 (2012).
Weymouth, G. & Triantafyllou, M. S. Ultra-fast escape of a deformable jet-propelled body. J. Fluid Mech. 721, 367–385 (2013).
Giorgio-Serchi, F., Arienti, A. & Laschi, C. Underwater soft-bodied pulsed-jet thrusters: actuator modeling and performance profiling. Int. J. Robot. Res. 35, 1308–1329 (2016).
Serchi, F. G., Lidtke, A. K. & Weymouth, G. A soft aquatic actuator for unsteady peak power amplification. IEEE–ASME Trans. Mechatron. 23, 2968–2973 (2018).
Vogt, D. M. et al. Shipboard design and fabrication of custom 3D-printed soft robotic manipulators for the investigation of delicate deep-sea organisms. PLoS ONE 13, e0200386 (2018).
Phillips, B. T. et al. A dexterous, glove-based teleoperable low-power soft robotic arm for delicate deep-sea biological exploration. Sci. Rep. 8, 14779 (2018).
Galloway, K. C. et al. Soft robotic grippers for biological sampling on deep reefs. Soft Robot. 3, 23–33 (2016).
Chen, Y. et al. Controlled flight of a microrobot powered by soft artificial muscles. Nature 575, 324–329 (2019).
Li, T. et al. Fast-moving soft electronic fish. Sci. Adv. 3, e1602045 (2017).
Christianson, C., Goldberg, N. N., Deheyn, D. D., Cai, S. & Tolley, M. T. Translucent soft robots driven by frameless fluid electrode dielectric elastomer actuators. Sci. Robot. 3, eaat1893 (2018).
Ji, X. et al. An autonomous untethered fast soft robotic insect driven by low-voltage dielectric elastomer actuators. Sci. Robot. 4, eaaz6451 (2019).
Acome, E. et al. Hydraulically amplified self-healing electrostatic actuators with muscle-like performance. Science 359, 61–65 (2018).
Chiba, S. et al. Electroactive polymer ‘artificial muscle’ operable in ultra-high hydrostatic pressure environment. IEEE Sens. J. 11, 3–4 (2011).
Yuk, H. et al. Hydraulic hydrogel actuators and robots optically and sonically camouflaged in water. Nat. Commun. 8, 14230 (2017).
Marchese, A. D., Onal, C. D. & Rus, D. Autonomous soft robotic fish capable of escape maneuvers using fluidic elastomer actuators. Soft Robot. 1, 75–87 (2014).
Morin, S. A. et al. Camouflage and display for soft machines. Science 337, 828–832 (2012).
Li, S., Vogt, D. M., Rus, D. & Wood, R. J. Fluid-driven origami-inspired artificial muscles. Proc. Natl Acad. Sci. USA 114, 13132–13137 (2017).
Wehner, M. et al. An integrated design and fabrication strategy for entirely soft, autonomous robots. Nature 536, 451–455 (2016).
Aubin, C. A. et al. Electrolytic vascular systems for energy-dense robots. Nature 571, 51–57 (2019).
Sandoval, J. A., Jadhav, S., Quan, H., Deheyn, D. D. & Tolley, M. T. Reversible adhesion to rough surfaces both in and out of water, inspired by the clingfish suction disc. Bioinspir. Biomim. 14, 066016 (2019).
Wang, Y. et al. A biorobotic adhesive disc for underwater hitchhiking inspired by the remora suckerfish. Sci. Robot. 2, eaan8072 (2017).
Koh, S. J. et al. Mechanisms of large actuation strain in dielectric elastomers. J. Polym. Sci. B 49, 504–515 (2011).
Zhao, X. & Suo, Z. Method to analyze programmable deformation of dielectric elastomer layers. Appl. Phys. Lett. 93, 251902 (2008).
Ma, Z. et al. Thermoplastic dielectric elastomer of triblock copolymer with high electromechanical performance. Macromol. Rapid Commun. 38, 1700268 (2017).
Passaglia, E. & Martin, G. M. Variation of glass temperature with pressure in polypropylene. J. Res. Natl Bur. Stand. A 68, 273–276 (1964).
Jones Parry, E. & Tabor, D. Effect of hydrostatic pressure on the mechanical properties of polymers: a brief review of published data. J. Mater. Sci. 8, 1510–1516 (1973).
Yang, C., Gao, X. & Luo, Y. End-block-curing ABA triblock copolymer towards dielectric elastomers with both high electro-mechanical performance and excellent mechanical properties. Chem. Eng. J. 382, 123037 (2020).
Joung, T. et al. A study on the pressure vessel design, structural analysis and pressure test of a 6000 m depth-rated unmanned underwater vehicle. Ships Offshore Struct. 3, 205–214 (2008).
Liang, C. et al. A study of diving depth on deep-diving submersible vehicles. Int. J. Press. Vessels Piping 75, 447–457 (1998).
Hernes, M. & Pittini, R. Enabling pressure tolerant power electronic converters for subsea applications. In 13th European Conference on Power Electronics and Applications https://ieeexplore.ieee.org/document/5278719 (IEEE, 2009).
Thiede, C. et al. An overall pressure tolerant underwater vehicle: DNS Pegel. In IEEE Oceans Conf. https://ieeexplore.ieee.org/document/5278313 (IEEE, 2009).
We acknowledge the support of the following: National Key R&D Program of China 2017 YFA0701100, National Natural Science Foundation of China (92048302, 11822207, 11525210, 91748209, 21636008, 41876179, 52075476, 91948303 and 21676244), Zhejiang Provincial Natural Science Foundation of China (R18A020004), China Postdoctoral Science Foundation (2020M671820), Zhejiang University, Zhejiang Lab Research Funds (113006-Z31801, 115002-AA2003), Fundamental Research Funds for the Central Universities, and Dr Li Dak Sum & Yip Yio Chin Fund for Stem Cell and Regenerative Medicine. The mechatronic equipment is supported by Shenzhen Dinghai Advanced Material Technology and Shaoxing Haivoo Power Supply Technology. We thank the crews of cruises from Guangzhou Marine Geological Survey, Westlake University and Shanghai Ocean University.
T.L. is applying for patents related to the described work. The other authors declare that they have no competing interests.
Peer review information Nature thanks Kaitlyn Becker, Cecilia Laschi, Brennan Philips and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
a, The soft robot was mounted onto a deep-sea lander to reach the bottom of the Mariana Trench (11.33° N, 142.19° E, depth 10,900 m). b, c, The soft robot has an embedded high-voltage amplifier (b) and a 2,500 mA h lithium-ion battery (c), generating the actuating a.c. voltages of 7 kV, 8 kV and 9 kV at 0.5 Hz in sequence. d, e, The experimental processes were recorded by the video cameras (d) and LED lights (e) protected by pressure vessels. f, Throughout field tests in the South China Sea, the soft robot was grasped by a robotic arm on the ROV HAIMA, reaching the bottom of the South China Sea (depth of 3,224 m). After being released, the soft robot demonstrated free-swimming locomotion with flapping fins, actuated by an onboard a.c. voltage of 8 kV and 1 Hz. This process was recorded by the video cameras and LED lights on the ROV.
a, Top and bottom views of a centralized electronic circuit, in which electronics such as microchips, transformer, capacitors and crystal oscillator are densely packed on a PCB. The electronics are encapsulated in a soft silicone matrix for the pressure test. b, c, After pressure test at 110 MPa, we removed the silicone matrix and investigated the functioning of the electronics. The crystal oscillator, which contains an air gap, collapses (b), and the MCU connection fails (c). d, Top view of decentralized electronics. A calibrated MSI clock from the STM32L4 microprocessor is used to replace the crystal oscillator. e, Electronic function tests of each stage (illustrated in Extended Data Fig. 5) for decentralized electronics. Blue curve is the battery output (channel 1; 1 V per grid), green curve is the boost circuit output (channel 4; 5 V per grid), cyan curve is the ‘flyback’ circuit output (channel 2; 500 V per grid), and purple curve is the final output (channel 3; 2 kV per grid). The results indicate that the decentralized electronics maintained their function after the 110-MPa pressure test.
a, FEA of detaching electronics. The electronic component A is modelled as a linear elastic unit with an elastic modulus of 120 GPa and a Poisson’s ratio of 0.35. Component B is modelled as a linear elastic unit with an elastic modulus of 22 GPa and a Poisson’s ratio of 0.39. The encapsulating silicone matrix C is modelled as a hyper-elastic unit with a shear modulus of 0.2 MPa and a bulk modulus of 2,000 MPa. The stress distribution at the interface illustrates the stress mitigation effect of detaching (components separated by a distance of 2 mm). The blue and red dashed lines highlight the paths on the interfaces used for stress investigation. The relation between the distance to the centre and the maximum shear stress is plotted to illustrate the effect of stress mitigation. The dimensions of components A and B are both 10 mm × 10 mm × 2 mm. b, A symmetric FEA model is built to simulate two adjacent electronic components on a PCB. The distance between the two components (d) varies from 0.4 mm to 2.4 mm. The highlighted red dashed line indicates the path for shear stress investigation. c, The shear stress distributions of the nodes on the path (red dashed line) with different d are plotted. The result shows that distancing neighbouring components on the PCB reduces the interfacial shear stress.
a, A SBAS DE membrane (initial thickness 1.5 mm) is clamped onto a stretching machine. b, The SBAS DE membrane is pre-stretched (3 × 3) and attached to four square brackets made of acrylonitrile butadiene styrene (border length of 13.2 cm and border width of 2 cm) to keep the pre-stretch. Then we cut off the four brackets, each with a pre-stretched DE membrane, to fabricate the DE actuator. c, The carbon grease electrode is brushed on the pre-stretched DE membrane. An aluminium foil feed line is placed on the electrode to connect it with the external circuit. d, Two frames are combined to protect the internal electrode. e, The laser-cut PET elastic frame (region indicated by blue dashed line, thickness of 0.35 mm, also shown in Fig. 3a) is glued to the DE membrane in one side. f, Bracing units (red dashed lines: (1) acrylic board with thickness 1 mm, (2) carbon fibre plate with thickness 0.3 mm and (3) PET with thickness 0.35 mm) are glued on the opposite side of the membrane. g, The precursor (Dragon Skin 20) is poured into the 3D-printed mould to encapsulate the electronics. h, The as-fabricated soft robot.
The electronic system of soft robot is mainly composed of a high-voltage amplifier and electronics for control. The high-voltage output is achieved by a three-stage circuit. The first stage is a boost circuit, which is dominated by a MP3431 chip to boost a 3.7-V battery voltage to 15 V. The second stage is the ‘flyback’ circuit, which boosts the voltage from 15 V to a maximum amplitude of 3 kV through a transformer. The third stage is a voltage-multiplier circuit, which can multiply the voltage output of the second stage, and finally produces the final voltage (maximum amplitude of 10 kV). The MCU tunes the voltage output with amplitude in the range of ~2–10 kV and frequency in the range of ~0.5–3.0 Hz.
a, b, The circular actuators of VHB (a) and SBAS (b) are composed of two DE membranes (thickness 1.5 mm) with pre-stretch of 3 × 3, fixed on a rigid ring. A carbon grease electrode is sandwiched by the two DE membranes with an initial radius (R) of 1.25 cm as the active region. The carbon grease electrode is connected to the power source as the high-voltage end. The external water surrounding the other sides of the DE membranes is connected to the power source as the ground end. When a high voltage is applied, the active region expands to the actuated radius of r. The area strain is calculated as (r2 − R2)/R2. c, The experimental set-ups for the performance test of the DE actuator. The circular actuators are installed in a pressure chamber and connected to an external high-voltage power source (Trek 610E). The voltage-induced area strain of the DE circular actuator is recorded from the top by a camera protected by a metallic vessel. With an identical actuating voltage of 8 kV at 1 Hz, we tested the actuation performances of VHB and SBAS under various temperatures (2.7 °C, 5.0 °C, 10.0 °C, 15.0 °C, 20.0 °C and 25.0 °C) and hydrostatic pressures (0 MPa, 25 MPa, 50 MPa, 75 MPa, 100 MPa and 110 MPa). d, The SBAS film (thickness of 1.5 mm) after solution casting and solvent evaporation process (320 g SBAS is dissolved in 3,000 ml tetrahydrofuran; the casting area is 1 m × 1 m). e, AFM phase image of the SBAS film (the bright islands indicate polystyrene nanodomains with high stiffness). f, The schematic shows that owing to the thermodynamic incompatibility of different blocks, the physically crosslinked SBAS elastomer has a sea–island microstructure via microphase separation of polystyrene blocks (blue) and poly(butyl acrylate) (orange). g, GPC curves of molecular weight distribution indicate the block-by-block chain extension process (black, then red, and finally the blue curve) during the synthesis of the triblock coploymer SBAS. PSt, polystyrene; PnBA, poly(n-butyl acrylate). h, Dynamic mechanical analysis of VHB (red) and SBAS (black). The glass transition temperature of the material can be determined from the tipping points on the curve of storage modulus versus temperature. SBAS material has a lower Tg (−17.2 °C) than VHB (0.3 °C). The blue dashed line indicates the temperature close to the Mariana Trench (2.7 °C). At this temperature, the VHB is close to its glass transition and has higher storage modulus than SBAS, resulting in its reduced actuation in low temperature and high pressure (Fig. 3d).
a, b, A high-pressure hydraulic pump (a) pumps water into a pressure chamber (b) with inner diameter 80 cm and height 2 m, to simulate the 110-MPa condition in the deep sea. c, d, The soft robot was first magnetically attracted by a electromagnet (c) and located at the left side of the pressure chamber (d). The electromagnet with electronic timer was encapsulated in a soft silicone matrix with the same decentralized design for pressure resilience. It can robustly operate in 110 MPa pressure to attract and release the soft robot. e, The experimental time-controlled sequence for the soft robot and the electromagnet. When the soft robot was released by the electromagnet, it was then actuated by a high-voltage signal (dashed blue line shows releasing point). f, After being released, the soft robot free-swam in the pressure chamber with an actuating voltage of 7 kV at 1 Hz.
a, The soft robot was gripped by an ROV to be lowered and then released in a deep lake. Multiple tests were conducted at depths of 8 m and 70 m. b, Free-swimming robot (side view) after release from the ROV.
: Flapping actuation of soft robot in the Mariana Trench at the depth of 10,900 m. This video shows deep sea field test of soft robot in the Mariana Trench. Mounted on a deep sea lander, the soft robot reached the bottom of Mariana Trench. At the depth of 10,900m in the Mariana Trench, the DE driven soft robot kept flapping its fins. The front view and side view of the soft robot were recorded by the deep sea cameras and LED lights in anti-pressure shells.
: Free swimming of soft robot in deep sea at the depth of 3,224 m. This video shows the deep-sea free swimming of soft robot in the South China Sea. The soft robot was grasped by a robotic arm on ‘HAIMA’ ROV and reached the bottom of the South China Sea (depth of 3,224 m). After the releasing, the soft robot was actuated with an on-board AC voltage of 8 kV at 1 Hz and demonstrated free swimming locomotion with its flapping fins. The front view and side view of swimming process were recorded by the cameras and LED lights on the ROV. This video shows the potential of soft robots in deep-sea exploration.
: FEA simulation of soft robot flapping actuation under the pressure of 110 MPa. This video shows the actuating mechanism of soft robot using FEA software ABAQUS 6.13. The resting state and actuating state under a hydrostatic pressure of 110 MPa are shown in the video. When a voltage is applied on the membrane, the bending angle of the wings decreases. Actuated with a cyclic voltage, the wing flaps periodically.
: Soft robot swimming in circle in pressure chamber. This video shows the swimming experiment in circle in a pressure chamber. The soft robot was attached to a passive, rotatable and low-friction rod. The hydrostatic pressures were 0 MPa and 110 MPa. Actuated with on-board AC voltages of 9 kV, 8 kV and 7 kV at 2 Hz, the soft robot flapped its wings and swam in circles. The pressure responsible foam (green dyed) was compressed from a diameter of 6 cm (0 MPa) to 1.4 cm (110 MPa).
: The free swimming tests of soft robot. This video shows free swimming experiment in of the soft robot under hydrostatic pressures of 0 MPa (in a pool) and 110 MPa (in a chamber). The soft was attached on a time-controlled electromagnet and released. After the releasing, the soft robot was actuated with an on-board AC voltage of 7 kV at 1Hz and demonstrated a free swimming locomotion.
: Soft robot free swimming in deep lake. This video shows field exploration of the robot in a deep lake (depth ~70 m). The soft robot was carried by a gripper on ‘Blue ROV’ and released at depth of 8 m and 70 m. After the releasing, the soft robot was actuated with an on-board AC voltage of 8 kV at 1Hz. The free swimming of soft robot was recorded by the internal camera and lights on the ROV, which further confirms the robustness of the robot in field exploration.
About this article
Cite this article
Li, G., Chen, X., Zhou, F. et al. Self-powered soft robot in the Mariana Trench. Nature 591, 66–71 (2021). https://doi.org/10.1038/s41586-020-03153-z