Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A bio-inspired and self-powered triboelectric tactile sensor for underwater vehicle perception

A bio-inspired and self-powered triboelectric tactile sensor for underwater vehicle perception www.nature.com/npjflexelectron ARTICLE OPEN A bio-inspired and self-powered triboelectric tactile sensor for underwater vehicle perception 1,7 1,7 1,7 1 2 1 1 1 1 Peng Xu , Jianhua Liu , Xiangyu Liu , Xinyu Wang , Jiaxi Zheng , Siyuan Wang , Tianyu Chen , Hao Wang , Chuan Wang , 3 4 5,6 1 ✉ ✉ ✉ Xianping Fu , Guangming Xie , Jin Tao and Minyi Xu Marine mammals relying on tactile perception for hunting are able to achieve a remarkably high prey capture rate without visual or acoustic perception. Here, a self-powered triboelectric palm-like tactile sensor (TPTS) is designed to build a tactile perceptual system for underwater vehicles. It is enabled by a three-dimensional structure that mimics the leathery, granular texture in the palms of sea otters, whose inner neural architecture provides additional clues indicating the importance of tactile information. With the assistance of palm structure and triboelectric nanogenerator technology, the proposed TPTS has the ability to detect and distinguish normal and shear external load in real-time and approximate the external stimulation area, especially not affected by the touch frequency, that is, it can maintain stable performance under high-frequency contact. The results show that the TPTS is a promising tool for integration into grippers mounted on underwater vehicles to complete numerous underwater tasks. npj Flexible Electronics (2022) 6:25 ; https://doi.org/10.1038/s41528-022-00160-0 INTRODUCTION fluid motion sensor to follow underwater wake using a piezo- electric material. With the assistance of the Hall effect element, an Autonomous underwater vehicles (AUVs) have been developed artificial whisker was proposed to distinguish two similar textures for years owing to their outstanding applicability in various by sensing variations in stiffness . industries, and different designs of the vehicles’ sensors have 1–4 To date, the triboelectric nanogenerator (TENG) has not been been developed to help AUVs accomplish specific tasks . These built with a palm-like sensor for gathering information about tasks include mapping the local seafloor with the aid of binocular 5,6 underwater physical objects. Noticeably, TENG-based sensors are stereo vision , measuring specific parameters in the water 7,8 compatible with tactile receptors because Young’s modulus of soft column with the aid of biosensors , and navigating with the 9–11 materials typically used in triboelectric sensors is at the same level aid of inertial sensors . Although AUVs can complete numerous as the Young’s modulus of the palms of marine mammals. Thus, tasks, they are still not an indispensable part of everyday work in TENG coupled with triboelectrification and electrostatic induction the marine field. One reason for this is the lack of AUVs with tactile has been developed as an electromechanical energy conversion perception capabilities equivalent to those of marine animals. 23–25 technology . Some triboelectric devices have the potential to More specifically, the tactile perception capacity for dexterous and be used in irregular and ultralow frequency blue energy harvest- in-hand manipulation is essential for situations, where visual 26–28 29–31 ing and self-powered mechanical sensing . Moreover, self- 12,13 detection is difficult or impossible . powered triboelectric sensors have shown responses to the area The number of tactile sensors designed specifically for marine 32–35 of external stimuli , the pressure change in intrabody catheter applications is still relatively small, although existing force/torque 36 37 balloons , acceleration, force, and rotational motion , and in vivo 14–16 is potentially applicable to marine measurements . Marine 38 microscale movements . Based on these results, a triboelectric mammals with tactile receptors can sense complex stimuli from palm-like tactile sensor (TPTS) may provide a simple scheme for organisms’ motion. Tactile perception is the main way of providing underwater vehicles with the ability to gather informa- perceiving the surrounding environment among most aquatic tion about physical objects. and semi-aquatic taxa, especially when hunting buried inverte- Here, we report a self-powered palm-like sensor based on brates or fishes . Moreover, seals, as well as other pinniped contact-separation channel TENG for use in building tactile mammals, possess the ability to detect waterborne disturbances perceptual systems of underwater vehicles, as shown in Fig. 1. using their facial whiskers . Surface otters are unique among This sensor is composed of triboelectric sensing units, a flexible marine mammals. Sea otters have two enhanced, complementary support, a fixedframe,anupper hatchcover,anelastomeric tactile structures that can be dexterously controlled: flexible paws O-Ring seal, and a flexible cover. In addition, the spinous and a complex array of facial whiskers. The paws may have greater structure on the sensing unit surface, composed of interlocked functional relevance than the whiskers in sea otters when visual hills and localized at the interface between the sensing unit and cues are reduced or absent . These findings suggest that flexible support, produces a localized and high-stress concentra- the development of animal ethology may provide the scheme tion near the receptors, playing a pivotal role in afferent stimuli for the structural design of an underwater tactile sensor. Inspired for enhanced pressure perception. The deformation of the 20,21 by the structure of whisker follicles , Beem et al. designed a flexible cover due to external stimuli results in contact- 1 2 Dalian Key Laboratory of Marine Micro/Nano Energy and Self-powered Systems, Marine Engineering College, Dalian Maritime University, Dalian 116026, China. Transportation Engineering College, Dalian Maritime University, Dalian 116026, China. School of Information Science and Technology, Dalian Maritime University, Dalian 116026, China. 4 5 Intelligent Biomimetic Design Lab, College of Engineering, Peking University, Beijing 100871, China. College of Artificial Intelligence, Nankai Universiy, Tianjin 300350, China. 6 7 Silo AI, Helsinki 00100, Finland. These authors contributed equally: Peng Xu, Jianhua Liu, Xiangyu Liu. email: xiegming@pku.edu.cn; taoj@nankai.edu.cn; xuminyi@dlmu.edu.cn Published in partnership with Nanjing Tech University 1234567890():,; P. Xu et al. Fig. 1 The structure, working mechanism, and application of TPTS. a Structural diagram of TPTS. Inset: top view of the sensor unit layout. b Structural diagram of the sensing unit and SEM image of the FEP membrane surface. c Charge distribution during contact separation between FEP membrane and conductive ink. d Simulation of potential distribution using COMSOL software. (e) Application of the TPTS in pipeline non-destructive evaluation. f Application of the TPTS in underwater autonomous target grasping. separation motion between the conductive ink layer and the RESULTS AND DISCUSSION fluorinated ethylene propylene (FEP) membrane for the tribo- Design, working principle, and simulation of TPTS electric sensing unit, producing charge transfer over the contact Inspired by the leathery, granular texture of sea otters’ palms in electrification process. Experimental results demonstrate the ref. , a TPTS is designed for the tactile perception system of potential for TPTS to be integrated into the manipulator underwater vehicles. The function of the TPTS is to capture mounted on underwater vehicles and used to complete information about underwater targets through physical touch. numerous underwater tasks. Figure 1a shows the structural diagram of the tactile sensor based npj Flexible Electronics (2022) 25 Published in partnership with Nanjing Tech University 1234567890():,; P. Xu et al. on TENG. This structure consists of a triboelectric sensing unit with Supplementary Fig. 7a shows the experimental setup, which is the spinosum structure, a flexible support, a fixed frame, an upper mainly composed of a linear motor, an LZ-WL2 pressure sensor hatch cover, an elastomeric O-Ring seal, and a flexible cover. with an intelligent display instrument, a Keithley 6514 electro- Specifically, the flexible cover and support (60 mm in length, 32 meter, and a computer. In particular, the pressure sensor is fixed mm in width, and 35 mm in height) are made of silicone rubber, on the linear motor using a corresponding circular-like fixture. enabling external stimuli to be converted into sensing unit Supplementary Fig. 7b, c present photos of the experimental deformation. Charge transfer is generated during the deformation setup. Because all the sensing units have the same structure, only process. Four square sensing units (18 mm in length, 8 mm in the performance of the sensing unit 1 is presented for the sake of width, and 40 mm in height) are symmetrically installed on the brevity. four inner surfaces of the square flexible support. For the sake of Figure 2a shows the deformation diagram of the sensing units ∘ ∘ N,0 E), in which clarity, it is worth noting that the sensing units are counter- under the vertical external stimuli at position (45 ∘ ∘ ∘ ∘ the deformation of the flexible cover from the vertical external clockwise angles coded as 1: 0 ,2:90 , 3: 180 , and 4: 270 .In stimuli F(t), with the aid of the internal flexible support, causes the addition, a circular-like sensing unit (8 mm in radius and 8 mm in triboelectric sensing unit to bend so that contact-separation height) is placed directly below the flexible cover, as shown in the motion between the conductive ink layer and FEP membrane top view of Fig. 1a. occurs for the sensing unit. To facilitate the exhibition of Figure 1b depicts the structure of the triboelectric sensing unit, experimental data, Fig. 2b shows labels on the flexible cover including silicone rubber with a spinosum structure aiming at surface using longitude and latitude lines. Figure 2c shows that producing a localized and high-stress concentration near recep- the voltage signal of the TPTS increases as the vertical external tors, a cast polypropylene (CPP) film for voiding electrostatic stimulus with a frequency of f = 0.8 Hz increases from 1 N to 5 N. interference, and FEP films sprayed with conductive ink. In order This is because, with the help of the spinosum structure, the to avoid contact between triboelectric layers and water, the increasing external load results in the increasing bending dimension of two ink-coated FEP membranes equals the CPP magnitude of the sensing unit, which makes the contact- membranes. Around them, there are two larger silicone layers separation movement between the films more sufficient and made of Dragonskin 30 so that the silicone rubber can seal two generates a larger output voltage. Furthermore, the leave-one-out triboelectric layers. The scanning electron microscope (SEM) cross-validation (LOOCV) strategy is used to obtain the relation- image of the FEP membrane surface, depicted in Fig. 1b bottom, ship between voltage and external stimuli (U = 0.14057 F + is characterized by a rough pattern. This nanostructure can 0.05935, R = 0.98609) in Fig. 2d, where the correlation coefficient enhance the surface charge density of the TPTS for higher R indicates that the external stimuli and the output voltage have sensitivity. an approximate linear relationship, with an error of less than Figure 1c illustrates the electron transfer process of sensing 1.78% at magnitude F = 4 N, and a relatively higher error of 9.40% units receiving external stimuli. The FEP film contacts the ink at magnitude F = 3 N. The reason for these errors may originate electrode, resulting in the electron clouds overlapping on the two from the high-frequency vibration generated by the linear motor layers. Due to FEP having greater electronegativity than ink, the during operation and strong electromagnetic interference from electrons of the ink layer enter into the deeper potential well of the surrounding equipment. Figure 2e, f also describes the the FEP film. Namely, the free electrons on the ink surface are influence of external magnitude F = 1–5 N on the current signal at transferred to the lowest molecular orbital at the FEP interface. f = 0.8 Hz. It is worth noting that the linearity between the current When the TPTS removes an external load, the FEP film is separated signal and the external load (I = 0.58082 F + 0.12191, R = from the ink electrode. Because the negative and positive friction 0.99455) is better than the linearity between the voltage signal charges no longer coincide on the same plane, a dipole moment and the external load. This result indicates that, compared to the and potential are generated between the two contact surfaces. output voltage responding to external magnitude, the output Therefore, the free electrons are transferred through the external current is less affected by environmental interference. circuit to balance the local electric field, generating a positive Subsequently, under the condition of external stimuli magni- charge on the conductive ink electrode. The flow of electrons tude F = 3 N, sensing unit 1 is tested for voltage and current continues until the distance between the two contact surfaces signals at different collision frequencies from 0.8 Hz to 1.6 Hz.As reaches the maximum. Meanwhile, the voltage difference shown in Fig. 3a, as the collision frequency increases, the output decreases, and the free electrons return through the external voltage remains almost unchanged, staying around 0.58 V.It can circuit. Finally, the charge distribution returns to its initial state, be seen that a changing collision frequency has little effect on the which completes the whole generation cycle. As shown in Fig. 1d, voltage output of the TPTS. Figure 3b shows that the current I COMSOL software is used to simulate the potential distribution increases as the collision frequency increases. The linear relation- results between the two films and verify the working principle. ship between collision frequency and current is fitted (I = Figure 1e shows the application of the TPTS for underwater 0.45601 f + 1.53521, R = 0.98694) in Fig. 3c. Figure 3d, e pipeline non-destructive evaluation. The TPTS is installed on the describes the relationship between the output signal (voltage end of underwater vehicles’ manipulators as the pressure feed- and current, respectively) of the TPTS, the load magnitude, and back device to prevent pipeline damage resulting from excessive the collision frequency. The response surface reflects the clamping force. Moreover, evaluation is performed along the path interaction between the two factors and the influence of each of the pipeline. The damage degree of a pipeline can be obtained factor on the output signal. The shape of the equal-height line from the characteristics of the sensor signal. Figure 1f demon- reflects the strength of the effect of the two factors. Moreover, strates the application of the TPTS for underwater autonomous the slope of the contour line for output current in the X-Y plane is target grasping. To ensure the integrity of the target, underwater greater than the case of output voltage, which indicates that the vehicles need to control the magnitude of the manipulator’s two factors are significant for the output current. Together with grasping force based on the electrical signal from the TPTS. The the influence of collision frequency and external load magnitude TPTS is regarded as a promising tool for building an underwater on the output signal, it is proved that the TPTS may be suitable tactile perception system based on these applications. for high-frequency sensing tasks, as it will still recognize the characteristics of the external load. For example, when the output Output characterization of the TPTS voltage is constant and the output current continually changes, In the experimental design, a linear motor and pressure sensors the external stimuli are characterized by time-varying frequency are used to simulate the external stimuli with different parameters. and constant magnitude. Finally, the durability of the TPTS is Published in partnership with Nanjing Tech University npj Flexible Electronics (2022) 25 P. Xu et al. Fig. 2 The results of the TPTS excited by a linear motor. a Deformation diagram of sensor units under external load. b Labels on the flexible cover surface using longitude and latitude lines. c The open circuit voltage responding to external load magnitude from 1 N to 5 N. d Fitted linear relationship between open circuit voltage and external load magnitude. e The short circuit current responding to external load magnitude from 1 N to 5 N. f Fitted linear relationship between short circuit current and external load magnitude. Fig. 3 Characterization of the sensing unit. a The open-circuit voltage responding to collision frequencies from 0.8 Hz to 1.6 Hz. b The short circuit current responding to collision frequencies from 0.8 Hz to 1.6 Hz. c Fitted linear relationship between short circuit current and collision frequencies. d Relationship between the output voltage of the TPTS, load magnitude, and collision frequency. e Relationship between the output current of the TPTS, load magnitude, and collision frequency. f The durability of the TPTS tested for 2800 cycles. tested for 2800 cycles in Fig. 3f. The TPTS maintains a stable plotted in Supplementary Figs. 8–10. The electronic signals are electrical signal output after repeated impacts from an external correlated with the characteristics of external stimuli for all load. Since other sensing units have the same structure as sensing sensing units. This indicates the sensing units can be used for unit 1, similar data along the sensing units 2, 3, 4, and 5 are signal feedback in AUV control systems. npj Flexible Electronics (2022) 25 Published in partnership with Nanjing Tech University P. Xu et al. Fig. 4 Response characteristics of TPTS under different load directions and locations. a The initial state of triboelectric sensing unit 1, and the angle α between the external load and the horizontal line. b The state of sensing unit 1 at the angle α = 0 , and the open-circuit voltage and short-circuit current responding to the external load. c The state of sensing unit 1 at the angle α = 22. 5 , and the open-circuit voltage and short-circuit current responding to the external load. d The state of sensing unit 1 at the angle α = 45 , and the open-circuit voltage and short- circuit current responding to the external load. e The state of sensing unit 1 at the angle α = 67. 5 , and the open-circuit voltage and short- circuit current responding to the external load. f The state of sensing unit 1 at the angle α = 90 , and the open-circuit voltage and short-circuit current responding to the external load. g The open circuit voltage for sensing unit 1 responding to stimuli at different positions under the conditions of F = 3 N, f = 0.8 Hz, and α = 90 . h The open circuit voltage for sensing unit 5 responding to stimuli at different positions under the conditions of F = 3 N, f = 0.8 Hz, and α = 90 . (i) The open circuit voltage responding to flexible covers with different curvatures under the conditions of F = 3 N, f = 0.8 Hz, and α = 90 . Interlocked hills localized at the interface between the top layer external stimuli and the horizontal line. Figure 4b–f depict the 3D of the sensing unit and flexible support are designed to enhance geometry of the hills and the anisotropic deformation of the top pressure perception. The other key benefit of this structure is that layer with applied tilt load. For example, in the case of α = 0 , the it has the ability to detect the direction of the external load. As direction of the tilt load is perpendicular to sensing unit 1, which shown in Fig. 4a, when an external load does not press the means that no shear force is applied to the spinosum structure. ∘ ∘ sensing unit, the triboelectric sensing unit and the flexible support Namely, the part located at position (45 N,0 E) is only affected by remain at the initial state. For the sake of brevity, we introduce the the bending component. As shown in Fig. 4b, the output voltage ∘ ∘ symbol α, ranging from 0 to 90 , as the angle between the in this state is about 0.52 V and the output current is about 2.1 nA. Published in partnership with Nanjing Tech University npj Flexible Electronics (2022) 25 P. Xu et al. Fig. 5 Experiments with TPTS controlling LED lights. a Experimental electronic setup. b Electronic module used for potential application demonstrations, such as controlling LED lights. c Demonstration of TPTS as a sensitive load switch control and its corresponding output voltage signal. In the case of α = 22. 5 , the shear component and the bending where f is internal load, m denotes torque, and ∘ ∘ rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi component work on the part located at position (45 N,0 E) qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi signðf mÞ 2 2 T interactively rather than independently. As shown in Fig. 4c, the pffiffi K ¼ ζ þ ζ þ 4R ðf mÞ ; 2R (2) output voltage in this state is about 0.62 V and the output current 2 2 2 ζ ¼kmk  R kfk : is about 2.3 nA. Evidently, the shear component increases the contact interface between the top layer of the sensing unit and In addition, we also carry out performance tests using flexible flexible support, which leads to a more significant increase in covers with different curvatures. Figure 4i shows the output transfer charge compared to cases in which there is only a voltage sensing unit 1 with three flexible covers of different bending component. As α increases to 45 , the shear component curvatures under the same external load (F = 3 N, f = 0.8 Hz, and causes the contact interface to interlock completely, at which time α = 90 ). As the curvature decreases from ρ = 0.04 to ρ = 0.015, the output voltage (0.66 V) and output current (2.75 nA) reach the output voltage increases from 0.46 V to 0.73 V. This is because their peak values, as shown in Fig. 4d. Figure 4e–f shows the the smaller the curvature of the flexible cover, the smaller its performance of sensing unit 1 in the cases of α = 67. 5 and α = thickness. The corresponding output current is shown in 90 , in which the the shear component directions are opposite Supplementary Fig. 14. In the case of ρ = 0, the flexible cover is ∘ ∘ compared to the cases of α = 22. 5 and α = 45 . Thus, the output a plane, and the TPTS loses its ability to sense a 3D spatial load. ∘ ∘ signal located at position (45 N,0 E) provides the ability to differentiate several types of applied external loads. Real-time control Furthermore, the TPTS can also approximate the external Figure 5a shows a photograph of the experimental electronic stimulation area based on how the output voltage responds to setup, wherein LEDs are installed around the TPTS at intervals of different position stimuli on the flexible cover surface. This 45 , and an Arduino Due R3 is used as a circuit board to perform functionality is amply illustrated in Fig. 4g–h via external stimuli signal sampling and data processing. Data processing consists of (F = 3 N, f = 0.8 Hz, and α = 90 ) applied to different positions of two stages: system initialization and event detection via judging the flexible cover surface. From Fig. 4g, it can be observed that the whether the peak voltage of the TPTS satisfies the setting value. magnitude of the output voltage for sensing unit 1 decreases as This framework provides a triggering capability for turning on LED the applied load moves away from the center of sensing unit 1. As lights in a corresponding direction. The schematic digraph is shown in Fig. 4h, the output voltage of sensing unit 5 is not depicted in Fig. 5b. If two sensing units are larger than the setting maximized at the installation position. This is because the value at the same time, the electronic circuit drives the LED thickness of the flexible cover at the pole is the largest, reducing between the sensing units. The stimuli starts at the position the ability to transmit load-induced deformation. As a result, the between sensing units 2 and 3 and rotate in a clockwise manner output voltage response to a load at the pole is smaller than by 45 during each step of the experiment. Figure 5b shows how the TPTS is used to control the on/off state of LED lights (also see the output voltage response to a load around the pole. Due to the Supplementary Video 1) by a corresponding voltage, where the similar structure and symmetrical distribution of the sensing units, peak voltage in the stimuli direction is larger than in other the characteristic data of sensing units 2, 3, and 4 are depicted in directions. This result indicates the TPTS’ great potential in Supplementary Figs. 11–13. Using the output voltage response to approximating the external stimulation directions. different positions, the contact centroid location can be obtained for a sphere surface of radius R in ref. , as shown in Application of TPTS in detecting hardness Supplementary Fig. 2, with the following equation: To explore the compatibility of the TPTS with underwater vehicle applications, a manipulator integrated with the TPTS and mounted 2 T c ¼ ðK m þ Kf ´ m þðf mÞfÞ; on a remotely operated vehicle (ROV) is prepared to perform (1) K þkfk hardness detection of various silicone samples. As shown in Fig. 6a, b, npj Flexible Electronics (2022) 25 Published in partnership with Nanjing Tech University P. Xu et al. Fig. 6 Experiments with TPTS mounted on an ROV. a, b The configuration of the gripper and sensors for detecting the hardness of a silicone sample. c The output signals of the TPTS when gripping various silicone samples under the condition of β = 73 . d Demonstration of the TPTS in a closed-loop control system for an ROV and the corresponding output voltage signal. the TPTS is guided by bolts on the manipulator surface and is fixed at For specific underwater objects, the desired peak voltage its center. To determine the relationship between the gripping angle corresponding to the grasping force is provided, where the goal and the output signal for various silicone samples, the gripping angle is to grasp objects without breaking them. Once the peak voltage β is defined to represent the angle between the horizontal line and from the TPTS reaches the setting value, the gripper will stop the gripper location. The Shore hardness of the silicone samples is increasing β and maintain its current state until the object is 30A (Dragonskin30),2A (DragonskinPX),and 00− 20 (Ecoflex 20). released. For example, when grasping the sample with Dragon The results show that the magnitude of flexible cover deformation skin 30, the desired peak voltage is given as 0.77 V. As the output increases with the incremental Shore hardness of the samples under voltage reaches the setting value, the gripping motion holds on the condition of β= 73 , resulting in larger output peaks generated with β = 73 in T , as shown in Fig. 6dI (also see Supplementary by the TPTS in one single actuation motion in Fig. 6c. Furthermore, Video 3). For samples with Dragon skin PX and Ecoflex 20, the when the sample with Dragon skin 30 is gripped at the palm position gripper completes this task in T and T , respectively. From Fig. 6d, 2 3 of the gripper, the slope of the output signal in sensing unit 5 (k )is T < T < T results from the samples with incremental Shore 1 2 3 higher than the others, as shown in Supplementary Fig. 15. It is worth hardness. Therefore, different types of underwater objects may be noting that under the relatively fast gripping speed, the gripping recognized when visual cues are reduced or absent, verifying the motion can still be clearly recognized from the generated output feasibility of applying the TPTS to underwater vehicles. peaks of the sensing unit (see Supplementary Video 2). In addition, the TPTS can also distinguish the gripping and releasing motions of Application of TPTS in non-destructive underwater pipeline the gripper from the rising edge and falling edge of the output evaluation signal. As a result, the slope of voltage signal increases with the Experiments are performed in an indoor water pool (3m*2m*1.5m) incremental Shore hardness for sensing unit 5 (k > k > k ), proving 1 2 3 with the goal of non-destructive pipeline evaluation based on the the feasibility of hardness detection with the output signals of the TPTS. developed TPTS, as shown in Fig. 7a. For additional stability in the Subsequently, a simple closed-loop control system is built for pool-based experiments, the pipeline (2 m in length and 5 mm in grasping objects underwater, as shown in Supplementary Fig. 16. radius) with three ruptures to be detected is mounted on the floor. Published in partnership with Nanjing Tech University npj Flexible Electronics (2022) 25 P. Xu et al. Fig. 7 Experiments with non-destructive underwater pipeline evaluation. a The configuration of the pipeline and sensors in use of non- destructive underwater pipeline evaluation. b The process flows from signal sampling to non-destructive underwater pipeline evaluation. c Demonstration of TPTS in non-destructive underwater pipeline evaluation and its corresponding output voltage signal. At the beginning of the task, the ROV is operated to explore the a first step toward utilizing the highly sensitive directional sensing nearby environment until the vision system detects the pipe. capabilities of the TPTS. Moreover, an underwater sensory tactile Subsequently, the ROV uses the manipulator to grasp the system is built by mounting a manipulator integrated with the pipeline. The magnitude of the grasping force depends on TPTS on an ROV to obtain information about grasped objects. whether the TPTS is in contact with the pipe. The process flow Based on the improved intelligence of the gripper, it has the of non-destructive pipeline evaluation is depicted in Fig. 7b. ability to detect the hardness of samples and complete grasp tasks After grasping the pipeline, the ROV moves forward along the while preventing breakage. Additionally, the underwater sensory pipeline until it reaches the end of the pipeline. The screen tactile system is used to conduct non-destructive underwater displays the real-time signals, as shown in Fig. 7c(also see pipeline evaluation, illustrating its potential future application in Supplementary Video 5). When the presence of a falling edge various fields, including underwater object monitoring and tactical stays for a period of time during the detection process, it can be surveillance. judged that there is a fracture at the location of the ROV at this time. In addition, the roughness of the pipeline surface can be METHODS determined by observing the vibration magnitude of the real- time signals. Fabrication of TPTS As shown in Supplementary Fig. 17, the triboelectric sensing unit consists of six layers assembled by lamination: (i) a bottom 2 mm thick silicon DISCUSSION substrate made of Dragonskin 30, (ii) an intermediate 1 mm thick CPP, a commercial product, cut into square pieces (18 mm in length, 8 mm in In summary, a TPTS for underwater tactile perception based on width), (iii)–(iv) 0.3 mm thick FEP films sprayed with conductive ink serving triboelectric nanogenerators is proposed and investigated in this as a dielectric layer, (v) an intermediate 1 mm thick CPP, and (vi) a top 2 paper. The TPTS uses the linear relationship between the output mm thick silicone layer with an array of spinosums. The flexible cover and signal and parameters to sense the magnitude, frequency, and flexible support are made of Dragonskin 30. Specifically, 30 ml part A and contact area of external stimuli by analyzing the contact- 30 ml part B of silicone rubber are mixed in a petri dish. Then, a vacuum separation pattern and the triboelectric output of multiple sensing pump is used to vacuum the mixture to 0.1 MP for 2.5 min. When the units. In addition, the TPTS has a spinosum structure to measure evacuation of the mixture is finished, the mixture is poured into a mold and distinguish normal and shear forces generated when manufactured by 3D printing. The molds of the upper hatch cover and interacting with an object in real time. The TPTS is then used to fixed frame made from polylactic acid are also manufactured by 3D control the on/off states of lights oriented in various directions as printing. In addition, The conductive ink No. is CH-8(MOD2) that is npj Flexible Electronics (2022) 25 Published in partnership with Nanjing Tech University P. Xu et al. produced by JUJO printing supplies and technology (Pinghu) Co. Ltd. The 16. Ju, F. et al. A miniature piezoelectric spiral tactile sensor for tissue hardness conductive ink is printed on the FEP membrane by the screen-printing palpation with catheter robot in minimally invasive surgery. Smart Mater. Struct. technology, then dried in the atmosphere temperature. Finally, the flexible 28, 025033 (2019). cover, fixed frame, and upper hatch cover parts are assembled using an 17. Dehnhardt, G. & Mauck, B. Mechanoreception in secondarily aquatic vertebrates. elastomeric O-Ring seal and several screws. In Sensory evolution on the threshold. 295–314 (2008). 18. Niesterok, B., Kruger, Y., Wieskotten, S., Dehnhardt, G. & Hanke, W. Hydrodynamic detection and localization of artificial flatfish breathing currents by harbour seals Electric measurement and characterization (phoca vitulina). J. Exp. Biol. 220, 174–185 (2017). Field-emission scanning electron microscopy (Hitachi SU8010) is used to 19. Strobel, S. M., Sills, J. M., Tinker, M. T. & Reichmuth, C. J. Active touch in sea otters: characterize the surface morphology of the nanostructured FEP film in-air and underwater texture discrimination thresholds and behavioral strategies sprayed with conductive ink. For the electric output measurement of the for paws and vibrissae. J. Exp. Biol. 221, jeb181347 (2018). TPTS, a linear motor (LINMOT EI200-P01) with a frequency modulation 20. Beem, H., Hildner, M. & Triantafyllou, M. Characterization of a harbor seal whisker- program is used to drive the pressure sensor (LZ-WL2) to collision for inspired flow sensor. In 2012 Oceans.1–4 (2012). quantified investigation. A Keithley (6514) electrometer is adopted to 21. Beem, H., Liu, Y., Barbastathis, G. & Triantafyllou, M. Vortex-induced vibration measure the voltage signal. NI-6259 is used for data collection. The measurements of seal whiskers using digital holography. In OCEANS 2014-TAIPEI. software platform is constructed on the basis of LabVIEW, which is 1–4 (2014). capable of realizing real-time data acquisition control and analysis. For 22. Wegiriya, H., Herzig, N., Abad, S. A., Sadati, S. H. & Nanayakkara, T. A stiffness example, an intelligent display instrument shows the simulated external controllable multimodal whisker sensor follicle for texture comparison. IEEE Sens. load generated by changing the setting parameters of the linear motor. J. 20, 2320–2328 (2019). When the TPTS collides with the pressure sensor formed by the linear 23. Fan, F. R., Tian, Z.-Q. & Wang, Z. L. Flexible triboelectric generator. Nano Energy. 1, motor, the Keithley electrometer samples the electric signal and 328–334 (2012). transmits the data to the computer for visualization with LabVIEW 24. Xia, X., Fu, J. J. & Zi, Y. L. A universal standardized method for output capability software. assessment of nanogenerators. Nat. Commun. 10, 4428 (2019). 25. Wang, H., Cheng, J., Wang, Z. Z., Ji, L. H. & Wang, Z. L. Triboelectric nanogen- erators for human-health care. Sci. Bull. 66, 490–511 (2020). DATA AVAILABILITY 26. Xie, W. B. et al. A nonresonant hybridized electromagnetic-triboelectric nano- generator for irregular and ultralow frequency blue energy harvesting. Research. The datasets generated during and/or analyzed during the current study are available https://doi.org/10.34133/2021/5963293 (2021). from the corresponding author on reasonable request. 27. Wang, Z. L., Jiang, T. & Xu, L. Toward the blue energy dream by triboelectric nanogenerator networks. Nano Energy. 39,9–23 (2017). Received: 5 December 2021; Accepted: 20 March 2022; 28. Shi, Q., Wang, H., Wu, H. & Lee, C. K. Self-powered triboelectric nanogenerator buoy ball for applications ranging from environment monitoring to water wave energy farm. Nano Energy. 40, 203–213 (2017). 29. Shi, Q. F., Wu, H., Wang, H., Wu, H. X. & Lee, C. K. Self-powered gyroscope ball using a triboelectric mechanism. Adv. Energy Mater. 7, 1701300 (2017). REFERENCES 30. Guo, H. et al. A highly sensitive, self-powered triboelectric auditory sensor for social robotics and hearing aids. Sci. Robot. 3, eaat2516 (2018). 1. Paull, L., Saeedi, S., Seto, M. & Li, H. Sensor-driven online coverage planning for 31. He, Q. et al. Triboelectric vibration sensor for a human-machine interface built on autonomous underwater vehicles. IEEE ASME Trans Mechatron. 18,1827–1838 (2012). ubiquitous surfaces. Nano Energy. 59, 689–696 (2019). 2. Melo, J. & Matos, A. Survey on advances on terrain based navigation for auton- 32. Jin, T. et al. Triboelectric nanogenerator sensors for soft robotics aiming at digital omous underwater vehicles. Ocean Eng. 139, 250–264 (2017). twin applications. Nat. Commun. 11, 5381 (2020). 3. He, B. et al. Autonomous navigation for autonomous underwater vehicles based 33. Xu, P. et al. A triboelectric-based artificial whisker for reactive obstacle avoidance on information filters and active sensing. Sens. 11, 10958–10980 (2011). and local mapping. Research. 2021, 9864967 (2021). 4. Jakuba, M. V. et al. Toward automatic classification of chemical sensor data from 34. Luo, Y. et al. Triboelectric bending sensor based smart glove towards intuitive autonomous underwater vehicles. In 2011 IEEE/RSJ International Conference on multi-dimensional human-machine interfaces. Nano Energy. 89, 106330 Intelligent Robots and Systems. 4722–4727 (2011). (2021). 5. Lin, Y. H., Shou, K. P. & Huang, L. J. The initial study of lls-based binocular stereo- 35. Li, J. R. et al. Triboelectric nanogenerators enabled internet of things: a survey. vision system on underwater 3d image reconstruction in the laboratory. J. Mar. Intelligent Converged Netw. 1, 115–141 (2020). Sci. Technol. 22, 513–532 (2017). 36. Kumar, K. S. et al. Stretchable capacitive pressure sensing sleeve deployable onto 6. Bruno, F., Bianco, G., Muzzupappa, M., Barone, S. & Razionale, A. V. Experi- catheter balloons towards continuous intra-abdominal pressure monitoring. mentation of structured light and stereo vision for underwater 3d reconstruction. Biosensors. 11, 156 (2021). ISPRS J. Photogramm. Remote Sens. 66, 508–518 (2011). 37. Wu, Z. Y. et al. Multifunctional sensor based on translational-rotary triboelectric 7. Dimitratos, S. D. et al. Biosensors to monitor water quality utilizing insect odorant- nanogenerator. Adv. Energy Mater. 9, 1901124 (2019). binding proteins as detector elements. Biosensors. 9, 62 (2019). 38. Cheng, B. L. et al. Mechanically asymmetrical triboelectric nanogenerator for self- 8. Kröger, S. & Law, R. J. Biosensors for marine applications: We all need the sea, but powered monitoring of in vivo microscale weak movement. Adv. Energy Mater. does the sea need biosensors? Biosens. Bioelectron. 20, 1903–1913 (2005). 10, 2000827 (2020). 9. Grenon, G., An, P. E., Smith, S. M. & Healey, A. J. Enhancement of the inertial 39. Bicchi, A., Salisbury, K. J. & Brock, D. L. Contact sensing from force measurements. navigation system for the morpheus autonomous underwater vehicles. J. Ocean. Int. J. Robot. Res. 12, 249–262 (1993). Eng. 26, 548–560 (2001). 10. Panish, R. & Taylor, M. Achieving high navigation accuracy using inertial navi- gation systems in autonomous underwater vehicles. In OCEANS 2011 1–7 (IEEE, Spain, 2011). ACKNOWLEDGEMENTS 11. Lee, P. M. & Jun, B. H. Pseudo long base line navigation algorithm for underwater The authors would like to thank the National Key R & D Project from the Minister of vehicles with inertial sensors and two acoustic range measurements. Ocean Eng. Science and Technology (2021YFA1201604), the Dalian Outstanding Young Scientific 34, 416–425 (2007). and Technological Talents Project (2021RJ11), the National Natural Science 12. Rooney, T., Pipe, A. G., Dogramadzi, S. & Pearson, M. Towards tactile sensing Foundation of China (62003175, 51879022), the Beijing Natural Science Foundation applied to underwater autonomous vehicles for near shore survey and de- (No. 4192026), and the Academy of Finland (Grant No. 315660). mining. In Conference Towards Autonomous Robotic Systems. 463–464 (2012). 13. Muscolo, G. G. & Cannata, G. A novel tactile sensor for underwater applications: limits and perspectives. In OCEANS 2015-Genova.1–7 (2015). AUTHOR CONTRIBUTIONS 14. Yohan, N. et al. A three-axial body force sensor for flexible manipulators. In 2014 IEEE International Conference on Robotics and Automation (ICRA). 6388–6393 M.X. designed the experiments. J.L. and X.L. synthesized the targets. P.X., J.L., and X.L. (2014). performed sample growth. P.X., J.L., X.W., and S.W. performed magnetism 15. Sun, Y., Liu, F., Yuan, Z. P., Huang,W., M. & Wang, B. W. A novel three-axial force measurements. P.X., J.L., X.L., J.Z., and T.C. performed sample structural characteriza- tactile sensor based on the fringing effect of electric field. IEEE Trans. Magn. 55, tion and data analysis. H.W. and C.W. performed theoretical calculations. P.X. wrote 1–5 (2019). the manuscript with significant contributions from J.L., X.W., J.Z., T.C., H.W., and C.W. Published in partnership with Nanjing Tech University npj Flexible Electronics (2022) 25 P. Xu et al. as well as contributions from all other authors. M.X., J.T., and G.X. supervised the Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims project. in published maps and institutional affiliations. COMPETING INTERESTS Open Access This article is licensed under a Creative Commons The authors declare no competing interests. Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative ADDITIONAL INFORMATION Commons license, and indicate if changes were made. The images or other third party Supplementary information The online version contains supplementary material material in this article are included in the article’s Creative Commons license, unless available at https://doi.org/10.1038/s41528-022-00160-0. indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory Correspondence and requests for materials should be addressed to Guangming Xie, regulation or exceeds the permitted use, you will need to obtain permission directly Jin Tao or Minyi Xu. from the copyright holder. To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/. Reprints and permission information is available at http://www.nature.com/ reprints © The Author(s) 2022 npj Flexible Electronics (2022) 25 Published in partnership with Nanjing Tech University http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png npj Flexible Electronics Springer Journals

A bio-inspired and self-powered triboelectric tactile sensor for underwater vehicle perception

Download PDF
10 pages

Loading next page...
 
/lp/springer-journals/a-bio-inspired-and-self-powered-triboelectric-tactile-sensor-for-i9gtaQVjen

References (43)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2022
eISSN
2397-4621
DOI
10.1038/s41528-022-00160-0
Publisher site
See Article on Publisher Site

Abstract

www.nature.com/npjflexelectron ARTICLE OPEN A bio-inspired and self-powered triboelectric tactile sensor for underwater vehicle perception 1,7 1,7 1,7 1 2 1 1 1 1 Peng Xu , Jianhua Liu , Xiangyu Liu , Xinyu Wang , Jiaxi Zheng , Siyuan Wang , Tianyu Chen , Hao Wang , Chuan Wang , 3 4 5,6 1 ✉ ✉ ✉ Xianping Fu , Guangming Xie , Jin Tao and Minyi Xu Marine mammals relying on tactile perception for hunting are able to achieve a remarkably high prey capture rate without visual or acoustic perception. Here, a self-powered triboelectric palm-like tactile sensor (TPTS) is designed to build a tactile perceptual system for underwater vehicles. It is enabled by a three-dimensional structure that mimics the leathery, granular texture in the palms of sea otters, whose inner neural architecture provides additional clues indicating the importance of tactile information. With the assistance of palm structure and triboelectric nanogenerator technology, the proposed TPTS has the ability to detect and distinguish normal and shear external load in real-time and approximate the external stimulation area, especially not affected by the touch frequency, that is, it can maintain stable performance under high-frequency contact. The results show that the TPTS is a promising tool for integration into grippers mounted on underwater vehicles to complete numerous underwater tasks. npj Flexible Electronics (2022) 6:25 ; https://doi.org/10.1038/s41528-022-00160-0 INTRODUCTION fluid motion sensor to follow underwater wake using a piezo- electric material. With the assistance of the Hall effect element, an Autonomous underwater vehicles (AUVs) have been developed artificial whisker was proposed to distinguish two similar textures for years owing to their outstanding applicability in various by sensing variations in stiffness . industries, and different designs of the vehicles’ sensors have 1–4 To date, the triboelectric nanogenerator (TENG) has not been been developed to help AUVs accomplish specific tasks . These built with a palm-like sensor for gathering information about tasks include mapping the local seafloor with the aid of binocular 5,6 underwater physical objects. Noticeably, TENG-based sensors are stereo vision , measuring specific parameters in the water 7,8 compatible with tactile receptors because Young’s modulus of soft column with the aid of biosensors , and navigating with the 9–11 materials typically used in triboelectric sensors is at the same level aid of inertial sensors . Although AUVs can complete numerous as the Young’s modulus of the palms of marine mammals. Thus, tasks, they are still not an indispensable part of everyday work in TENG coupled with triboelectrification and electrostatic induction the marine field. One reason for this is the lack of AUVs with tactile has been developed as an electromechanical energy conversion perception capabilities equivalent to those of marine animals. 23–25 technology . Some triboelectric devices have the potential to More specifically, the tactile perception capacity for dexterous and be used in irregular and ultralow frequency blue energy harvest- in-hand manipulation is essential for situations, where visual 26–28 29–31 ing and self-powered mechanical sensing . Moreover, self- 12,13 detection is difficult or impossible . powered triboelectric sensors have shown responses to the area The number of tactile sensors designed specifically for marine 32–35 of external stimuli , the pressure change in intrabody catheter applications is still relatively small, although existing force/torque 36 37 balloons , acceleration, force, and rotational motion , and in vivo 14–16 is potentially applicable to marine measurements . Marine 38 microscale movements . Based on these results, a triboelectric mammals with tactile receptors can sense complex stimuli from palm-like tactile sensor (TPTS) may provide a simple scheme for organisms’ motion. Tactile perception is the main way of providing underwater vehicles with the ability to gather informa- perceiving the surrounding environment among most aquatic tion about physical objects. and semi-aquatic taxa, especially when hunting buried inverte- Here, we report a self-powered palm-like sensor based on brates or fishes . Moreover, seals, as well as other pinniped contact-separation channel TENG for use in building tactile mammals, possess the ability to detect waterborne disturbances perceptual systems of underwater vehicles, as shown in Fig. 1. using their facial whiskers . Surface otters are unique among This sensor is composed of triboelectric sensing units, a flexible marine mammals. Sea otters have two enhanced, complementary support, a fixedframe,anupper hatchcover,anelastomeric tactile structures that can be dexterously controlled: flexible paws O-Ring seal, and a flexible cover. In addition, the spinous and a complex array of facial whiskers. The paws may have greater structure on the sensing unit surface, composed of interlocked functional relevance than the whiskers in sea otters when visual hills and localized at the interface between the sensing unit and cues are reduced or absent . These findings suggest that flexible support, produces a localized and high-stress concentra- the development of animal ethology may provide the scheme tion near the receptors, playing a pivotal role in afferent stimuli for the structural design of an underwater tactile sensor. Inspired for enhanced pressure perception. The deformation of the 20,21 by the structure of whisker follicles , Beem et al. designed a flexible cover due to external stimuli results in contact- 1 2 Dalian Key Laboratory of Marine Micro/Nano Energy and Self-powered Systems, Marine Engineering College, Dalian Maritime University, Dalian 116026, China. Transportation Engineering College, Dalian Maritime University, Dalian 116026, China. School of Information Science and Technology, Dalian Maritime University, Dalian 116026, China. 4 5 Intelligent Biomimetic Design Lab, College of Engineering, Peking University, Beijing 100871, China. College of Artificial Intelligence, Nankai Universiy, Tianjin 300350, China. 6 7 Silo AI, Helsinki 00100, Finland. These authors contributed equally: Peng Xu, Jianhua Liu, Xiangyu Liu. email: xiegming@pku.edu.cn; taoj@nankai.edu.cn; xuminyi@dlmu.edu.cn Published in partnership with Nanjing Tech University 1234567890():,; P. Xu et al. Fig. 1 The structure, working mechanism, and application of TPTS. a Structural diagram of TPTS. Inset: top view of the sensor unit layout. b Structural diagram of the sensing unit and SEM image of the FEP membrane surface. c Charge distribution during contact separation between FEP membrane and conductive ink. d Simulation of potential distribution using COMSOL software. (e) Application of the TPTS in pipeline non-destructive evaluation. f Application of the TPTS in underwater autonomous target grasping. separation motion between the conductive ink layer and the RESULTS AND DISCUSSION fluorinated ethylene propylene (FEP) membrane for the tribo- Design, working principle, and simulation of TPTS electric sensing unit, producing charge transfer over the contact Inspired by the leathery, granular texture of sea otters’ palms in electrification process. Experimental results demonstrate the ref. , a TPTS is designed for the tactile perception system of potential for TPTS to be integrated into the manipulator underwater vehicles. The function of the TPTS is to capture mounted on underwater vehicles and used to complete information about underwater targets through physical touch. numerous underwater tasks. Figure 1a shows the structural diagram of the tactile sensor based npj Flexible Electronics (2022) 25 Published in partnership with Nanjing Tech University 1234567890():,; P. Xu et al. on TENG. This structure consists of a triboelectric sensing unit with Supplementary Fig. 7a shows the experimental setup, which is the spinosum structure, a flexible support, a fixed frame, an upper mainly composed of a linear motor, an LZ-WL2 pressure sensor hatch cover, an elastomeric O-Ring seal, and a flexible cover. with an intelligent display instrument, a Keithley 6514 electro- Specifically, the flexible cover and support (60 mm in length, 32 meter, and a computer. In particular, the pressure sensor is fixed mm in width, and 35 mm in height) are made of silicone rubber, on the linear motor using a corresponding circular-like fixture. enabling external stimuli to be converted into sensing unit Supplementary Fig. 7b, c present photos of the experimental deformation. Charge transfer is generated during the deformation setup. Because all the sensing units have the same structure, only process. Four square sensing units (18 mm in length, 8 mm in the performance of the sensing unit 1 is presented for the sake of width, and 40 mm in height) are symmetrically installed on the brevity. four inner surfaces of the square flexible support. For the sake of Figure 2a shows the deformation diagram of the sensing units ∘ ∘ N,0 E), in which clarity, it is worth noting that the sensing units are counter- under the vertical external stimuli at position (45 ∘ ∘ ∘ ∘ the deformation of the flexible cover from the vertical external clockwise angles coded as 1: 0 ,2:90 , 3: 180 , and 4: 270 .In stimuli F(t), with the aid of the internal flexible support, causes the addition, a circular-like sensing unit (8 mm in radius and 8 mm in triboelectric sensing unit to bend so that contact-separation height) is placed directly below the flexible cover, as shown in the motion between the conductive ink layer and FEP membrane top view of Fig. 1a. occurs for the sensing unit. To facilitate the exhibition of Figure 1b depicts the structure of the triboelectric sensing unit, experimental data, Fig. 2b shows labels on the flexible cover including silicone rubber with a spinosum structure aiming at surface using longitude and latitude lines. Figure 2c shows that producing a localized and high-stress concentration near recep- the voltage signal of the TPTS increases as the vertical external tors, a cast polypropylene (CPP) film for voiding electrostatic stimulus with a frequency of f = 0.8 Hz increases from 1 N to 5 N. interference, and FEP films sprayed with conductive ink. In order This is because, with the help of the spinosum structure, the to avoid contact between triboelectric layers and water, the increasing external load results in the increasing bending dimension of two ink-coated FEP membranes equals the CPP magnitude of the sensing unit, which makes the contact- membranes. Around them, there are two larger silicone layers separation movement between the films more sufficient and made of Dragonskin 30 so that the silicone rubber can seal two generates a larger output voltage. Furthermore, the leave-one-out triboelectric layers. The scanning electron microscope (SEM) cross-validation (LOOCV) strategy is used to obtain the relation- image of the FEP membrane surface, depicted in Fig. 1b bottom, ship between voltage and external stimuli (U = 0.14057 F + is characterized by a rough pattern. This nanostructure can 0.05935, R = 0.98609) in Fig. 2d, where the correlation coefficient enhance the surface charge density of the TPTS for higher R indicates that the external stimuli and the output voltage have sensitivity. an approximate linear relationship, with an error of less than Figure 1c illustrates the electron transfer process of sensing 1.78% at magnitude F = 4 N, and a relatively higher error of 9.40% units receiving external stimuli. The FEP film contacts the ink at magnitude F = 3 N. The reason for these errors may originate electrode, resulting in the electron clouds overlapping on the two from the high-frequency vibration generated by the linear motor layers. Due to FEP having greater electronegativity than ink, the during operation and strong electromagnetic interference from electrons of the ink layer enter into the deeper potential well of the surrounding equipment. Figure 2e, f also describes the the FEP film. Namely, the free electrons on the ink surface are influence of external magnitude F = 1–5 N on the current signal at transferred to the lowest molecular orbital at the FEP interface. f = 0.8 Hz. It is worth noting that the linearity between the current When the TPTS removes an external load, the FEP film is separated signal and the external load (I = 0.58082 F + 0.12191, R = from the ink electrode. Because the negative and positive friction 0.99455) is better than the linearity between the voltage signal charges no longer coincide on the same plane, a dipole moment and the external load. This result indicates that, compared to the and potential are generated between the two contact surfaces. output voltage responding to external magnitude, the output Therefore, the free electrons are transferred through the external current is less affected by environmental interference. circuit to balance the local electric field, generating a positive Subsequently, under the condition of external stimuli magni- charge on the conductive ink electrode. The flow of electrons tude F = 3 N, sensing unit 1 is tested for voltage and current continues until the distance between the two contact surfaces signals at different collision frequencies from 0.8 Hz to 1.6 Hz.As reaches the maximum. Meanwhile, the voltage difference shown in Fig. 3a, as the collision frequency increases, the output decreases, and the free electrons return through the external voltage remains almost unchanged, staying around 0.58 V.It can circuit. Finally, the charge distribution returns to its initial state, be seen that a changing collision frequency has little effect on the which completes the whole generation cycle. As shown in Fig. 1d, voltage output of the TPTS. Figure 3b shows that the current I COMSOL software is used to simulate the potential distribution increases as the collision frequency increases. The linear relation- results between the two films and verify the working principle. ship between collision frequency and current is fitted (I = Figure 1e shows the application of the TPTS for underwater 0.45601 f + 1.53521, R = 0.98694) in Fig. 3c. Figure 3d, e pipeline non-destructive evaluation. The TPTS is installed on the describes the relationship between the output signal (voltage end of underwater vehicles’ manipulators as the pressure feed- and current, respectively) of the TPTS, the load magnitude, and back device to prevent pipeline damage resulting from excessive the collision frequency. The response surface reflects the clamping force. Moreover, evaluation is performed along the path interaction between the two factors and the influence of each of the pipeline. The damage degree of a pipeline can be obtained factor on the output signal. The shape of the equal-height line from the characteristics of the sensor signal. Figure 1f demon- reflects the strength of the effect of the two factors. Moreover, strates the application of the TPTS for underwater autonomous the slope of the contour line for output current in the X-Y plane is target grasping. To ensure the integrity of the target, underwater greater than the case of output voltage, which indicates that the vehicles need to control the magnitude of the manipulator’s two factors are significant for the output current. Together with grasping force based on the electrical signal from the TPTS. The the influence of collision frequency and external load magnitude TPTS is regarded as a promising tool for building an underwater on the output signal, it is proved that the TPTS may be suitable tactile perception system based on these applications. for high-frequency sensing tasks, as it will still recognize the characteristics of the external load. For example, when the output Output characterization of the TPTS voltage is constant and the output current continually changes, In the experimental design, a linear motor and pressure sensors the external stimuli are characterized by time-varying frequency are used to simulate the external stimuli with different parameters. and constant magnitude. Finally, the durability of the TPTS is Published in partnership with Nanjing Tech University npj Flexible Electronics (2022) 25 P. Xu et al. Fig. 2 The results of the TPTS excited by a linear motor. a Deformation diagram of sensor units under external load. b Labels on the flexible cover surface using longitude and latitude lines. c The open circuit voltage responding to external load magnitude from 1 N to 5 N. d Fitted linear relationship between open circuit voltage and external load magnitude. e The short circuit current responding to external load magnitude from 1 N to 5 N. f Fitted linear relationship between short circuit current and external load magnitude. Fig. 3 Characterization of the sensing unit. a The open-circuit voltage responding to collision frequencies from 0.8 Hz to 1.6 Hz. b The short circuit current responding to collision frequencies from 0.8 Hz to 1.6 Hz. c Fitted linear relationship between short circuit current and collision frequencies. d Relationship between the output voltage of the TPTS, load magnitude, and collision frequency. e Relationship between the output current of the TPTS, load magnitude, and collision frequency. f The durability of the TPTS tested for 2800 cycles. tested for 2800 cycles in Fig. 3f. The TPTS maintains a stable plotted in Supplementary Figs. 8–10. The electronic signals are electrical signal output after repeated impacts from an external correlated with the characteristics of external stimuli for all load. Since other sensing units have the same structure as sensing sensing units. This indicates the sensing units can be used for unit 1, similar data along the sensing units 2, 3, 4, and 5 are signal feedback in AUV control systems. npj Flexible Electronics (2022) 25 Published in partnership with Nanjing Tech University P. Xu et al. Fig. 4 Response characteristics of TPTS under different load directions and locations. a The initial state of triboelectric sensing unit 1, and the angle α between the external load and the horizontal line. b The state of sensing unit 1 at the angle α = 0 , and the open-circuit voltage and short-circuit current responding to the external load. c The state of sensing unit 1 at the angle α = 22. 5 , and the open-circuit voltage and short-circuit current responding to the external load. d The state of sensing unit 1 at the angle α = 45 , and the open-circuit voltage and short- circuit current responding to the external load. e The state of sensing unit 1 at the angle α = 67. 5 , and the open-circuit voltage and short- circuit current responding to the external load. f The state of sensing unit 1 at the angle α = 90 , and the open-circuit voltage and short-circuit current responding to the external load. g The open circuit voltage for sensing unit 1 responding to stimuli at different positions under the conditions of F = 3 N, f = 0.8 Hz, and α = 90 . h The open circuit voltage for sensing unit 5 responding to stimuli at different positions under the conditions of F = 3 N, f = 0.8 Hz, and α = 90 . (i) The open circuit voltage responding to flexible covers with different curvatures under the conditions of F = 3 N, f = 0.8 Hz, and α = 90 . Interlocked hills localized at the interface between the top layer external stimuli and the horizontal line. Figure 4b–f depict the 3D of the sensing unit and flexible support are designed to enhance geometry of the hills and the anisotropic deformation of the top pressure perception. The other key benefit of this structure is that layer with applied tilt load. For example, in the case of α = 0 , the it has the ability to detect the direction of the external load. As direction of the tilt load is perpendicular to sensing unit 1, which shown in Fig. 4a, when an external load does not press the means that no shear force is applied to the spinosum structure. ∘ ∘ sensing unit, the triboelectric sensing unit and the flexible support Namely, the part located at position (45 N,0 E) is only affected by remain at the initial state. For the sake of brevity, we introduce the the bending component. As shown in Fig. 4b, the output voltage ∘ ∘ symbol α, ranging from 0 to 90 , as the angle between the in this state is about 0.52 V and the output current is about 2.1 nA. Published in partnership with Nanjing Tech University npj Flexible Electronics (2022) 25 P. Xu et al. Fig. 5 Experiments with TPTS controlling LED lights. a Experimental electronic setup. b Electronic module used for potential application demonstrations, such as controlling LED lights. c Demonstration of TPTS as a sensitive load switch control and its corresponding output voltage signal. In the case of α = 22. 5 , the shear component and the bending where f is internal load, m denotes torque, and ∘ ∘ rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi component work on the part located at position (45 N,0 E) qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi signðf mÞ 2 2 T interactively rather than independently. As shown in Fig. 4c, the pffiffi K ¼ ζ þ ζ þ 4R ðf mÞ ; 2R (2) output voltage in this state is about 0.62 V and the output current 2 2 2 ζ ¼kmk  R kfk : is about 2.3 nA. Evidently, the shear component increases the contact interface between the top layer of the sensing unit and In addition, we also carry out performance tests using flexible flexible support, which leads to a more significant increase in covers with different curvatures. Figure 4i shows the output transfer charge compared to cases in which there is only a voltage sensing unit 1 with three flexible covers of different bending component. As α increases to 45 , the shear component curvatures under the same external load (F = 3 N, f = 0.8 Hz, and causes the contact interface to interlock completely, at which time α = 90 ). As the curvature decreases from ρ = 0.04 to ρ = 0.015, the output voltage (0.66 V) and output current (2.75 nA) reach the output voltage increases from 0.46 V to 0.73 V. This is because their peak values, as shown in Fig. 4d. Figure 4e–f shows the the smaller the curvature of the flexible cover, the smaller its performance of sensing unit 1 in the cases of α = 67. 5 and α = thickness. The corresponding output current is shown in 90 , in which the the shear component directions are opposite Supplementary Fig. 14. In the case of ρ = 0, the flexible cover is ∘ ∘ compared to the cases of α = 22. 5 and α = 45 . Thus, the output a plane, and the TPTS loses its ability to sense a 3D spatial load. ∘ ∘ signal located at position (45 N,0 E) provides the ability to differentiate several types of applied external loads. Real-time control Furthermore, the TPTS can also approximate the external Figure 5a shows a photograph of the experimental electronic stimulation area based on how the output voltage responds to setup, wherein LEDs are installed around the TPTS at intervals of different position stimuli on the flexible cover surface. This 45 , and an Arduino Due R3 is used as a circuit board to perform functionality is amply illustrated in Fig. 4g–h via external stimuli signal sampling and data processing. Data processing consists of (F = 3 N, f = 0.8 Hz, and α = 90 ) applied to different positions of two stages: system initialization and event detection via judging the flexible cover surface. From Fig. 4g, it can be observed that the whether the peak voltage of the TPTS satisfies the setting value. magnitude of the output voltage for sensing unit 1 decreases as This framework provides a triggering capability for turning on LED the applied load moves away from the center of sensing unit 1. As lights in a corresponding direction. The schematic digraph is shown in Fig. 4h, the output voltage of sensing unit 5 is not depicted in Fig. 5b. If two sensing units are larger than the setting maximized at the installation position. This is because the value at the same time, the electronic circuit drives the LED thickness of the flexible cover at the pole is the largest, reducing between the sensing units. The stimuli starts at the position the ability to transmit load-induced deformation. As a result, the between sensing units 2 and 3 and rotate in a clockwise manner output voltage response to a load at the pole is smaller than by 45 during each step of the experiment. Figure 5b shows how the TPTS is used to control the on/off state of LED lights (also see the output voltage response to a load around the pole. Due to the Supplementary Video 1) by a corresponding voltage, where the similar structure and symmetrical distribution of the sensing units, peak voltage in the stimuli direction is larger than in other the characteristic data of sensing units 2, 3, and 4 are depicted in directions. This result indicates the TPTS’ great potential in Supplementary Figs. 11–13. Using the output voltage response to approximating the external stimulation directions. different positions, the contact centroid location can be obtained for a sphere surface of radius R in ref. , as shown in Application of TPTS in detecting hardness Supplementary Fig. 2, with the following equation: To explore the compatibility of the TPTS with underwater vehicle applications, a manipulator integrated with the TPTS and mounted 2 T c ¼ ðK m þ Kf ´ m þðf mÞfÞ; on a remotely operated vehicle (ROV) is prepared to perform (1) K þkfk hardness detection of various silicone samples. As shown in Fig. 6a, b, npj Flexible Electronics (2022) 25 Published in partnership with Nanjing Tech University P. Xu et al. Fig. 6 Experiments with TPTS mounted on an ROV. a, b The configuration of the gripper and sensors for detecting the hardness of a silicone sample. c The output signals of the TPTS when gripping various silicone samples under the condition of β = 73 . d Demonstration of the TPTS in a closed-loop control system for an ROV and the corresponding output voltage signal. the TPTS is guided by bolts on the manipulator surface and is fixed at For specific underwater objects, the desired peak voltage its center. To determine the relationship between the gripping angle corresponding to the grasping force is provided, where the goal and the output signal for various silicone samples, the gripping angle is to grasp objects without breaking them. Once the peak voltage β is defined to represent the angle between the horizontal line and from the TPTS reaches the setting value, the gripper will stop the gripper location. The Shore hardness of the silicone samples is increasing β and maintain its current state until the object is 30A (Dragonskin30),2A (DragonskinPX),and 00− 20 (Ecoflex 20). released. For example, when grasping the sample with Dragon The results show that the magnitude of flexible cover deformation skin 30, the desired peak voltage is given as 0.77 V. As the output increases with the incremental Shore hardness of the samples under voltage reaches the setting value, the gripping motion holds on the condition of β= 73 , resulting in larger output peaks generated with β = 73 in T , as shown in Fig. 6dI (also see Supplementary by the TPTS in one single actuation motion in Fig. 6c. Furthermore, Video 3). For samples with Dragon skin PX and Ecoflex 20, the when the sample with Dragon skin 30 is gripped at the palm position gripper completes this task in T and T , respectively. From Fig. 6d, 2 3 of the gripper, the slope of the output signal in sensing unit 5 (k )is T < T < T results from the samples with incremental Shore 1 2 3 higher than the others, as shown in Supplementary Fig. 15. It is worth hardness. Therefore, different types of underwater objects may be noting that under the relatively fast gripping speed, the gripping recognized when visual cues are reduced or absent, verifying the motion can still be clearly recognized from the generated output feasibility of applying the TPTS to underwater vehicles. peaks of the sensing unit (see Supplementary Video 2). In addition, the TPTS can also distinguish the gripping and releasing motions of Application of TPTS in non-destructive underwater pipeline the gripper from the rising edge and falling edge of the output evaluation signal. As a result, the slope of voltage signal increases with the Experiments are performed in an indoor water pool (3m*2m*1.5m) incremental Shore hardness for sensing unit 5 (k > k > k ), proving 1 2 3 with the goal of non-destructive pipeline evaluation based on the the feasibility of hardness detection with the output signals of the TPTS. developed TPTS, as shown in Fig. 7a. For additional stability in the Subsequently, a simple closed-loop control system is built for pool-based experiments, the pipeline (2 m in length and 5 mm in grasping objects underwater, as shown in Supplementary Fig. 16. radius) with three ruptures to be detected is mounted on the floor. Published in partnership with Nanjing Tech University npj Flexible Electronics (2022) 25 P. Xu et al. Fig. 7 Experiments with non-destructive underwater pipeline evaluation. a The configuration of the pipeline and sensors in use of non- destructive underwater pipeline evaluation. b The process flows from signal sampling to non-destructive underwater pipeline evaluation. c Demonstration of TPTS in non-destructive underwater pipeline evaluation and its corresponding output voltage signal. At the beginning of the task, the ROV is operated to explore the a first step toward utilizing the highly sensitive directional sensing nearby environment until the vision system detects the pipe. capabilities of the TPTS. Moreover, an underwater sensory tactile Subsequently, the ROV uses the manipulator to grasp the system is built by mounting a manipulator integrated with the pipeline. The magnitude of the grasping force depends on TPTS on an ROV to obtain information about grasped objects. whether the TPTS is in contact with the pipe. The process flow Based on the improved intelligence of the gripper, it has the of non-destructive pipeline evaluation is depicted in Fig. 7b. ability to detect the hardness of samples and complete grasp tasks After grasping the pipeline, the ROV moves forward along the while preventing breakage. Additionally, the underwater sensory pipeline until it reaches the end of the pipeline. The screen tactile system is used to conduct non-destructive underwater displays the real-time signals, as shown in Fig. 7c(also see pipeline evaluation, illustrating its potential future application in Supplementary Video 5). When the presence of a falling edge various fields, including underwater object monitoring and tactical stays for a period of time during the detection process, it can be surveillance. judged that there is a fracture at the location of the ROV at this time. In addition, the roughness of the pipeline surface can be METHODS determined by observing the vibration magnitude of the real- time signals. Fabrication of TPTS As shown in Supplementary Fig. 17, the triboelectric sensing unit consists of six layers assembled by lamination: (i) a bottom 2 mm thick silicon DISCUSSION substrate made of Dragonskin 30, (ii) an intermediate 1 mm thick CPP, a commercial product, cut into square pieces (18 mm in length, 8 mm in In summary, a TPTS for underwater tactile perception based on width), (iii)–(iv) 0.3 mm thick FEP films sprayed with conductive ink serving triboelectric nanogenerators is proposed and investigated in this as a dielectric layer, (v) an intermediate 1 mm thick CPP, and (vi) a top 2 paper. The TPTS uses the linear relationship between the output mm thick silicone layer with an array of spinosums. The flexible cover and signal and parameters to sense the magnitude, frequency, and flexible support are made of Dragonskin 30. Specifically, 30 ml part A and contact area of external stimuli by analyzing the contact- 30 ml part B of silicone rubber are mixed in a petri dish. Then, a vacuum separation pattern and the triboelectric output of multiple sensing pump is used to vacuum the mixture to 0.1 MP for 2.5 min. When the units. In addition, the TPTS has a spinosum structure to measure evacuation of the mixture is finished, the mixture is poured into a mold and distinguish normal and shear forces generated when manufactured by 3D printing. The molds of the upper hatch cover and interacting with an object in real time. The TPTS is then used to fixed frame made from polylactic acid are also manufactured by 3D control the on/off states of lights oriented in various directions as printing. In addition, The conductive ink No. is CH-8(MOD2) that is npj Flexible Electronics (2022) 25 Published in partnership with Nanjing Tech University P. Xu et al. produced by JUJO printing supplies and technology (Pinghu) Co. Ltd. The 16. Ju, F. et al. A miniature piezoelectric spiral tactile sensor for tissue hardness conductive ink is printed on the FEP membrane by the screen-printing palpation with catheter robot in minimally invasive surgery. Smart Mater. Struct. technology, then dried in the atmosphere temperature. Finally, the flexible 28, 025033 (2019). cover, fixed frame, and upper hatch cover parts are assembled using an 17. Dehnhardt, G. & Mauck, B. Mechanoreception in secondarily aquatic vertebrates. elastomeric O-Ring seal and several screws. In Sensory evolution on the threshold. 295–314 (2008). 18. Niesterok, B., Kruger, Y., Wieskotten, S., Dehnhardt, G. & Hanke, W. Hydrodynamic detection and localization of artificial flatfish breathing currents by harbour seals Electric measurement and characterization (phoca vitulina). J. Exp. Biol. 220, 174–185 (2017). Field-emission scanning electron microscopy (Hitachi SU8010) is used to 19. Strobel, S. M., Sills, J. M., Tinker, M. T. & Reichmuth, C. J. Active touch in sea otters: characterize the surface morphology of the nanostructured FEP film in-air and underwater texture discrimination thresholds and behavioral strategies sprayed with conductive ink. For the electric output measurement of the for paws and vibrissae. J. Exp. Biol. 221, jeb181347 (2018). TPTS, a linear motor (LINMOT EI200-P01) with a frequency modulation 20. Beem, H., Hildner, M. & Triantafyllou, M. Characterization of a harbor seal whisker- program is used to drive the pressure sensor (LZ-WL2) to collision for inspired flow sensor. In 2012 Oceans.1–4 (2012). quantified investigation. A Keithley (6514) electrometer is adopted to 21. Beem, H., Liu, Y., Barbastathis, G. & Triantafyllou, M. Vortex-induced vibration measure the voltage signal. NI-6259 is used for data collection. The measurements of seal whiskers using digital holography. In OCEANS 2014-TAIPEI. software platform is constructed on the basis of LabVIEW, which is 1–4 (2014). capable of realizing real-time data acquisition control and analysis. For 22. Wegiriya, H., Herzig, N., Abad, S. A., Sadati, S. H. & Nanayakkara, T. A stiffness example, an intelligent display instrument shows the simulated external controllable multimodal whisker sensor follicle for texture comparison. IEEE Sens. load generated by changing the setting parameters of the linear motor. J. 20, 2320–2328 (2019). When the TPTS collides with the pressure sensor formed by the linear 23. Fan, F. R., Tian, Z.-Q. & Wang, Z. L. Flexible triboelectric generator. Nano Energy. 1, motor, the Keithley electrometer samples the electric signal and 328–334 (2012). transmits the data to the computer for visualization with LabVIEW 24. Xia, X., Fu, J. J. & Zi, Y. L. A universal standardized method for output capability software. assessment of nanogenerators. Nat. Commun. 10, 4428 (2019). 25. Wang, H., Cheng, J., Wang, Z. Z., Ji, L. H. & Wang, Z. L. Triboelectric nanogen- erators for human-health care. Sci. Bull. 66, 490–511 (2020). DATA AVAILABILITY 26. Xie, W. B. et al. A nonresonant hybridized electromagnetic-triboelectric nano- generator for irregular and ultralow frequency blue energy harvesting. Research. The datasets generated during and/or analyzed during the current study are available https://doi.org/10.34133/2021/5963293 (2021). from the corresponding author on reasonable request. 27. Wang, Z. L., Jiang, T. & Xu, L. Toward the blue energy dream by triboelectric nanogenerator networks. Nano Energy. 39,9–23 (2017). Received: 5 December 2021; Accepted: 20 March 2022; 28. Shi, Q., Wang, H., Wu, H. & Lee, C. K. Self-powered triboelectric nanogenerator buoy ball for applications ranging from environment monitoring to water wave energy farm. Nano Energy. 40, 203–213 (2017). 29. Shi, Q. F., Wu, H., Wang, H., Wu, H. X. & Lee, C. K. Self-powered gyroscope ball using a triboelectric mechanism. Adv. Energy Mater. 7, 1701300 (2017). REFERENCES 30. Guo, H. et al. A highly sensitive, self-powered triboelectric auditory sensor for social robotics and hearing aids. Sci. Robot. 3, eaat2516 (2018). 1. Paull, L., Saeedi, S., Seto, M. & Li, H. Sensor-driven online coverage planning for 31. He, Q. et al. Triboelectric vibration sensor for a human-machine interface built on autonomous underwater vehicles. IEEE ASME Trans Mechatron. 18,1827–1838 (2012). ubiquitous surfaces. Nano Energy. 59, 689–696 (2019). 2. Melo, J. & Matos, A. Survey on advances on terrain based navigation for auton- 32. Jin, T. et al. Triboelectric nanogenerator sensors for soft robotics aiming at digital omous underwater vehicles. Ocean Eng. 139, 250–264 (2017). twin applications. Nat. Commun. 11, 5381 (2020). 3. He, B. et al. Autonomous navigation for autonomous underwater vehicles based 33. Xu, P. et al. A triboelectric-based artificial whisker for reactive obstacle avoidance on information filters and active sensing. Sens. 11, 10958–10980 (2011). and local mapping. Research. 2021, 9864967 (2021). 4. Jakuba, M. V. et al. Toward automatic classification of chemical sensor data from 34. Luo, Y. et al. Triboelectric bending sensor based smart glove towards intuitive autonomous underwater vehicles. In 2011 IEEE/RSJ International Conference on multi-dimensional human-machine interfaces. Nano Energy. 89, 106330 Intelligent Robots and Systems. 4722–4727 (2011). (2021). 5. Lin, Y. H., Shou, K. P. & Huang, L. J. The initial study of lls-based binocular stereo- 35. Li, J. R. et al. Triboelectric nanogenerators enabled internet of things: a survey. vision system on underwater 3d image reconstruction in the laboratory. J. Mar. Intelligent Converged Netw. 1, 115–141 (2020). Sci. Technol. 22, 513–532 (2017). 36. Kumar, K. S. et al. Stretchable capacitive pressure sensing sleeve deployable onto 6. Bruno, F., Bianco, G., Muzzupappa, M., Barone, S. & Razionale, A. V. Experi- catheter balloons towards continuous intra-abdominal pressure monitoring. mentation of structured light and stereo vision for underwater 3d reconstruction. Biosensors. 11, 156 (2021). ISPRS J. Photogramm. Remote Sens. 66, 508–518 (2011). 37. Wu, Z. Y. et al. Multifunctional sensor based on translational-rotary triboelectric 7. Dimitratos, S. D. et al. Biosensors to monitor water quality utilizing insect odorant- nanogenerator. Adv. Energy Mater. 9, 1901124 (2019). binding proteins as detector elements. Biosensors. 9, 62 (2019). 38. Cheng, B. L. et al. Mechanically asymmetrical triboelectric nanogenerator for self- 8. Kröger, S. & Law, R. J. Biosensors for marine applications: We all need the sea, but powered monitoring of in vivo microscale weak movement. Adv. Energy Mater. does the sea need biosensors? Biosens. Bioelectron. 20, 1903–1913 (2005). 10, 2000827 (2020). 9. Grenon, G., An, P. E., Smith, S. M. & Healey, A. J. Enhancement of the inertial 39. Bicchi, A., Salisbury, K. J. & Brock, D. L. Contact sensing from force measurements. navigation system for the morpheus autonomous underwater vehicles. J. Ocean. Int. J. Robot. Res. 12, 249–262 (1993). Eng. 26, 548–560 (2001). 10. Panish, R. & Taylor, M. Achieving high navigation accuracy using inertial navi- gation systems in autonomous underwater vehicles. In OCEANS 2011 1–7 (IEEE, Spain, 2011). ACKNOWLEDGEMENTS 11. Lee, P. M. & Jun, B. H. Pseudo long base line navigation algorithm for underwater The authors would like to thank the National Key R & D Project from the Minister of vehicles with inertial sensors and two acoustic range measurements. Ocean Eng. Science and Technology (2021YFA1201604), the Dalian Outstanding Young Scientific 34, 416–425 (2007). and Technological Talents Project (2021RJ11), the National Natural Science 12. Rooney, T., Pipe, A. G., Dogramadzi, S. & Pearson, M. Towards tactile sensing Foundation of China (62003175, 51879022), the Beijing Natural Science Foundation applied to underwater autonomous vehicles for near shore survey and de- (No. 4192026), and the Academy of Finland (Grant No. 315660). mining. In Conference Towards Autonomous Robotic Systems. 463–464 (2012). 13. Muscolo, G. G. & Cannata, G. A novel tactile sensor for underwater applications: limits and perspectives. In OCEANS 2015-Genova.1–7 (2015). AUTHOR CONTRIBUTIONS 14. Yohan, N. et al. A three-axial body force sensor for flexible manipulators. In 2014 IEEE International Conference on Robotics and Automation (ICRA). 6388–6393 M.X. designed the experiments. J.L. and X.L. synthesized the targets. P.X., J.L., and X.L. (2014). performed sample growth. P.X., J.L., X.W., and S.W. performed magnetism 15. Sun, Y., Liu, F., Yuan, Z. P., Huang,W., M. & Wang, B. W. A novel three-axial force measurements. P.X., J.L., X.L., J.Z., and T.C. performed sample structural characteriza- tactile sensor based on the fringing effect of electric field. IEEE Trans. Magn. 55, tion and data analysis. H.W. and C.W. performed theoretical calculations. P.X. wrote 1–5 (2019). the manuscript with significant contributions from J.L., X.W., J.Z., T.C., H.W., and C.W. Published in partnership with Nanjing Tech University npj Flexible Electronics (2022) 25 P. Xu et al. as well as contributions from all other authors. M.X., J.T., and G.X. supervised the Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims project. in published maps and institutional affiliations. COMPETING INTERESTS Open Access This article is licensed under a Creative Commons The authors declare no competing interests. Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative ADDITIONAL INFORMATION Commons license, and indicate if changes were made. The images or other third party Supplementary information The online version contains supplementary material material in this article are included in the article’s Creative Commons license, unless available at https://doi.org/10.1038/s41528-022-00160-0. indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory Correspondence and requests for materials should be addressed to Guangming Xie, regulation or exceeds the permitted use, you will need to obtain permission directly Jin Tao or Minyi Xu. from the copyright holder. To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/. Reprints and permission information is available at http://www.nature.com/ reprints © The Author(s) 2022 npj Flexible Electronics (2022) 25 Published in partnership with Nanjing Tech University

Journal

npj Flexible Electronics – Springer Journals

Published: Apr 25, 2022

There are no references for this article.