Open Access
Add to BibSonomy
Abstract
In this paper, an omnidirectional underwater wireless optical communication (UWOC) system is proposed, including six lens-free transceivers. An omnidirectional communication with a data rate of 5 Mbps in a 7-m underwater channel is experimentally demonstrated. The optical communication system is integrated into a self-designed robotic fish, and the signal is processed real-time through an integrated micro-control unit (MCU). In addition, it is experimentally demonstrated that the proposed system could establish a stable communication link between two nodes, regardless of the nodes’ locomotion and attitude, with a data rate of 2 Mbps and a communication range up to 7 m. In particular, the optical communication system features small footprint and low power consumption, which is suitable for integration in autonomous underwater vehicle (AUV) swarms to achieve an omnidirectional information transmission with low latency, high security, and high data rate compared with its acoustic counterpart.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Humans have never stopped exploring and exploiting the ocean. With the rapid development of technology, more and more advanced equipment have been applied to improve exploration capabilities and expand the range of activities. Using a swarm of low-cost and compact autonomous underwater vehicles (AUVs) instead of a single but complex AUV has become a trend with the advantages of high efficiency and robustness [1–3]. AUV swarms are widely used in underwater resource extraction, search and rescue, maintenance of underwater infrastructures, and exploration of susceptible ecological environments [4–6]. Similar to unmanned aerial vehicles, underwater swarm robotics require stable and reliable inter-vehicle communications for data exchange [7–9]. One of the significant challenges is the lack of effective communication methods associated with underwater swarm robotics. Technologies including high-frequency radio, infrared radiation, and radar which are widely used in the air, are not so applicable due to their high attenuation in the water [10]. Underwater acoustic communication (UAC) is mature and widely used, with high reliability and long communication distance [11]. Using a miniature acoustic communication module, a soft robotic fish named SoFi was proposed, which could receive control commands with a data rate of 20 bps within 15 m [12]. However, the narrow bandwidth and high transmission delay of UAC make it insufficient when the swarm size is large [13,14]. Underwater wireless optical communication (UWOC), owing to its high data rate, low latency, and high power efficiency, is an ideal alternative for underwater swarm robotics to establish broadband inter-vehicle communication [15,16].
In recent decades, significant progress has been made in UWOC. Researchers mainly focused on increasing the data rate and extending the transmission distance of UWOC systems [17–19]. To achieve such goals, laser diode (LD)-based system is an ideal choice due to its high bandwidth and small divergence angle. In 2019, a 500-Mbps UWOC link through a 100-m tap-water channel was established with nonlinear equalization using a green LD and on-off keying (OOK) modulation [20]. In 2021, Y. Dai successfully conducted a 200-m/500-Mbps UWOC experiment using a sparse nonlinear equalizer [21]. In 2022, C. Fei experimentally demonstrated a 100-m/3-Gbps UWOC system using a wideband photomultiplier tube (PMT) [22]. These significant progresses show the great potential of UWOC.
However, link alignment between the transmitter and the receiver is indispensable for achieving stable information transmission. Even worse, most of the reported LD-based systems require high-precision alignment, which is extremely challenging in practical underwater dynamic environments. Recently, some studies on alignment tolerance in UWOC were conducted. The alignment requirement can be relaxed by using high-sensitivity arrayed receivers [23] or using light-emitting diodes (LEDs) as the light source. J. Rao achieved data rates of 4 Mbps and 1 Mbps over 8.4-m and 22-m underwater channels, respectively, by using an LED array with a convergent optical system [24]. In 2013, a 38-Kbps UWOC system with a large communication angle was demonstrated in a swimming pool at a distance of 7 m using a single blue LED and a photodiode [25]. In 2019, a single photon avalanche diode (SPAD) was adopted by [26] as a receiver to achieve 1.5-m/1-Mbps real-time communication. In 2020, [27] adopted a PMT as a receiver and achieved 3.7-m/500-Kbps offline communication in the lake. In [28], the UWOC system based on FPGA is designed, which can realize 10-m/1-Mbps full-duplex real-time communication. In 2022, an avalanche photodiode (APD) was adopted by [29] as a receiver to achieve error-free communication of up to 120 Mbps in 30-m air. More than 50-m UWOC in the deep sea can also be realized with a communication rate not lower than 80 Mbps. Such design makes it much easier to establish a UWOC link in a practical underwater environment especially when the transmitter and the receiver keep moving. However, due to the absorption, scattering, and line-of-sight propagation properties of light, the LED-based systems usually introduce optical lens groups to obtain higher power optical signal at the receiver to offset the geometric loss brought by large divergence angles, which undoubtedly increases the difficulty of alignment [30].
The high requirement on link alignment limits its application, which is further exacerbated by the constantly changing locomotion and attitude of the AUVs. Therefore, it is necessary to design an LED array and a detector array to realize omnidirectional emission and detection of optical signals. The CoCoRo underwater swarm project introduces omnidirectional UWOC to swarm robotics [31]. The AUV named Jeff developed by this project is integrated with 6 sets of optical communication modules, each of which contains two photodiodes and two pairs of LEDs with different divergence angles, thus enabling an omnidirectional emission and detection of optical signals within 1 m [32]. In addition, [33] proposed an underwater robot that integrates two vertically placed LEDs and a pair of cameras with a nearly full field of view (FOV), introducing signaling through composite behavior, where a robot flashes its LEDs at 15 Hz and other robots detect that. Although these studies established the omnidirectional optical communication link, its data rate and transmission distance are limited. In addition, hybrid optical-acoustic communications were also studied to combine the advantages of UAC’s long distance with UWOC’s high data rate [34]. Using UAC for remote control and assisted positioning, a stable optical communication link was established to transmit large amounts of data [35]. Nevertheless, the aforementioned systems still need acquisition and tracking units to establish and maintain the link. Such units usually contain moving parts, making it hard to miniaturize the transceiver, which is crucial for robots with limited size, weight, and energy consumption.
This paper proposes an omnidirectional UWOC system for underwater swarm robotics to establish stable, reliable, and high-speed inter-vehicle communication links. The system uses LEDs with large divergence angles as light sources and photodiodes with large FOVs as detectors. Six groups of light sources and detectors form the LED array and the detector array, respectively, to achieve an omnidirectional optical signal emission and detection. A beam-shaping structure is proposed to unify the spatial radiation distribution of the LED array. Thanks to the uniform spatial radiation distribution, the detection responses at different angles are roughly equal, greatly relaxing the alignment requirement. Together with the remarkable increase in signal-to-noise ratio (SNR) brought by an equal gain combined (EGC) circuit, a 7-m omnidirectional UWOC system with a data rate of up to 5 Mbps was successfully achieved. When the data rate is reduced to 2 Mbps, the transmission distance of the system can reach 9.5 m. The measured optical power at the transmitter and the receiver is 26.98 dBm and -30.58 dBm, respectively, implying an impressive link budget of 57.56 dB, which is necessary to compensate for the large geometric loss in an omnidirectional UWOC system. Finally, the system was integrated into the self-designed robotic fish to explore its applicability in underwater swarm robotics. Real-time modulation and demodulation is realized by a micro-control unit (MCU) inside the robotic fish. The results show that the robotic fish, with continues changing in locomotion and attitude, can maintain a stable communication link with the fixed node within 7 m. Information or control commands can be transmitted at a data rate of 2 Mbps without errors. The proposed omnidirectional UWOC system is compatible with the serial communication protocol, which reduces the requirement for hardware equipment as well as system complexity. Compared with the reported omnidirectional UWOC system for swarm robotics, the proposed scheme achieves significant improvements in data rate and transmission distance.
The rest of this paper is organized as follows: Section 2 introduces the principles of each part of the circuit in detail. Section 3 describes the implementation of omnidirectional transmitter and receiver. A series of experiments that were carried out to examine the BER performance of the system are described in Sec. 4. Finally, Sec. 5 summarizes the design methods and improvements for the system.
2. System implementation
2.1 Omnidirectional illuminance distribution
The system consists of six sets of optical communication modules, each of which is composed of an LED and a photodiode. The modules are placed on the six sides of the robotic fish to form the LED array and the photodiode array, as shown in Fig. 1. As the distances between different modules and the geometric center of the robotic fish are not the same, the effect on the spatial radiation distribution of the LED array and the FOV of the photodiode array should be considered.
Fig. 1. Optical communication module distribution. Inset: (i) the module with the beam-shaping structure.
The light source is a high-power LED with a central wavelength of 450 nm. The radiation distribution of a single LED measured at a distance of 1 m is shown as the solid line in Fig. 2(a). It can be seen that the radiation decreases slowly as the angle increases. The measured full width at half maximum (FWHM) divergence angle of an LED is approximately 130°. As shown in Fig. 1, No.1 and No. 3 modules are 150 mm away from the center, No. 2 and No. 4 modules are 50 mm away from the center, and No.5 and No.6 modules are 70 mm away from the center. The expected distance of the proposed optical communication system is from 0.5 m to 7 m, so it can be considered that the distances from No. 2, No. 4, No. 5, and No. 6 modules to the center are approximately equal. Only the horizontal plane is considered when measuring the radiation intensity of multiple LEDs. The calculated normalized illuminance distribution generated by No. 1 to No. 4 LEDs is shown as the solid line in Fig. 2(b). The results show that the spatial radiation distribution is uneven. As the FWHM divergence angle of a single LED exceeds 120°, the radiation illuminance at the overlapping area of two LEDs exceeds that of a single LED. Such an uneven distribution will make the signal response unstable when the robotic fish’s attitude changes, thus affecting the performance of the UWOC system.
Fig. 2. (a) Measured normalized illuminance distribution of a single LED with and without the beam-shaping structure, (b) measured normalized illuminance distribution of four LEDs in the horizontal section with and without the beam-shaping structure.
A beam-shaping structure is proposed to solve the problem of uneven radiation distribution, as shown in Fig. 3. It can be seen as a circular aperture placed on the light-emitting surface of the LED, as shown in Fig. 3(c). Taking the right edge of the emitting surface as the reference point and setting the desired divergence angle to be ±45°, it can be derived that:
where k is the distance from the edge of the emitting surface to the internal wall of the aperture, and h is the height of the aperture. As the emitting surface is relatively large (with a length g of 4 mm), it cannot be regarded as a point source, and thus the beam emitted from the center point of the surface will exceed the limiting angle. For this reason, the point in the middle of the right half of line segment g is set as the reference point, and Eq. (1) can be rewritten as:Fig. 3. (a) Optical communication module, (b) the module with beam-shaping structure, (c) the sectional view of the beam-shaping structure.
Finally, k is set to 1.5 mm and h to 2.5 mm. The designed beam-shaping structure is manufactured using 3D printing, and its effect on the radiation distribution is measured. The radiation distribution of a single LED with the structure is shown as the dash line in Fig. 2(a). The results demonstrate that the radiation changes slowly from -45° to +45° and decreases rapidly once the angle exceeds this range. The spatial radiation distribution of four LEDs is shown as the dash line in Fig. 2(b). It can be concluded that the radiation distribution is roughly uniform. The calculated results show that the maximum value of the illuminance is about 1-dB higher than the minimum value. Therefore, only the case of a single LED with beam-shaping structure emitting vertically needs to be considered when measuring the performance of the optical communication system.
2.2 UWOC transmitter
The transmitter of the UWOC system includes a signal modulation unit, a driving circuit, and LEDs, as shown in Fig. 4. For compatibility with the robot's control system, commercially available STM32 series MCUs are adopted for signal processing in this paper. Considering the limited computing resources provided by the MCU, the simple but robust non-return-to-zero on-off keying (NRZ-OOK) is chosen as the signal modulation format of the system. Due to the limited load capacity of the MCU, it cannot be used to drive high-power LEDs directly. Therefore, it is necessary to design an LED driving circuit.
A metal-oxide-semiconductor field-effect transistor (MOSFET)-based driving scheme is adopted in this work as shown in Fig. 5. Q1 is an enhanced N-channel MOSFET. R1 and C5 form a resistor-capacitor based pre-equalizer, which can compensate for high-frequency attenuation and improve the modulation bandwidth at the transmitter. The parasitic capacitance between the gate-source and the gate-drain affects the switching speed of the MOSFET. Therefore, a pull-down resistor R2 is connected to the gate of Q1. The charge at the gate could release quickly when the input flips from high to low. R3 is a current-limiting resistor to prevent the MOSFET overheating. There are three sets of LED arrays in the transmitter (two LEDs in each set), which are driven by three MOSFETs. Moreover, to effectively integrate with the serial communication protocol, a reverse circuit is designed to solve the problem that the LED keeps glowing when the serial port is idle. This scheme reduces the complexity of the circuits and saves power consumption.
2.3 UWOC receiver
A positive-intrinsic-negative (PIN) photodiode (Thorlabs, FDS1010) is used as the detector. A reverse bias voltage circuit is designed to increase the response speed of the PIN photodiode, as shown in Fig. 6. Due to the large internal resistance of the PIN photodiode, the output current is very small, which is usually in the order of microamps. Therefore, a low-noise amplifier (LNA) with a high gain is needed. The amplifier designed in this paper consists of two cascaded wide-band amplifiers (ERA-8SM+, ERA-4SM+), as shown in Fig. 6. The gain of the circuit is about 46-dB in the frequency range of 0.1-10 MHz. The output noise of LNA is 2.2 mVRMS in a dark environment and 2.8 mVRMS in a working environment (68 lux).
A low pass filter (LPF) is used to filter out the high frequency noise. It is designed with reference to the Butterworth filter, which has a flat frequency response in the passband. A 7-order normalized Butterworth LPF is used as the reference filter with a cut-off frequency of 1⁄(2π) Hz and a characteristic impedance of 1 Ω. The circuit diagram of the normalized Butterworth LPF is shown in Fig. 7(a), and the values of the components are calculated as follows:
where n is the order and k is the number of components. Further, the values of the elements with a cut-off frequency of 10 MHz and a characteristic impedance of 50 Ω are calculated. Ultimately, the calculated values and finally used values of the inductances and capacitances are listed in Table 1. The LPF was fabricated and measured, and the frequency response is shown in Fig. 7(b).
Table 1. Parameters of each component on the LPF
The amplitude of the signal output from the LNA is only tens of millivolts, which could not reach the ideal dynamic range. A two-stage cascaded operational amplifier (OPA) is utilized to amplify the LPF output signal further, as shown in Fig. 8. By optimizing the resistance values of R4 and R5, the output of the OPA is adjusted to the appropriate range. The output noise of OPA is 65 mVRMS in a dark environment and 76 mVRMS under daylight illumination conditions (68 lux).
The signal processing circuit of a single detector is illustrated in Fig. 6 to Fig. 8. To realize omnidirectional optical detection, six detectors are placed on the robotic fish to receive the optical signals from different directions. An addition circuit is designed to combine the output signals of six photodiodes, as shown in Fig. 9. The relationship between the output voltage and the input voltage is:
In order to simplify the system, the six signal paths are combined with equal gain, so the input resistances R6 to R11 are set to be the same. The output noise of the addition circuit is 227 mVRMS in a dark environment and 242 mVRMS in the working environment (68 lux). It is worth mentioning that an adjustable direct current voltage is connected to the positive port to control the amplitude range of the output signal, which is convenient for the post-stage to compensate for the signal distortion. Due to the influence of the underwater channel and device bandwidth, the signal will be distorted when it reaches the receiver. Therefore, a comparator is designed to restore the signal to the standard OOK format. The comparator works on the same principle as the reverse circuit at the transmitter. The difference is that the signal is connected to the positive port, and the reference voltage can be adjusted to be equal to the average value of the signal to improve the performance of the UWOC system.
3. Experimental setup
The robotic fish platform carrying the proposed omnidirectional UWOC system is shown in Fig. 10. It is a miniature AUV with a size of 300 ${\times} $ 140 ${\times} $ 100 $\textrm{m}{\textrm{m}^3}$ and a weight of 2.4 kg, which integrates the omnidirectional UWOC system to establish communication links with neighboring vehicles or underwater fixed nodes for information exchange. The designed printed circuit board (PCB) of circuit diagram is shown in Fig. 10, which includes the optical communication board and control board. The optical communication board is subdivided into a mainboard and six vice-boards. The LPF and the front stage circuit of the receiver are designed on the vice-board with a smaller area so that they can be placed as close as possible to the photodiode, which could shorten the path between the photodiode and the signal port and reduce the noise introduced in the process of weak signal transmission as much as possible.
Fig. 10. Robotic fish system overview: (a) shell, (b) bow propulsion unit, (c) aft propulsion units, (d) motor driver, (e) optical communication module, (f) PCBs of communication board and control board, (g) buoyancy system, (h) battery pack, (j) pressure sensor.
The experimental setup of the proposed omnidirectional UWOC system is illustrated in Fig. 11. The experiment was conducted in a 7-m water tank, filled with tap water. Firstly, a pseudo-random binary sequence with a length of $2 \times {10^5}$ was generated offline by MATLAB and loaded into an arbitrary waveform generator (AWG, Tektronix 70002A) to generate a baseband electrical signal with OOK format. The N-channel MOSFET-based driver was used to drive the LED to convert the electrical signal to an optical signal. Secondly, four PIN photodiodes are fixed on a rotatable platform, and their relative position on the platform is consistent with that on the robotic fish. The signal was captured by an oscilloscope (OSC, Tektronix 71254C) and demodulated offline by a computer to calculate the BER. The receiver was placed on one side of the water tank, and the transmitter was placed in the water tank. The transmission performance over different distances was measured by changing the position of the LED. Likewise, the transmission performance at different data rates and angles was measured. Finally, the omnidirectional UWOC system was integrated into the robotic fish, and the MCU was used to replace the AWG and OSC for real-time signal procession. The transmitter repeatedly sent data packets with fixed lengths at short intervals. The receiver detected and demodulated the signal. The BER was calculated and transmitted to the upper computer through serial communication.
Fig. 11. Experimental setup of the proposed omnidirectional optical communication system. Insets: (i) receiver, (ii) the water tank, (iii) transmitter.
4. Results and discussion
At first, the frequency response of the proposed system was measured. As the designed driving circuit utilizes the on/off characteristics of MOSFETs, it can only output high and low levels. Therefore, square wave signals with different frequencies were generated using an AWG, with a step size of 0.1 MHz, to measure the frequency response of the system in the range of 0.1-13 MHz. The result is shown in Fig. 12, indicating that the -3-dB bandwidth is approximately 2.6 MHz.
Fig. 12. Normalized frequency response of the system. Insets: waveforms under (i) 0.2 MHz, (ii) 2 MHz, (iii) 6.5 MHz.
Secondly, the transmitter was placed in the water tank 7 m away from the receiver. The BER with different data rates is shown in Fig. 13(a). The maximum data rate of the system under the forward error correction (FEC) threshold is 5.8 Mbps. Subsequently, the maximum data rates of the system at different distances were measured by changing the transmission distance of the signal. As shown in Fig. 13(b), the maximum data rate of the system is increased from 5.8 Mbps to 7.8 Mbps when the distance decreases from 7 m to 1 m.
Fig. 13. (a) BER performance versus data rate, (b) data rate versus distance under the FEC threshold. Insets: eye diagrams under (i) 7 m/6.8 Mbps, (ii) 7 m/4.6 Mbps, (iii) 3 m/7.8 Mbps.
Moreover, the transmission performance was measured with different reception angles, with a 4-MHz square wave being sent from the transmitter. For each of the four detectors, the response was measured when the platform rotated from 0° to 350°. The output peak-to-peak voltage of each of the four detectors is shown in Fig. 14(a). It can be seen that the peak-to-peak voltages of No. 1 and No. 3 detectors are slightly higher than those of No. 2 and No. 4 detectors because they are closer to the transmitter. The output peak-to-peak voltage of the addition circuit is also shown in Fig. 14(a). It can be seen that the voltage is slightly higher at the reception angles where two detectors can detect the signals simultaneously. Furthermore, the BER performance of the system is measured with the addition circuit. The BER fluctuates in a small range at different angles. And it is better at angles where two detectors can detect the optical signal, such as 45° and 135°. The result can be explained from two aspects. Firstly, the output voltage is higher at such reception angles. Secondly, according to the principle of the EGC algorithm, the performance is optimal when the SNR of the received signals from each path is close to each other [36]. The results demonstrate that the designed omnidirectional UWOC system can achieve a maximum data rate of 5 Mbps within a distance of 7 m at any angle of the receiver.
Fig. 14. (a) The output peak-to-peak voltage of each detector and the addition circuit versus the reception angle, (b) the BER performance of four detectors combined with equal weights versus the reception angle.
In addition, the experiments were conducted in the air to measure the maximum transmission distance. The results are shown in Fig. 15(c). It can be seen that the transmission distance of the system in the air channel can reach 14 m at a data rate of 2 Mbps. The optical power at the receiver is -30.58 dBm measured by an optical power meter (Thorlabs PM100) with a detection area of 78.5 $\textrm{m}{\textrm{m}^2}$. The BER performance is worse when the data rate is 1 Mbps because part of the signal processing circuit of the receiver is coupled by several 100-nF capacitors, including the LNA’s input capacitor C6 (as shown in Fig. 6), output capacitor C7, and cascaded capacitor C8, as well as the OPA’s cascaded capacitor C9 (as shown in Fig. 8). These capacitors attenuate the low-frequency signal. The attenuation coefficient of water was measured by a blue laser and an optical power meter, as shown in Fig. 15(b). The value is 0.444 dB/m (0.102 ${\textrm{m}^{ - 1}}$) after removing the geometric loss. The estimated transmission distance with the same received optical power in the underwater channel is 9.5 m. Therefore, the maximum communication distance of the system in the water is 9.5 m with a data rate of 2 Mbps.
Fig. 15. (a) Received optical power versus distance in the air, (b) received optical power versus distance in the water, (c) BER performance in the air versus distance, with different data rates. Inset: (i) eye diagram under 14 m/2 Mbps.
Finally, the modulation and demodulation units of the system were replaced with the MCU (STM32F103RCT6). The data packet was set to 100 bytes in length, where the first five bytes are redundant symbol. The 6th byte is the start symbol. The 7th to 9th bytes are identification symbol, followed by 90 bytes of valid data, and the last byte is the stop symbol, as shown in Fig. 16. The redundant symbol solves the problem that the first few bytes are prone to errors after a long idle time and ensure that the circuit is in an ideal state when demodulating valid information. The identification symbol indicates the source and destination of the data packet. The start and stop symbols are one byte long, helping the MCUs confirm the range of valid data information. The serial port’s baud rate was set to 2 MBaud, and the data packet is sent every 50 ms. The distance between the transmitter and the receiver is 7 m. The results show that the packet loss rate is less than 10-3 when continuously demodulating 105 data packets. Such a real-time communication link is reliable, and the rare packet loss can be easily resolved through information retransmission. The transmission performance of the system was then tested on the robotic fish platform. The robotic fish sent a data packet every 50 ms. A fixed UWOC node was placed at one side of the water tank to receive the packet. The signal-sending task was implemented through timer interrupts, with a special timer configured to execute the task every 50 ms. The signal detection task was implemented through receiving interrupts, which are triggered when signals are detected. At the beginning, the fixed node and the underwater robot were placed together, and a synchronization command was sent by a third device. Both devices received the command simultaneously. Then one of them reset the timer, and the other set the timer’s initial value to the one corresponding to 25 ms. In this way, the two devices sent signals alternately without affecting each other. The synchronization error of this method is less than 50 µs. Compared with the 25-ms frame interval, the negative impact on communication performance is negligible. It is also possible to change the movement state of the robotic fish by sending control commands through the fixed node. The process of the experiment is shown in Fig. 17. Only four LEDs and photodiodes were assembled on the horizontal plane of the robotic fish. It should be mentioned that the distance between the robotic fish and the fixed node should be larger than 0.5 m to prevent entering the blind area.
5. Conclusion
In this paper, an omnidirectional UWOC system for underwater swarm robotics is proposed, and its feasibility is verified by using a self-designed robotic fish. To integrate into AUVs with limited size and with limited computational power, the classical OOK modulation scheme is used. The driving and signal processing circuits are designed to improve the communication performance of the system. Moreover, the spatial radiation distribution is analyzed and improved so that it is uniform at all angles. Experiments were implemented to measure the communication performance of the system under different distances, data rates, and reception angles. The results show that the system can realize omnidirectional communication at 7 m with a data rate of 5 Mbps under offline processing. The maximum communication distance at a data rate of 2 Mbps is 9.5 m. Eventually, the system was integrated into the robotic fish. The experiments prove that a stable and reliable communication link can be achieved between a fixed node and the robotic fish, regardless of its attitude. The performance of the omnidirectional optical communication system is sufficient, and there is no need to introduce any optical lens to reshape the beam. It is worth mentioning that this work uses TDM techniques to establish a bidirectional link. In the future, it is expected to access more vehicles so that each vehicle in the swarm can transmit information without affecting the others.
Funding
National Key Research and Development Program of China (2022YFB2903403, 2022YFC2808200); National Natural Science Foundation of China (61971378); Strategic Priority Research Program of the Chinese Academy of Sciences (XDA22030208); Zhoushan-Zhejiang University Joint Research Project (2019C81081).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. A. E. Hassanien and E. Emary, Swarm Intelligence: Principles, Advances, and Applications (CRC press, 2018). [CrossRef]
2. Y. Zhu and X. Tang, “Overview of swarm intelligence,” in 2010 International Conference on Computer Application and System Modeling (ICCASM 2010) (2010).
3. R. Cui, S. S. Ge, B. V. E. How, and Y. S. Choo, “Leader–follower formation control of underactuated autonomous underwater vehicles,” Ocean Eng. 37(17-18), 1491–1502 (2010). [CrossRef]
4. J. S. Jaffe, P. J. Franks, P. L. Roberts, D. Mirza, C. Schurgers, R. Kastner, and A. Boch, “A swarm of autonomous miniature underwater robot drifters for exploring submesoscale ocean dynamics,” Nat. Commun. 8(1), 14189 (2017). [CrossRef]
5. F. Schill, A. Bahr, and A. Martinoli, “Vertex: A new distributed underwater robotic platform for environmental monitoring,” in Distributed Autonomous Robotic Systems: The 13th International Symposium (2018). [CrossRef]
6. P. Zahadat and T. Schmickl, “Division of labor in a swarm of autonomous underwater robots by improved partitioning social inhibition,” Adapt. Behav. 24(2), 87–101 (2016). [CrossRef]
7. S. Zhu, X. Chen, X. Liu, G. Zhang, and P. Tian, “Recent progress in and perspectives of underwater wireless optical communication,” Prog. Quantum Electron. 73, 100274 (2020). [CrossRef]
8. M. A. Khalighi and M. Uysal, “Survey on free space optical communication: A communication theory perspective,” IEEE Commun. Surv. Tutorials 16(4), 2231–2258 (2014). [CrossRef]
9. Y. Tan, Handbook of Research on Design, Control, and Modeling of Swarm Robotics (IGI Global, 2016).
10. M. A. Joordens and M. Jamshidi, “Consensus control for a system of underwater swarm robots,” IEEE Systems Journal 4(1), 65–73 (2010). [CrossRef]
11. M. Stojanovic and J. Preisig, “Underwater acoustic communication channels: Propagation models and statistical characterization,” IEEE Commu. Mag. 47(1), 84–89 (2009). [CrossRef]
12. R. K. Katzschmann, J. DelPreto, R. MacCurdy, and D. Rus, “Exploration of underwater life with an acoustically controlled soft robotic fish,” Sci. Robot. 3(16), 3449 (2018). [CrossRef]
13. P. Kumar, V. K. Trivedi, and P. Kumar, “Recent trends in multicarrier underwater acoustic communications,” IEEE Underwater Technol. 1, (2015).
14. C. M. Gussen, P. S. Diniz, M. L. Campos, W. A. Martins, F. M. Costa, and J. N. Gois, “A survey of underwater wireless communication technologies,” J. Commun. Inf. Sys 31, 242 (2016). [CrossRef]
15. H. J. Son, H. S. Choi, N. H. Tran, J. H. Ha, D. H. Ji, and J. Y. Kim, “Study on underwater wireless communication system using LED,” Mod. Phys. Lett. B 29(06n07), 1540023 (2015). [CrossRef]
16. M. A. Khalighi, C. Gabriel, T. Hamza, S. Bourennane, P. Leon, and V. Rigaud, “Underwater wireless optical communication; recent advances and remaining challenges,” in International Conference on Transparent Optical Networks (ICTON) (2014).
17. S. Hu, L. Mi, T. Zhou, and W. Chen, “35.88 attenuation lengths and 3.32 bits/photon underwater optical wireless communication based on photon-counting receiver with 256-PPM,” Opt. Express 26(17), 21685 (2018). [CrossRef]
18. C. Lu, J. Wang, S. Li, and Z. Xu, “60 m/2.5 Gbps underwater optical wireless communication with NRZ-OOK modulation and digital nonlinear equalization,” in Conference on Lasers and Electro-Optics (CLEO) (2019).
19. X. Chen, W. Lyu, Z. Zhang, J. Zhao, and J. Xu, “56-m/3.31-Gbps underwater wireless optical communication employing Nyquist single carrier frequency domain equalization with noise prediction,” Opt. Express 28(16), 23784 (2020). [CrossRef]
20. J. Wang, C. Lu, S. Li, and Z. Xu, “100 m/500 Mbps underwater optical wireless communication using an NRZ-OOK modulated 520 nm laser diode,” Opt. Express 27(9), 12171 (2019). [CrossRef]
21. Y. Dai, X. Chen, X. Yang, Z. Tong, Z. Du, W. Lyu, C. Zhang, H. Zhang, H. Zou, and Y. Cheng, “200-m/500-Mbps underwater wireless optical communication system utilizing a sparse nonlinear equalizer with a variable step size generalized orthogonal matching pursuit,” Opt. Express 29(20), 32228 (2021). [CrossRef]
22. C. Fei, Y. Wang, J. Du, R. Chen, N. Lv, G. Zhang, J. Tian, X. Hong, and S. He, “100-m/3-Gbps underwater wireless optical transmission using a wideband photomultiplier tube (PMT),” Opt. Express 30(2), 2326 (2022). [CrossRef]
23. M. Zhao, X. Li, X. Chen, Z. Tong, W. Lyu, Z. Zhang, and J. Xu, “Long-reach underwater wireless optical communication with relaxed link alignment enabled by optical combination and arrayed sensitive receivers,” Opt. Express 28(23), 34450 (2020). [CrossRef]
24. J. Rao, W. Wei, F. Wang, and X. Zhang, “An underwater optical wireless communication system based on LED source,” in Photonics and Optoelectronics Meetings (POEM)170 (2012).
25. B. Tian, F. Zhang, and X. Tan, “Design and development of an LED-based optical communication system for autonomous underwater robots,” in IEEE/ASME International Conference on Advanced Intelligent Mechatronics1558 (2013).
26. Q. Yan, Z. Li, Z. Hong, T. Zhan, and Y. Wang, “Photon-counting underwater wireless optical communication by recovering clock and data from discrete single photon pulses,” IEEE Photonics J. 11(5), 1–15 (2019). [CrossRef]
27. W. Wei, C. Zhang, W. Zhang, W. Jiang, C. Shu, and Q. Xiaorui, “LED-based underwater wireless optical communication for small mobile platforms: Experimental channel study in highly-turbid lake water,” IEEE Access 8, (2020).
28. J. Li, B. Yang, D. Ye, L. Wang, K. Fu, J. Piao, and Y. Wang, “A real-time, full-duplex system for underwater wireless optical communication: hardware structure and optical link model,” IEEE Access 8, (2020).
29. H. Zhou, M Zhang, X Wang, and X Ren, “Design and implementation of more than 50 m real-time underwater wireless optical communication system,” J. Lightwave Technol. 40(12), 3564–3668 (2022). [CrossRef]
30. X. Yang, Z. Tong, H. Zhang, Y. Zhang, Y. Dai, C. Zhang, X. Chen, and J. Xu, “7-M/130-Mbps LED-to-LED underwater wireless optical communication based on arrays of series-connected LEDs and a coaxial lens group,” J. Lightwave Technol. 40(17), 5901–5909 (2022). [CrossRef]
31. T. Schmickl, R. Thenius, C. Moslinger, J. Timmis, A. Tyrrell, M. Read, J. Hilder, J. Halloy, A. Campo, and C. Stefanini, “CoCoRo–The self-aware underwater swarm,” in Fifth IEEE Conference on Self-Adaptive and Self-Organizing Systems Workshops120 (2011).
32. S. Mintchev, E. Donati, S. Marrazza, and C. Stefanini, “Mechatronic design of a miniature underwater robot for swarm operations,” in IEEE International Conference on Robotics and Automation (ICRA)2938 (2014).
33. F. Berlinger, M. Gauci, and R. Nagpal, “Implicit coordination for 3D underwater collective behaviors in a fish-inspired robot swarm,” Sci. Robot. 6(50), eabd8668 (2021). [CrossRef]
34. S. Han, Y. Noh, R. Liang, R. Chen, Y. J. Cheng, and M. Gerla, “Evaluation of underwater optical-acoustic hybrid network,” China Commun. 11(5), 49–59 (2014). [CrossRef]
35. J. He, J. Li, X. Zhu, S. Xiong, and F. Chen, “Design and analysis of an optical–acoustic cooperative communication system for an underwater remote-operated vehicle,” Appl. Sci. 12(11), 5533 (2022). [CrossRef]
36. W. Liu, Z. Xu, and L. Yang, “SIMO detection schemes for underwater optical wireless communication under turbulence,” Photonics Res. 3(3), 48 (2015). [CrossRef]
