Xiang-Rui Huang, Liang-Bi Chen
©SHUTTERSTOCK.COM/ROMOLO TAVANI
In recent years, underwater robots have been an essential focus of marine science and technology applications. Whether it is the application of military tasks or general civil affairs, underwater robots have played a significant role. For example, unmanned underwater vehicles (UUVs) can work in the sea for a long time, have high maneuverability, and perform various underwater tasks. There are many subcategories of UUVs, such as autonomous underwater vehicles (AUVs), remotely operated vehicles (ROVs), and autonomous and remotely operated vehicles (ARVs) (Kong et al., 2020).
In the military, underwater robots can be used for coastal positioning, coastal cruises (Yao et al., 2020), port surveillance of ships, and tracking vessels. In civilian applications, they can be used for underwater building inspection, underwater equipment inspection, underwater disaster rescue, and marine engineering. Especially in rescue operations, underwater rescue robots improve the efficiency of search and rescue, reduce the exposure of rescuers to dangerous underwater environments, and increase rescue safety. Oceanographic applications can be used for 3D mapping of the seabed environment (Salavasidis et al., 2021), collecting marine resources (Hu et al., 2020), and exploring marine resources. With the rise of intelligent aquaculture in recent years, underwater robots can also be used for fish activities, force monitoring, and breeding net cage inspection.
Before the development of underwater robots, divers relied on performing underwater tasks, such as inspecting subsea pipelines, cable breakage repairs, and oil rig inspections, and they took high risks to dive into the deep sea to perform repairs, reviews, and even salvage tasks. However, with the advancement of science and technology, these high-risk tasks have gradually been transferred to underwater robots, including oil exploration; subsea mineral exploration; subsea search and salvage operations; or the inspection of subsea pipelines, subsea structures, subsea cables, and aquatic drilling facilities (such as oil platforms and harbor terminals), and military operations have significantly reduced the risks faced by personnel when performing these tasks.
The most widely used underwater robot is usually the ROV, which is small, easy to control, and has muscular flexibility. The low-cost and small ROV faces many uncertainties when operating underwater, such as the existence of currents and waves in the ocean. Interference—these external disturbances—is a significant challenge for ROV designers. Therefore, ROVs emphasize the design of structure and power. Once the underwater robot loses control due to unknown environmental changes, it easily collides. The uncertain factor of this ecological change is water. It is difficult for robots to perform tasks accurately. Currently, much of the literature on ROVs is concerned with solving the loss of control due to unknown environmental changes, such as adaptive control, robust control, sliding mode control, fuzzy logic control, and other methods to improve the proportional-integral-derivative controller of the underwater robot. Overall, today’s ROVs are low cost, small, and flexible, solving the uncontrollable situation of unknown environmental changes. Therefore, this article focuses on the underwater application of ROVs. Table 1 also provides a specification example of an existing popular-market commercial ROV (CHASING M2).
Digital ocean collection is a tool that provides people with knowledge about the sea to meet the needs of developing and sustainable oceans in the 21st century. Digital ocean collection records the ocean in images and text, covering a range of animals, plants, and marine ecology in the sea; as we know, the ocean covers nearly 71% of Earth’s surface. The sea contains many treasures and secrets that humans have not yet discovered. Approximately 4.2 billion years ago, the world began to cool, and the atmospheric temperature dropped to form the ocean. At that time, the sea had already started to breed many organisms, and this continues today. To date, 210,000 kinds of marine organisms have been discovered, but the actual number of marine organisms may be tens of thousands or hundreds of thousands of times greater (Census Of Marine Life, 2003). More than 300 scientists from more than 53 countries work together; the marine census project for identifying and compiling marine organisms has recorded the ocean from paper to digital.
More than 15,000 species of fish have been listed in this marine organism census database. However, why make survey records on these hundreds of thousands of marine creatures? Because of the investigation of these marine creatures and ecology, we can understand and even predict the climate. For example, solar energy will affect the temperature of the seawater, and the seawater will then become a temperature regulator; from this marine census project, it is possible to identify endangered marine life and develop new chemical drugs as new marine species are discovered. However, as mentioned, this marine digital collection comes from marine experts and scholars.
Those with insufficient nautical knowledge who want to record marine life or discover unknown marine species still need to rely on professional oceanographers for identification. Exploration is focused on marine experts. In addition, there is no professional submarine equipment for conducting submarine exploration. Therefore, this article introduces how the general person can safely and conveniently avoid costly subsea equipment.
Because of the dark, cold, and unstable water pressure of the ocean’s interior, most of the deep sea is beyond human exploration. The traditional way to explore the ocean is to have individuals wearing diving and photography equipment dive into the water for photography. Even in the ancient Greek era, Aristotle’s zoology recorded 170 species of underwater animals in the Aegean Sea. Wagernel (human) diving involves risks, such as sunburn, dehydration, and marine life attacks; if diving equipment, such as underwater gas cylinders, is damaged, divers could lack oxygen. Long-term activities in violent currents can also easily cause divers to die. There are risks of diving diseases, such as ear barotrauma, decompression sickness, and pulmonary barotrauma. Divers need specific training to avoid the risk of diving diseases. Diving methods are divided into scuba diving and free diving. Scuba diving refers to using underwater breathing equipment for diving, and free diving relies on the diver’s breathing to dive.
Therefore, diving involves a high degree of skill and a certain degree of risk and requires complete diving course training before diving activities. Diving requires specific equipment, such as masks, snorkels, fins, wetsuits, buoyancy control devices, breathing apparatuses (regulators), and cylinders, and a complete set of diving equipment plus underwater photography equipment will cost US$1,700. Therefore, the traditional labor method of exploring the ocean is costly, and preparation and training before diving are time-consuming.
With the assistance of high technology, underwater robots that can withstand environmental instability can help humans explore the ocean in and out of the high-pressure and dark marine environment. Underwater robots can be divided into two types: ROVs and AUVs. ROVs are connected to large mother ships on the sea using a rope with copper wires or optical fiber bundles. The ship provides the power required for underwater robot movement, transmits control commands, and sends back information collected under the sea. AUVs are equipped with acoustic sensors, rely on batteries, and navigate the ocean autonomously. Today, ROVs have gradually replaced divers and become the mainstream of underwater operations. ROVs can also be equipped with robotic arms to perform the tasks of sample collection and seabed salvage (Cai et al., 2020).
Compared with conventional robots running on the ground, the structural design of underwater robots is challenging (Fernandez et al., 2019), and there is still a lack of development tools for underwater robots that can truly simulate the underwater environment because underwater robots need to face the deep sea. In harsh environments, the parameters in the seabed vary with time, the sea area, and the depth of the sea. There are also unknown changes in the seabed, such as currents and sea creatures, which will affect the underwater robot to perform seabed tasks. Therefore, to operate in the ocean, the underwater robot must have a solid and streamlined structural design, and the hardware structure design of the underwater robot must be able to adapt to the high-pressure environment of the sea. Advanced technologies include measurement and control technology, signal processing, dynamic estimation, navigation and positioning (Karmozdi et al., 2020; Xu et al., 2021), colocation (Li et al., 2020), and communications.
Underwater robots use acoustic responders to confirm their location. During the process, the underwater robot transmits an acoustic signal to the acoustic transponder, and each transponder responds to the underwater robot with its unique movement. From the information reflected by the three transponders, the underwater robot can use trigonometric functions to navigate and locate. Through sonar scanning (Kim et al., 2020), the underwater robot sends sound waves of a fixed frequency to the four directions of the seabed below and then listens to the difference in the sound frequency when they bounce back to calculate its speed.
Generally, the underwater robot shuttles between the sea surface and the seabed by the action of gravity and buoyancy. When diving from the sea surface, it uses gravity to go down, and when it reaches the vicinity of the target point, it throws out some of the weighted lead blocks. Balance the buoyancy, turn on the propellers’ thrust, and start diving. The underwater robot can use the ultrasonic sensor to detect the distance between itself and the obstacle. Then, the camera mounted on the bottom of the robot body is activated, and the captured underwater image data are sent back to the host. When the power is finally exhausted, the robot will throw away the last weight of the lead and float gently, returning to the sea where it started.
The problem of the underwater robot power supply has always been one of the challenges of underwater robot technology. The underwater robot battery must adapt to different sea conditions and perform observation tasks over a large sea area to provide a stable power supply for underwater robots. However, the battery of a single underwater robot cannot adapt to different sea conditions. As a result, it cannot load the power provided by the observation mission of the vast sea area.
Therefore, the underwater robot system is a concept that integrates a variety of underwater robots to form an ocean detection team (Connor et al., 2021). For example, multiple low-cost, low-power, compact, and mass-produced underwater robots conduct operations at fixed points in vast sea areas. For detection, the detection data of these various units are integrated into a large-scale observation result. The underwater robot system uses mutual positioning, connection, communication, and coordination between adjacent underwater robots to achieve fast and effective detection tasks (Lin et al., 2020); this type of underwater robot does not need to load expensive positioning equipment and uses the concept of multimachine division of labor to complete marine tasks, which is also the development trend of future underwater robots. This concept of multimachine division of labor can increase the efficiency of observation tasks in a large sea area and solve the problem of underwater robot power supply.
However, there are still challenges in the communication technology of underwater robots, and they need to be able to communicate with each other underwater. In robotics communication, this is usually done via Wi-Fi, Bluetooth, or infrared (IR). However, since Wi-Fi, Bluetooth, or IR transmits in the underwater environment, energy is absorbed, and the high volume of changes in the underwater environment can cause interference. In addition, sound waves are another possible method, but refraction is caused by changes in the seabed environment, which affects communication.
This section compares two methods of exploring the ocean. The traditional labor method of exploring the sea is expensive, and preparation and training before diving are time-consuming. An ROV can help humans enter and exit the sea in a high-pressure, dark marine environment. Humans no longer need to take high risks to perform ocean tasks; they can explore the ocean in a low-cost way and do not need time-consuming diving training. Therefore, the ROV method is superior to the traditional labor method of exploring the ocean.
Table 2 shows a comparison of ways to explore the sea. In terms of communication, the ROV is equipped with a communication cable and can send information back from the ocean any time. Compared with labor exploration of the sea, the labor investigation underwater is closed. Underwater surveys in the form of labor are closed because divers must rely on diving equipment and cannot freely explore the ocean. In comparison, underwater robots are highly maneuverable and can instantly transmit images to the robot’s operator. During the process, divers must wear underwater camera equipment to shoot. After shooting, the images and image information will be sorted out. In terms of power, underwater robots can continuously explore the marine environment if batteries are replaced. Generally, underwater robots must rely on batteries for energy. If the battery dies, the underwater robot must return to the robot’s operator to replace the battery. In terms of mobility, underwater robots are small and highly maneuverable. In comparison, in labor ocean exploration, divers wear large amounts of diving equipment and have poor mobility. In terms of movement, underwater robots rely on powered propellers to move forward. Conversely, in labor ocean exploration, divers must wear fins and use human sliding currents to move forward. Overall, the safety of underwater robots is higher than that of labor ocean exploration.
Table 2. A comparison of two ways of exploring the ocean.
As shown in Fig. 1, we can use underwater robots to explore the ocean and carry out digital collection in the sea. We can use underwater robots to explore various creatures and ecologies on the seabed, and we can explore the ocean without special equipment for diving. It is an emerging technology, and the cross-disciplinary combination of marine science enables those who are passionate about the ocean to use underwater robots to study marine life at any time; the marine life photographed in the ocean can be digitally archived, bringing nautical knowledge closer to popular science rather than the field of marine experts.
Fig 1 An underwater robot combined with marine digital collection.
The 2015 United Nations Sustainable Development Summit put forward 17 Sustainable Development Goals, of which the 14th goal was “Conservation and sustainable use of oceans and marine resources.” The sustainable development of oceans has become a global issue. The purpose is not to invade the ocean but to understand marine ecology and protect ocean marine life, use underwater robots to explore the ocean, and establish maritime digital classics for marine life, which can improve the current situation of marine conservation and sustainably manage the world. The ocean is one of the essential sources of food, and it provides food for the growing population on Earth. Marine digital collection aims to protect the diversity of marine life and promote the concept of sustainable ocean development.
In recent years, information and communication technology has been continuously improved, and the combination of the Internet of Things (IoT) and artificial intelligence (AI) has become the leading technology trend. The AI of Things (AIoT) means the application of AI combined with the IoT. The IoT is a network formed among hardware devices through sensors, remote controls, and others. Compared with traditional equipment, the AIoT has the function of interconnecting and exchanging data among devices. However, in addition to the connections among devices, AI can enable devices to have the ability to analyze data, upgrade the IoT to the AIoT, and achieve more immediate and intelligent demand responses. It can independently identify creatures in the ocean, record the growth of unique marine animals, and send back the identification images to the maritime digital archive to make the overall system intelligent. In addition, sensors can be installed in the underwater robot to detect the marine environment, that is, to establish a combination underwater robot system with IoT and AI technologies.
This article introduces a method different from the traditional manual method: using an underwater robot ROV method to explore the ocean and carry out digital ocean collection. We can explore various creatures and ecology on the seabed with underwater robots. However, there are still challenges in the design of the overall system—for example, the storage problem of the data returned by the underwater robot ROV and the battery problem of the underwater robot ROV. In terms of future applications, as the design and manufacturing technology of underwater robots at this stage are becoming increasingly mature, we can rely on underwater robots to join the task of deep sea exploration to find unknown marine creatures and ecological environments. In future developments, we hope to use emerging underwater robot technology to get close to the ocean, treat the sea kindly, and achieve the goal of a sustainable ocean.
This work is supported in part by the Higher Education Sprout Project, Ministry of Education, Taiwan, under Grant MOE 111G0009-1.
• F. Kong, Y. Guo, and W. Lyu, “Dynamics modeling and motion control of a new unmanned underwater vehicle,” IEEE Access, vol. 8, pp. 30,119–30,126, Feb. 2020, doi: 10.1109/ACCESS.2020.2972336.
• P. Yao, Z. Zhao, and Q. Zhu, “Path planning for autonomous underwater vehicles with simultaneous arrival in ocean environment,” IEEE Syst. J., vol. 14, no. 3, pp. 3185–3193, Sep. 2020, doi: 10.1109/JSYST.2019.2937346.
• G. Salavasidis et al., “Terrain-aided navigation with coarse maps—Toward an arctic crossing with an AUV,” IEEE J. Ocean. Eng., vol. 46, no. 4, pp. 1192–1212, Oct. 2021, doi: 10.1109/JOE.2021.3085941.
• C. Hu, Y. Pu, F. Yang, R. Zhao, A. Alrawais, and T. Xiang, “Secure and efficient data collection and storage of IoT in smart ocean,” IEEE Internet Things J., vol. 7, no. 10, pp. 9980–9994, Oct. 2020, doi: 10.1109/JIOT.2020.2988733.
• “How many fish in the sea? Census of marine life launches first report,” ScienceDaily. [Online] . Available: https://www.sciencedaily.com/releases/2003/10/031024064333.htm
• M. Cai, Y. Wang, S. Wang, R. Wang, Y. Ren, and M. Tan, “Grasping marine products with hybrid-driven underwater vehicle-manipulator system,” IEEE Trans. Autom. Sci. Eng. (From July 2004), vol. 17, no. 3, pp. 1443–1454, Jul. 2020, doi: 10.1109/TASE.2019.2957782.
• A. Karmozdi, M. Hashemi, H. Salarieh, and A. Alasty, “INS-DVL navigation improvement using rotational motion dynamic model of AUV,” IEEE Sensors J., vol. 20, no. 23, pp. 14,329–14,336, Dec. 2020, doi: 10.1109/JSEN.2020.3007929.
• Z. Xu, M. Haroutunian, A. J. Murphy, J. Neasham, and R. Norman, “An integrated visual odometry system for underwater vehicles,” IEEE J. Ocean. Eng., vol. 46, no. 3, pp. 848–863, Jul. 2021, doi: 10.1109/JOE.2020.3036710.
• S. Li, B. Xu, L. Wang, and A. A. Razzaqi, “Improved maximum correntropy cubature Kalman filter for cooperative localization,” IEEE Sensors J., vol. 20, no. 22, pp. 13,585–13,595, Nov. 2020, doi: 10.1109/JSEN.2020.3006026.
• B. Kim et al., “AUV-based multi-view scanning method for 3-D reconstruction of underwater object using forward scan sonar,” IEEE Sensors J., vol. 20, no. 3, pp. 1592–1606, Feb. 2020, doi: 10.1109/JSEN.2019.2946587.
• J. Connor, B. Champion, and M. A. Joordens, “Current algorithms, communication methods and designs for underwater swarm robotics: A review,” IEEE Sensors J., vol. 21, no. 1, pp. 153–169, Jan. 2021, doi: 10.1109/JSEN.2020.3013265.
• C. Lin, G. Han, J. Du, Y. Bi, L. Shu, and K. Fan, “A path planning scheme for AUV flock-based Internet-of-Underwater-Things systems to enable transparent and smart ocean,” IEEE Internet Things J., vol. 7, no. 10, pp. 9760–9772, Oct. 2020, doi: 10.1109/JIOT.2020.2988285.
• R. A. Suarez Fernandez, D. Grande, A. Martins, L. Bascetta, S. Dominguez, and C. Rossi, “Modeling and control of underwater mine explorer robot UX-1,” IEEE Access, vol. 7, pp. 39432–39447, 2019, doi: 10.1109/ACCESS.2019.2907193.
Xiang-Rui Huang (cren30475@gmail.com) is currently earning his B.S. degree in computer science and information engineering at National Penghu University of Science and Technology, Magong, Penghu 880011, Taiwan. His current research interests include deep learning, smart aquaculture, fish-feeding systems design, robots, Artificial Intelligence of Things systems design, and computer vision for the Internet of Things.
Liang-Bi Chen (liangbi.chen@gmail.com) earned his Ph.D. degree in electronic engineering from Southern Taiwan University of Science and Technology, Tainan, Taiwan. Currently, he is an assistant professor in the Department of Computer Science and Information Engineering at National Penghu University of Science and Technology, Penghu 880011, Taiwan. He serves as an associate editor for IEEE Access and as a chair of the Internet of Things Technical Committee of the IEEE Consumer Technology Society. He is a Senior Member of IEEE.
Digital Object Identifier 10.1109/MPOT.2022.3233713