Apparatus and Method for Autonomous and Swarm Navigation Control of Unmanned Ships

The digital twin system trains unmanned vessels to navigate autonomously and form swarms by simulating marine environments, addressing navigation and obstacle avoidance challenges, thereby enhancing their operational efficiency and safety.

KR103045997B1Active Publication Date: 2026-11-10LEEAHN
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
KR1020240123035
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2024-09-10
Publication Date
2026-11-10
Estimated Expiration
2044-09-10

AI Technical Summary

Technical Problem

Existing unmanned vessels lack effective methods for autonomous navigation and swarm driving, particularly in dynamic marine environments, and are limited in their ability to optimize navigation paths and avoid obstacles efficiently.

Method used

A digital twin-based system that simulates marine environments and vessel operations to train autonomous navigation algorithms, using reinforcement learning to derive optimal movement paths and obstacle avoidance strategies, enabling vessels to operate cooperatively in swarms.

Benefits of technology

Enhances the autonomous navigation capabilities of unmanned vessels by optimizing path planning, reducing collision risks, and improving energy efficiency through continuous learning and adaptation to environmental changes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112024099366968-PAT00001_ABST
    Figure 112024099366968-PAT00001_ABST
Patent Text Reader

Abstract

One embodiment relates to a technology for managing autonomous and swarm driving of unmanned vessels by applying a digital twin technique, comprising the steps of: collecting physical information of a vessel and marine environment monitoring information; receiving operation algorithms and operation status information of at least one unmanned vessel; creating a digital twin model to enable unmanned vessel operation simulation by reflecting the marine environment monitoring information collected in the collecting step and the unmanned vessel operation algorithms and operation status information received in the receiving step into a virtual model; and deriving an unmanned vessel movement path to derive an unmanned vessel learning algorithm according to the results of the unmanned vessel operation simulation in the digital twin model created in the creating step. By the method for managing autonomous and swarm driving of unmanned vessels, a customized 3D fetal avatar can be created by setting various similarity ratios between the mother and father.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The embodiment relates to a marine unmanned vessel monitoring technology, and more specifically, to a technology for managing the autonomous and swarm navigation of unmanned vessels by applying a digital twin technique. Background Technology

[0003] Unmanned vessels, such as unmanned surface vessels, receive a mission plan including given waypoints prior to operation, and perform the mission after autonomously navigating to a target point without operator intervention.

[0004] These unmanned vessels are being utilized in civilian sectors, such as monitoring illegal fishing, supporting rescue and salvage operations, and conducting marine surveys, through their path-following, obstacle avoidance, and specific object tracking capabilities. Furthermore, they are being used in the military sector for coastal surveillance, reconnaissance, mine detection, and combat missions, enabling them to be utilized for force enhancement without the loss of combat personnel.

[0005] In particular, various marine research institutions, fisheries companies, and technology firms are conducting diverse experiments and projects related to fish school detection using unmanned vessels.

[0006] In some advanced fisheries nations, unmanned fish school detection vessels are being used to rapidly locate fish schools and establish optimal fishing routes, thereby increasing fishing efficiency and reducing costs; consequently, the introduction of unmanned vessels is on the rise to enhance the automation and efficiency of the fisheries industry.

[0007] In addition, unmanned vessels can be utilized to monitor illegal fishing and collect important data for the protection and management of the marine environment. For example, the National Oceanic and Atmospheric Administration (NOAA) uses unmanned vessels to monitor the migration patterns of endangered species and collect data to protect marine ecosystems. Prior art literature

[0009] (Patent Document 0001) KR 10-1997-0026810 A(Patent Document 0002) KR 10-2015-0112166 A The problem to be solved

[0010] The present invention is derived from this technical background and aims to provide an autonomous and swarm driving management device and method for unmanned vessels capable of training and optimizing an autonomous driving algorithm for unmanned vessels so that they can navigate autonomously in the sea, avoid obstacles, and, when necessary, form a swarm to operate cooperatively. means of solving the problem

[0012] The present invention for achieving the above objectives includes the following configuration.

[0013] That is, a method for managing the autonomous and swarm driving of an unmanned vessel according to one embodiment of the present invention is a method performed on a computing device having one or more processors and a memory storing one or more programs executed by said one or more processors, and comprises the steps of: collecting physical information of a vessel and marine environment monitoring information; receiving operation algorithms and operation status information of at least one unmanned vessel; creating a digital twin model that enables unmanned vessel operation simulation by reflecting the marine environment monitoring information collected in the collecting step and the unmanned vessel operation algorithms and operation status information received in the receiving step into a virtual model; and deriving an unmanned vessel movement path to derive an unmanned vessel learning algorithm according to the results of the unmanned vessel operation simulation in the digital twin model created in the creating step.

[0014] Meanwhile, an autonomous and swarm driving management device for an unmanned vessel according to one embodiment is a computer device having one or more processors and a memory for storing one or more programs executed by said one or more processors, and includes an information collection unit for collecting marine environment monitoring information and physical information of a vessel, an operation status information receiving unit for receiving operation algorithms and operation status information of at least one unmanned vessel, a digital twin model creation unit for creating a digital twin model that enables unmanned vessel operation simulation by reflecting the marine environment monitoring information and physical information of a vessel collected by the information collection unit and the unmanned vessel operation algorithms and operation status information received by the operation status information receiving unit into a virtual model, and a movement path derivation unit for deriving an unmanned vessel movement path to derive an unmanned vessel learning algorithm in the digital twin model created by the digital twin model creation unit. Effects of the invention

[0016] According to one embodiment of the present invention, there is an effect of providing an autonomous and swarm driving management device and method for unmanned vessels that can generate a customized 3D fetal avatar using fetal ultrasound images by setting various similarity ratios between the mother and father.

[0017] In addition, the effect of providing an autonomous and swarm driving management device and method for unmanned vessels is derived, which can provide services that help with prenatal care by applying various services such as a prenatal diary using a generated customized 3D fetal avatar to interact more realistically with the fetus. Brief explanation of the drawing

[0019] FIG. 1 is a block diagram illustrating an autonomous and swarm driving management device for an unmanned vessel according to one embodiment of the present invention. FIG. 2 is an example of a Unity project creation screen according to one embodiment. Figures 3a and 3b are example diagrams illustrating key assets for creating a digital twin model. FIG. 4 is an example diagram of a marine environment simulation configuration screen according to one embodiment. FIGS. 5A and 5B are exemplary diagrams illustrating the modeling process of an unmanned vessel in a twin model production unit in one embodiment. Figure 6 is an example diagram illustrating the construction of an autonomous pest control learning environment. Figure 7 is an example diagram illustrating the construction of a destination navigation learning environment. Figure 8 is an example of a grid sensor used by an unmanned vessel. FIG. 9 is an example diagram illustrating the granting of rewards according to the deviation angle according to one embodiment. FIG. 10 is an example diagram of a destination navigation learning environment configuration according to one embodiment of the present invention. FIG. 11 is an example diagram showing the TensorBoard learning results according to one embodiment of the present invention. Figure 12 is an example diagram illustrating a cluster autonomous fish school detection scenario. FIG. 13 is a flowchart of a method for managing autonomous and swarm driving of an unmanned vessel according to one embodiment of the present invention. Specific details for implementing the invention

[0020] It should be noted that the technical terms used in this invention are used merely to describe specific embodiments and are not intended to limit the invention. Furthermore, unless specifically defined otherwise in this invention, the technical terms used in this invention should be interpreted in the sense generally understood by those skilled in the art to which this invention pertains, and should not be interpreted in an overly broad or overly narrow sense.

[0022] Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the attached drawings.

[0023] The autonomous and swarm driving management device for an unmanned vessel according to embodiments of the present invention may be implemented by at least one computer device, and the autonomous and swarm driving management method for an unmanned vessel according to embodiments of the present invention may be performed through at least one computer device included in the autonomous and swarm driving management device for an unmanned vessel. At this time, a computer program according to one embodiment of the present invention may be installed and run on the computer device, and the computer device may perform the autonomous and swarm driving management method for an unmanned vessel according to embodiments of the present invention under the control of the run computer program. The above-described computer program may be stored on a computer-readable recording medium to be combined with the computer device to execute the autonomous and swarm driving management method for an unmanned vessel on a computer.

[0025] An autonomous and swarm driving management device for an unmanned vessel according to one embodiment proposes a function to autonomously navigate in the sea, avoid obstacles, and, when necessary, form a swarm to operate cooperatively. To this end, autonomous driving technology through reinforcement learning and a digital twin can be configured to enable effective operation in various marine environments.

[0026] That is, the autonomous and swarm driving management device for an unmanned vessel according to one embodiment can utilize a digital twin model to derive an optimal movement path based on the results of the unmanned vessel's operation simulation, and reflect this in a learning algorithm to improve autonomous navigation performance. Through this, the unmanned vessel can be trained to operate more efficiently and safely in various maritime environments.

[0027] An autonomous and swarm driving management device for an unmanned vessel according to one embodiment performs a simulation of the operation of the unmanned vessel under various conditions in a digital twin environment. For example, it can simulate scenarios such as navigating along a specific path, collision avoidance, obstacle detection, and reaching a target point.

[0028] In addition, data obtained from simulations is analyzed to evaluate the operational performance of the ship, including route optimization, energy efficiency, collision avoidance capability, and responsiveness to environmental changes.

[0029] Furthermore, by using data derived from simulation results to develop or improve learning algorithms for unmanned vessels, such as reinforcement learning algorithms, unmanned vessels can acquire autonomous navigation capabilities that allow them to independently learn optimal movement paths and adapt to new environmental conditions.

[0030] In this process, the learning algorithm derives the optimal travel path based on simulation data and various pathfinding results. For example, it can calculate a route that minimizes fuel consumption, reduces the risk of collision, and reaches the destination most quickly. The travel path is continuously updated according to the vessel's motion model and environmental conditions, thereby optimizing the autonomous navigation capabilities of the unmanned vessel.

[0031] Furthermore, the derived learning algorithm can be verified in a real-world environment. To this end, the performance of the algorithm in a virtual environment can be applied to actual unmanned vessel operations to compare results, and the algorithm's performance can be evaluated through actual testing to perform additional adjustments and improvements if necessary.

[0032] FIG. 1 is a block diagram illustrating an autonomous and swarm driving management device for an unmanned vessel according to one embodiment of the present invention.

[0033] As illustrated in FIG. 1, the autonomous and swarm driving management device (10) of an unmanned vessel according to one embodiment specifically includes a communication interface (110), a memory (120), an input / output interface (130), and a processor (140).

[0034] The communication interface (110) can provide a function for the autonomous and swarm driving management device (10) of the unmanned vessel to communicate with other devices. For example, the processor (140) of the autonomous and swarm driving management device (10) of the unmanned vessel can transmit requests, commands, data, files, etc., generated according to program code stored in a recording device such as memory (120) to other devices, such as at least one user terminal (30), through a network (20) under the control of the communication interface (110).

[0035] Conversely, signals, commands, data, files, etc. from other devices may be received via a network through the communication interface (110) of the autonomous and swarm driving management device (10) of the unmanned vessel. Signals, commands, data, etc. received through the communication interface (110) may be transmitted to a processor (140) or memory (120), and files, etc. may be stored in a storage medium (the permanent storage device described above) that the autonomous and swarm driving management device (10) of the unmanned vessel may further include.

[0036] A network (20) refers to a connection structure capable of exchanging information between each node, such as multiple terminals and servers, and examples of such a network include a Local Area Network (LAN), a Wide Area Network (WAN), the Internet (WWW: World Wide Web), a wired and wireless data network, a telephone network, a wired and wireless television network, etc. Examples of wireless data communication networks include, but are not limited to, 3G, 4G, 5G, 3GPP (3rd Generation Partnership Project), 5GPP (5th Generation Partnership Project), LTE (Long Term Evolution), WIMAX (World Interoperability for Microwave Access), Wi-Fi, Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network), RF (Radio Frequency), Bluetooth network, NFC (Near-Field Communication) network, satellite broadcasting network, analog broadcasting network, DMB (Digital Multimedia Broadcasting) network, etc.

[0037] In the following, the term "at least one" is defined as a term including both singular and plural forms, and it will be obvious that even if the term "at least one" does not exist, each component may exist in a singular or plural form and may mean singular or plural. Furthermore, whether each component is provided in a singular or plural form may be changed according to the embodiment.

[0038] At least one user terminal (30) may be implemented as a computer capable of connecting to a remote server or terminal via a network. Here, the computer may include, for example, a laptop, desktop, or laptop equipped with a navigation system or a web browser. At this time, at least one user terminal (30) may be implemented as a terminal capable of connecting to a remote server or terminal via a network.

[0039] At least one user terminal (30) may include all kinds of handheld-based wireless communication devices, such as navigation, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet) terminal, smartphone, smartpad, tablet PC, etc.

[0040] At least one user terminal (30) may be a terminal possessed by a ship operation manager, etc., who uses a community or platform by using a web page, app page, program, or application related to the autonomous and swarm driving management service of an unmanned vessel according to one embodiment.

[0041] Memory (120) is a computer-readable recording medium and may include RAM (random access memory), ROM (read only memory), and permanent mass storage devices such as disk drives and servers. Here, permanent mass storage devices such as ROM and disk drives may be included in the autonomous and swarm driving management device (10) of an unmanned vessel as separate permanent storage devices distinct from memory (120).

[0042] Additionally, an operating system and at least one program code may be stored in the memory (120). These software components may be loaded into the memory (120) from a computer-readable recording medium separate from the memory (120). This separate computer-readable recording medium may include a floppy drive, disk, tape, DVD / CD-ROM drive, memory card, SSD, USB, etc.

[0043] In another embodiment, software components may be loaded into memory (120) via a communication interface (110) rather than a computer-readable recording medium. For example, software components may be loaded into memory (120) of the autonomous and swarm driving management device (10) of an unmanned vessel based on a computer program installed by files received over a network.

[0044] The input / output interface (130) may be a means for interfacing with an input / output device. For example, the input device of the input / output device may include a device such as a microphone, keyboard, or mouse, and the output device may include a device such as a display or speaker. As another example, the input / output interface (130) may be a means for interfacing with a device in which the functions for input and output are integrated into one, such as a touchscreen. The input / output device may be composed of a single device with the autonomous and swarm driving management device (10) of the unmanned vessel.

[0045] The processor (140) may be configured to process instructions of a computer program by performing basic arithmetic, logic, and input / output operations. Instructions may be provided to the processor (140) via memory (120) or a communication interface (110). For example, the processor (140) may be configured to execute instructions received according to program code stored in a recording device such as memory (120).

[0046] As illustrated in FIG. 1, the processor (140) of the autonomous and swarm driving management device (10) of an unmanned vessel according to one embodiment includes an information collection unit (1410), a driving status information receiving unit (1420), a digital twin model creation unit (1430), a movement path derivation unit (1440), and a reward granting unit (1450).

[0047] The information collection unit (1410) collects marine environment monitoring information and physical information of the vessel.

[0048] Marine environment monitoring information includes physical environment information, such as water temperature, salinity, wave height and wave period, ocean current data, wind data, and water depth data, as well as weather conditions such as rainfall, fog, typhoons, and storms.

[0049] It also includes information on light intensity and visibility, such as sunlight, length of day and night, and visibility distance; information on the marine ecosystem of a specific sea area; information related to maritime traffic, such as the location, speed, and size of other vessels; and information on the location and characteristics of floating objects, icebergs, and offshore structures.

[0050] And the physical information of the vessel includes vessel dimensions and shape, hull structure information, material properties, weight distribution of the vessel, and location information of the center of gravity (CG) and center of buoyancy (CB).

[0051] In addition, mechanical characteristics of the ship, including propulsion system information, steering system information, and power system information, can also be collected.

[0052] And the operation status information receiving unit (1420) receives operation algorithms and operation status information of at least one unmanned vessel.

[0053] The operation algorithm of an unmanned vessel includes various control and decision-making logic that enables autonomous navigation. Specifically, the operation algorithm of an unmanned vessel may include a path planning algorithm that calculates the optimal path from a starting point to a target point and modifies the path in real time according to environmental changes (e.g., occurrence of obstacles, weather changes); a collision avoidance algorithm that detects obstacles in real time using sensors such as LiDAR, radar, cameras, and sonar, calculates avoidance paths, analyzes maritime traffic conditions to predict the position, speed, and direction of surrounding vessels, and determines whether there is a risk of collision; an autonomous navigation algorithm that adjusts the rudder and thrusters to maintain the vessel's designated path, monitors the marine environment (weather, currents, tides, etc.) in real time, and adjusts the navigation strategy accordingly; an energy management algorithm; and communication and data processing algorithms.

[0054] In addition, operational status information of unmanned vessels is real-time data collected through various sensors and data sources, and is used to accurately reflect the actual operational status of the unmanned vessels in the digital twin model.

[0055] Operation status information may include location and navigation information, environmental sensing information such as radar, LiDAR, and sonar data, operation status information, system status and health information, collision and obstacle situation information, route and mission information, etc.

[0056] The digital twin model production unit (1430) creates a digital twin model that enables unmanned vessel operation simulation by reflecting the marine environment monitoring information and physical information of the vessel collected from the information collection unit (1410), and the unmanned vessel operation algorithm and operation status information received from the operation status information receiving unit (1420) into a virtual model.

[0057] And the movement path derivation unit (1440) derives a movement path for unmanned vessels to derive an unmanned vessel learning algorithm based on the operation simulation results of unmanned vessels in the digital twin model produced by the digital twin model production unit (1430).

[0058] A digital twin refers to a simulation environment that allows for testing and learning the performance of unmanned vessels by virtually recreating conditions and situations identical to the actual marine environment.

[0059] FIG. 2 is an example of a Unity project creation screen according to one embodiment.

[0060] The digital twin model creation department (1430) creates a Unity project as shown in Fig. 2 and implements a sea environment in Unity.

[0061] In one embodiment, a Universal Render Pipeline (URP) project is created in Unity Hub as shown in FIG. 2. The reason for selecting URP is that HDRP does not support Android and iOS, and it was chosen to allow for testing and operation on mobile platforms. By creating a new project, preparations can be made to configure the ocean environment.

[0062] In one aspect, the digital twin model creation department (1430) creates a digital twin using DWP2 (Dynamic Water Physics 2) and Crest Ocean System assets.

[0063] Figures 3a and 3b are example diagrams illustrating key assets for creating a digital twin model.

[0064] Figure 3a is an example illustration to explain the DWP (Dynamic Water Physics 2) asset. DWP is an asset that calculates buoyancy and fluid dynamics using mesh data, which allows for a more realistic representation of an object floating and moving on water.

[0065] Figure 3b is an example diagram illustrating the Crest Ocean System asset.

[0066] The Crest Ocean System is an asset in Unity designed to implement visual representations of the ocean and wave simulations. This asset is available in URP and HDRP, and we selected URP for mobile support. By integrating the wave mesh data generated by the Crest Ocean System with DWP, it is possible to simulate physical reactions such as the undulation of the real ocean.

[0067] FIG. 4 is an example diagram of a marine environment simulation configuration screen according to one embodiment.

[0068] The digital twin model creation department (1430) creates an empty game object and adds the OceanRenderer.cs script to configure the ocean environment, and uses the CrestWaterDataProvider.cs script to configure it to interact with the DWP, thereby simulating the dynamic characteristics of the ocean. Additionally, Crest Shape FFT.cs can be added to another empty object to pass parameters of the wave simulation to OceanRenderer.cs.

[0069] Additionally, the digital twin model production department (1430) performs modeling of the unmanned vessel for the actual application of the autonomous driving algorithm.

[0070] The digital twin model creation department (1430) can create a model of an actual unmanned vessel within the Unity environment and integrate it into a simulation to build a more realistic test environment. The unmanned vessel is designed to reflect its unique physical characteristics and operational performance, thereby enabling the optimization of autonomous driving capabilities through reinforcement learning.

[0071] At this time, the 3D modeling of the unmanned vessel is created based on the size and shape of the actual operating vessel. Through this, the results simulated in the digital twin environment can be reflected as closely as possible to the operating conditions of the real world.

[0072] The twin model production department can implement details such as the ship's hull, propulsion system, and sensors during the 3D modeling process, thereby laying the foundation for testing and improving operational performance in actual marine environments.

[0073] And the digital twin model creation department (1430) can create a model of an unmanned vessel in a Unity environment and then apply DWP2 assets to more realistically implement the movement of the unmanned vessel on the sea. Since DWP2 calculates buoyancy and fluid dynamics using mesh data, it can reproduce the appearance of the unmanned vessel floating and moving on the water in a manner similar to reality. Through this, it is possible to simulate the movement, rotation, and reaction to waves of the unmanned vessel, and provide an environment where an optimal autonomous driving algorithm can be trained through reinforcement learning.

[0074] FIGS. 5A and 5B are exemplary diagrams illustrating the unmanned vessel modeling process in the twin model production unit of one embodiment.

[0075] As shown in FIG. 5a, the propulsion system of the unmanned vessel consists of a total of two motors, one on each side. These motors are used to control the direction and speed of the unmanned vessel, and the output of each motor is adjusted according to an autonomous driving algorithm to enable the unmanned vessel to move accurately toward a target point. This configuration helps the unmanned vessel operate stably under various environmental conditions. FIG. 5b shows the propulsion system of an unmanned vessel with two motors applied.

[0076] In one aspect of the present invention, the movement path derivation unit (1440) learns an unmanned vessel learning algorithm so that the unmanned vessel detects and avoids obstacles and moves toward a target point in an autonomous control environment built from a digital twin model created by the digital twin model creation unit (1430).

[0077] Figure 6 is an example diagram illustrating the construction of an autonomous pest control learning environment.

[0078] In other words, an autonomous pollution control environment can be established to train the autonomous navigation algorithms of unmanned vessels using reinforcement learning. In this environment, the vessel learns by using grid sensors to detect and avoid obstacles, and by configuring it to move toward a target point. This learning environment can help train the unmanned vessel to effectively avoid obstacles and reach its target even in real-world marine conditions.

[0079] Figure 7 is an example diagram illustrating the construction of a destination navigation learning environment.

[0080] As shown in Fig. 7, the destination navigation learning environment provides a scenario in which an unmanned vessel moves toward a given destination while simultaneously avoiding marine debris or other obstacles. In this environment, the unmanned vessel can learn a path to reach the destination while avoiding obstacles in real time.

[0081] In a characteristic aspect of the present invention, the movement path derivation unit (1440) learns that the unmanned vessel detects the surrounding environment using a grid sensor and calculates a path to a destination by avoiding obstacles using the detection data of the grid sensor.

[0082] Figure 8 is an example of a grid sensor used by an unmanned vessel.

[0083] The unmanned vessel detects its surrounding environment using grid sensors. The grid sensors distinguish between other vessels (blue) and obstacles (red) in the vicinity, helping the vessel avoid them and find a safe route. Based on this sensor data, the unmanned vessel calculates a safe path and can move efficiently to a given destination.

[0084] At this time, in order for the unmanned vessel to navigate accurately to the destination, observation data is used that includes the Y-axis rotation value of the main vessel indicating the unmanned vessel's direction of travel, a direction vector between the vessel and the destination representing the direction the unmanned vessel is facing the destination, and the distance to the destination measured between the current unmanned vessel position and the destination.

[0085] In one embodiment, the dual path derivation unit (1440) can calculate a safe path using observation data including the Y-axis rotation value of the vessel, a direction vector between the vessel and the destination, and the distance to the destination. These observation data can enable the unmanned vessel to adjust its position and direction in real time and reach the destination via an optimal path.

[0086] The dual path derivation unit (1440) can provide the operating algorithm information and operating situation information of the unmanned vessel collected for the digital twin model, which are necessary data to accurately simulate and optimize the autonomous operation performance of the actual vessel in a virtual environment.

[0087] In a further aspect of the present invention, the reward granting unit (1450) may provide various forms of reward provision functions to effectively proceed with learning in the movement path derivation unit ().

[0088] In one embodiment, the reward granting unit (1450) grants a reward based on the angle of deviation or collision with an obstacle relative to the destination direction for the derived unmanned vessel movement path.

[0089] FIG. 9 is an example diagram illustrating the granting of rewards according to the deviation angle according to one embodiment.

[0090] As shown in FIG. 9, when the unmanned vessel moves toward a destination, the reward granting unit (1450) grants a reward based on the difference in angle between the direction of travel and the direction of the destination. By granting a higher reward as the difference in angle becomes smaller, the unmanned vessel is enabled to move toward the destination more accurately. Additionally, it can be implemented so that points are deducted in proportion to the angle of deviation from the straight line to the destination.

[0091] Additionally, the reward granting unit (1450) detects when the unmanned vessel collides with an obstacle, a wall, or another vessel using Unity's Collision event, immediately grants a penalty reward, and then terminates the episode. This mechanism induces the unmanned vessel to learn to avoid collisions and enables it to select and maintain a safe navigation path.

[0092] And the reward granting unit (1450) can reinforce correct behavior by granting a high reward when the unmanned vessel successfully reaches its destination.

[0093] FIG. 10 is an example diagram of a destination navigation learning environment configuration according to one embodiment of the present invention.

[0094] A digital twin model creation unit (1430) according to one embodiment can configure a destination navigation learning environment as shown in FIG. 10 and provide a screen.

[0095] In one embodiment, the movement path derivation unit (1440) repeated the learning several times in the designed learning environment to enhance the autonomous navigation capability of the unmanned vessel to the destination. The learning process can be monitored using TensorBoard.

[0096] FIG. 11 is an example diagram showing the TensorBoard learning results according to one embodiment of the present invention.

[0097] In a characteristic aspect of the present invention, the movement path derivation unit (1440) can derive an unmanned vessel movement path by learning an unmanned vessel learning algorithm according to a scenario in which at least one unmanned vessel forms a set formation to detect a school of fish.

[0098] Figure 12 is an example diagram illustrating a cluster autonomous fish school detection scenario.

[0099] In one embodiment, the swarm autonomous fish school detection learning is implemented based on an existing destination autonomous navigation module to additionally learn the function of unmanned vessels forming a swarm and cooperatively detecting fish schools. In this process, scenarios can be set in which unmanned vessels form various formations to reach a destination or detect fish schools.

[0100] As shown in FIG. 12, at least one unmanned vessel can be implemented to autonomously detect schools of fish by forming a cluster in one of the following formations: a straight formation, a crane-wing formation, or a circular formation.

[0101] Specifically, the linear formation is a formation in which unmanned vessels line up in a single file and move while scanning a wide area forward. All vessels move in the same direction and maintain a constant distance to avoid obstacles and advance efficiently.

[0102] The Crane Wing Formation is a formation in which unmanned vessels move cooperatively in a V-shape, making it useful for navigating narrow paths or intensively detecting specific targets. In this formation, the lead unmanned vessel finds the path, while the following vessels maintain the formation and advance from their respective positions.

[0103] A circular formation is a formation in which unmanned vessels form a circle to protect a center point or move while scanning around it. It can be effective for detecting areas with dense fish schools or protecting the center from external dangers.

[0104] In one embodiment, the movement path derivation unit (1440) can analyze and determine which formation of the unmanned vessels is advantageous for detecting schools of fish according to the situation.

[0105] For example, a linear formation is selected when monitoring a wide area, a crane-wing formation is selected when intensively detecting specific targets within a predetermined area, and a circular formation is selected when scanning the center; the system learns to move by forming a cluster in the selected formation and can derive a movement path.

[0106] More specifically, when a linear formation is selected, the unmanned vessels are trained to focus on maintaining a linear formation while detecting a wide range using their respective sensors.

[0107] The unmanned vessel uses grid sensors to detect its surroundings. These sensors distinguish between other vessels and obstacles nearby, helping the vessel find a safe route while avoiding them. Based on this sensor data, the unmanned vessel calculates a safe path and can move efficiently to a given destination.

[0108] In a linear formation, unmanned vessels measure the distance to other vessels located ahead, behind, left, and right in real time to determine the distance and relative position between vessels. Through this, the unmanned vessels can form a linear formation while maintaining a constant spacing. In particular, each unmanned vessel is implemented to continuously maintain the formation by calculating its relative position with surrounding vessels based on the position within the formation it must maintain.

[0109] In a linear formation, unmanned vessels detect the location and distance of surrounding obstacles in real time using grid sensors. When an obstacle approaches, the unmanned vessel plans an immediate avoidance action to prevent collision; ideally, this avoidance action is designed to ensure a safe distance from the obstacle while maintaining the formation.

[0110] In addition, in a linear formation, the unmanned vessels calculate the direction vector and distance to the destination, advance in a consistent direction while maintaining the linear formation, and thereby enable the unmanned vessels to move efficiently to the target point.

[0111] When moving in a linear formation, a higher reward is provided for maintaining the linear formation accurately, that is, for maintaining a constant distance between vessels. Additionally, if an unmanned vessel collides with another vessel or an obstacle, a penalty reward is given, and a high reward is paid for successfully reaching the destination, thereby reinforcing the behavior of efficiently reaching the destination while maintaining the formation.

[0112] Swarming in a crane-wing formation involves unmanned vessels arranged in a V-shape to move cooperatively, making it suitable for navigating narrow paths or intensively exploring specific areas.

[0113] When moving in a cluster in a crane-wing formation, the unmanned vessels use grid sensors to detect the surrounding environment. The grid sensors distinguish other vessels and obstacles (in red) nearby, helping to avoid them and find a safe route.

[0114] When moving in a cluster in the Crane Wing formation, the follow-up UAVs track the position and direction of the lead UAV in real time. Then, the follow-up UAVs maintain a V-shaped formation by adjusting their positions and angles based on the lead UAV's path. Through this information, the UAVs can flexibly adjust their formation and move.

[0115] In addition, when moving in a cluster in a crane-wing formation, the unmanned ships maintain the formation by measuring the distance to other unmanned ships in front of and behind them. In particular, to form a V-shaped formation, they maintain an even distance between ships and can adjust their relative positions according to the position of the leading unmanned ship.

[0116] It plans avoidance paths by observing the location and distance of obstacles in real time, and is implemented so that the unmanned vessels move while maintaining their formation as much as possible even when avoiding obstacles.

[0117] Additionally, rewards are granted when unmanned vessels accurately maintain a V-shape during cluster movement in the Crane Wing formation. In other words, higher rewards may be awarded as the follow-on unmanned vessels maintain a precise V-shape centered around the lead unmanned vessel. This enables the unmanned vessels to move cooperatively while maintaining their formation.

[0118] Furthermore, penalty rewards are assigned if unmanned vessels disrupt the formation or collide, thereby teaching them to maintain the formation while avoiding collisions. Additionally, high rewards are given for reaching the destination while maintaining the formation, which can reinforce autonomous driving learning to maintain the correct shape.

[0119] Accordingly, unmanned vessels can learn to improve their ability to cooperatively reach a destination while maintaining a V-shape to avoid obstacles and having subsequent unmanned vessels accurately adjust their positions along the path of the lead unmanned vessel.

[0120] In the circular formation, unmanned vessels form a circle and move in concentric circles around the center; each unmanned vessel learns to maintain a circular arrangement while keeping a distance from neighboring unmanned vessels.

[0121] Even when moving in a large, circular formation, the unmanned vessels use grid sensors to detect their surroundings. The grid sensors distinguish between other vessels and obstacles nearby, helping the vessels avoid them and find a safe route. Based on this sensor data, the unmanned vessels calculate a safe path to efficiently navigate to a given destination.

[0122] In addition, when moving in a circular formation, the unmanned vessels measure the distance to other unmanned vessels located to the left and right in real time to maintain a constant distance, and each unmanned vessel calculates its relative position with surrounding vessels according to the position of the formation it must maintain, thereby continuously maintaining the formation.

[0123] Furthermore, each unmanned vessel monitors its distance in real time from the center point of the circular formation (a virtual point representing the center). This distance value enables the unmanned vessel to move within a fixed distance from the center point while maintaining the radius of the circle. Additionally, by calculating the direction vector and distance to the destination, the unmanned vessels advance in a consistent direction while maintaining the circular formation. This allows for efficient movement to the target point.

[0124] When moving in a circular formation, unmanned vessels are given a higher reward if they maintain a consistent distance from neighboring unmanned vessels, as this allows them to maintain the circular formation more accurately. Additionally, rewards are given if unmanned vessels maintain a constant distance from the center point, and the reward may be reduced as they move further away or closer to the center point. If the direction of movement of an unmanned vessel aligns with the direction of movement along the circumference of the circle, additional rewards are given, and rewards are deducted as the difference in direction increases, and deducted rewards are given in the event of a collision with another vessel or obstacle.

[0125] In other words, through swarm movement in a circular formation, the unmanned vessels can adjust their positions relative to each other to stably maintain the formation, enabling them to perform missions such as intensively detecting specific targets or protecting the center.

[0126] FIG. 13 is a flowchart of a method for managing autonomous and swarm driving of an unmanned vessel according to one embodiment of the present invention.

[0127] A method for managing autonomous and swarm driving of an unmanned vessel according to one embodiment first collects physical information of the vessel and marine environment monitoring information (S200).

[0128] And, receive operation algorithm and operation status information of at least one unmanned vessel (S210).

[0129] Subsequently, a digital twin model is created by reflecting the marine environment monitoring information collected in the collection step and the unmanned vessel operation algorithm and operation status information received in the receiving step into a virtual model to enable unmanned vessel operation simulation (S220).

[0130] In one embodiment, the step of creating a digital twin model involves creating a digital twin using DWP2 (Dynamic Water Physics 2) and Crest Ocean System assets.

[0131] Then, in the production stage, a movement path for an unmanned vessel is derived based on the results of the operation simulation of the unmanned vessel from the digital twin model produced to derive an unmanned vessel learning algorithm (S230).

[0132] In one aspect of the present invention, the step of deriving an unmanned vessel movement path involves deriving an unmanned vessel movement path by learning an unmanned vessel learning algorithm according to a scenario in which at least one unmanned vessel forms a set formation to detect a school of fish.

[0133] And in the step of deriving the unmanned vessel's movement path, an unmanned vessel learning algorithm is trained so that the unmanned vessel detects and avoids obstacles and moves toward a target point in the autonomous pollution control environment built from the digital twin model produced in the manufacturing stage.

[0134] In addition, the step of deriving the movement path of an unmanned vessel involves the unmanned vessel detecting the surrounding environment using a grid sensor and learning to calculate a path to a destination by avoiding obstacles using the detection data from the grid sensor.

[0135] In one embodiment, the step of deriving the movement path of an unmanned vessel can derive the movement path of the unmanned vessel to the destination using observation data including the Y-axis rotation value of the vessel, a direction vector between the vessel and the destination, and the distance to the destination.

[0136] In a further aspect of the present invention, in the step of deriving an unmanned vessel movement path, a reward is given to the unmanned vessel movement path derived based on the direction of the destination, depending on the angle of deviation, whether there is a collision with an obstacle, or whether the destination is reached (S240).

[0138] The above-described method may be implemented as an application or in the form of program instructions that can be executed through various computer components and recorded on a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, etc., either individually or in combination.

[0139] The program instructions recorded on the above-mentioned computer-readable recording medium are those specifically designed and configured for the present invention, but may also be those known and available to those skilled in the art of computer software.

[0140] Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, and flash memory.

[0141] Examples of program instructions include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. The hardware device may be configured to operate as one or more software modules to perform processing according to the present invention, and vice versa.

[0143] Meanwhile, a stain-resistant coating layer made of a stain-resistant coating composition may be applied to the upper surface of the communication interface (110) to improve stain resistance.

[0144] The above anti-stain coating composition contains alkoxysilane and laurylamidopropyl betaine in a molar ratio of 1:0.01 to 1:2, and the total content of alkoxysilane and laurylamidopropyl betaine is 1 to 12 weight% with respect to the total aqueous solution.

[0145] The above alkoxysilane and laurylamidopropyl betaine are preferably in a molar ratio of 1:0.01 to 1:2. If the molar ratio falls outside this range, the applicability of the anti-fouling coating composition is reduced, or moisture adsorption on the surface increases after application, causing the coating film to be removed.

[0146] The above alkoxysilane and laurylamidopropyl betaine are preferably present in an amount of 1 to 12 weight percent of the total aqueous composition. If the amount is less than 1 weight percent, there is a problem that the coating performance of the anti-fouling coating composition is reduced, and if the amount exceeds 12 weight percent, crystal precipitation is likely to occur due to an increase in the thickness of the coating film.

[0147] Meanwhile, it is preferable to apply the present anti-fouling coating composition onto the communication interface (110) by a spray method. Additionally, the final coating film thickness on the communication interface (110) is preferably 700 to 2000 Å. If the thickness of the coating film is less than 700 Å, there is a problem of deterioration in the case of high-temperature heat treatment, and if it exceeds 2000 Å, there is a disadvantage that crystal precipitation is likely to occur on the coating surface.

[0148] In addition, the present stain-resistant coating composition can be prepared by adding 0.1 mol of alkoxysilane and 0.05 mol of laurylamidopropyl betaine to 1000 ml of distilled water and then stirring.

[0149] The reason the ratio of the above components and the thickness of the coating film were numerically limited as described above is that, based on the analysis of test results after repeated failures by the inventor, the optimal anti-fouling coating effect was observed at the above ratios.

[0150] In addition, a coating layer may be formed on the processor (140) to prevent corrosion and improve wear resistance of the metal surface.

[0151] The coating material of this coating layer is composed of 26 wt% polymer 188, 25 wt% salicylic acid, 10 wt% hafnium, 12 wt% magnesium organic acid, 9 wt% titanium oxide (TiO2), 12 wt% aluminum oxide (AIO2), and 6 wt% co-solvent, and the coating thickness can be formed to 6 μm.

[0152] Polysamer 188 and salicylic acid salts serve to prevent corrosion, weathering, and discoloration, while hafnium, as a transition metal element with wear resistance and weathering properties, plays a role in providing excellent water resistance and corrosion resistance.

[0153] Magnesium organic acid serves to impart alkali resistance and sliding properties to the surface of the coating film, while titanium oxide and aluminum oxide are added for purposes such as refractoriness and chemical stability.

[0154] The reason the ratio of the above-mentioned components and the coating thickness were numerically limited as described above is that, based on the analysis of test results after repeated failures by the inventor, the above ratios demonstrated optimal effects in improving weather resistance and wear resistance.

[0155] In addition, a sound-absorbing layer can be formed inside the case of the above-mentioned service providing device (10).

[0156] The above sound-absorbing layer may use a Co-PET / PET bicomponent fiber structure nonwoven fabric.

[0157] It is preferable that the thickness of the sound-absorbing layer be 0.7 to 15 mm. If the thickness of the sound-absorbing layer is less than 0.7 mm, a sufficient sound absorption effect is not obtained, and if it exceeds 15 mm, it is not desirable because it has the disadvantage of increasing the temperature of the service-providing device (10) as sufficient space is not secured.

[0158] The unit weight of the above sound-absorbing layer is 13 to 450 g / m² 2 It is preferable to do so. 13g / m² 2 Below this level, a sufficient sound absorption effect is not obtained, and also 450 g / m² 2 If it exceeds [amount], the lightness of the service providing device (10) cannot be secured, so it is not desirable.

[0159] It is preferable that the fineness of the fibers constituting the sound-absorbing layer be in the range of 0.5 to 25 decitex. It is not desirable if it is less than 0.5 decitex because it is difficult to absorb low-frequency noise and cushioning properties are reduced, and it is not desirable if it exceeds 25 decitex because it is difficult to absorb high-frequency noise.

[0160] The tensile strength of the nonwoven fabric of the above sound-absorbing layer is formed to be 7 Kgf / ㎠, and the noise attenuation coefficient (NRC) is formed to be 0.690.

[0161] Since the above sound-absorbing layer is provided inside the case of the service providing device (10), the noise of the service providing device (10) can be reduced.

[0162] The reason for limiting the constituent components, such as the nonwoven fabric, thickness, weight, and fineness constituting the sound-absorbing layer, and the numerical values ​​of the composition ratios is that, based on the analysis of test results obtained by the inventor through repeated failures, the optimal sound absorption effect was achieved at the aforementioned constituent components and numerical ratios.

[0163] Although the invention has been described above with reference to embodiments, those skilled in the art will understand that various modifications and changes can be made to the invention without departing from the spirit and scope of the invention as set forth in the following claims. Explanation of the symbols

[0165] 10 : Service provider device 20 : Network 30 : User terminal 110 : Communication interface 120: Memory 130: I / O interface 140: Processor 1410: Information Gathering Unit 1420: Operation Status Information Receiving Unit 1430: Digital Twin Model Creation Unit 1440: Movement Path Derivation Unit 1450: Reward Granting Unit

Claims

Claim 1 A method performed on a computing device having one or more processors and a memory for storing one or more programs executed by said one or more processors, comprising: a step of collecting physical information of a vessel and marine environment monitoring information; a step of receiving operation algorithms and operation status information of at least one unmanned vessel; a step of creating a digital twin model that enables unmanned vessel operation simulation by reflecting the marine environment monitoring information collected in the collecting step and the unmanned vessel operation algorithms and operation status information received in the receiving step into a virtual model; and a step of deriving an unmanned vessel movement path to derive an unmanned vessel learning algorithm according to the results of the unmanned vessel operation simulation in the digital twin model created in the creating step.The method includes, but is not limited to, the step of creating the digital twin model, which includes completing the modeling of the unmanned vessel in a Unity environment, applying the DWP2 (Dynamic Water Physics 2) asset to simulate the movement, rotation, and reaction to waves of the unmanned vessel on the sea, and simulating the physical reaction of the sea by linking wave mesh data generated through the Crest Ocean System asset with the DWP2 asset; the step of deriving the unmanned vessel's movement path, which includes training an unmanned vessel learning algorithm in an autonomous pollution control environment or destination navigation learning environment built in the digital twin model so that the unmanned vessel detects the surrounding environment using a grid sensor and calculates a path to the destination by avoiding obstacles using the detection data from the grid sensor, wherein the grid sensor distinguishes other vessels and obstacles around the unmanned vessel, and the step of deriving the unmanned vessel's movement path, which uses observation data including the Y-axis rotation value of the main vessel indicating the direction of travel of the unmanned vessel, a direction vector between the vessel and the destination representing the direction the unmanned vessel is heading toward the destination as a vector, and the distance to the destination measured between the current unmanned vessel position and the destination to determine a safe path A method for managing autonomous and swarm driving of unmanned vessels, comprising calculating, granting a reward based on the deviation angle relative to the destination direction or whether there is a collision with an obstacle regarding the unmanned vessel's movement path, and granting a reward when the unmanned vessel reaches the destination, wherein the step of deriving the unmanned vessel's movement path involves deriving the unmanned vessel's movement path by learning an unmanned vessel learning algorithm according to a scenario in which at least one unmanned vessel forms a set formation to detect a school of fish. Claim 2 delete Claim 3 delete Claim 4 delete Claim 5 delete Claim 6 delete Claim 7 delete Claim 8 A computer device comprising one or more processors and a memory for storing one or more programs executed by said one or more processors, the device comprising: an information collection unit for collecting marine environment monitoring information and physical information of a vessel; an operation status information receiving unit for receiving operation algorithms and operation status information of at least one unmanned vessel; a digital twin model creation unit for creating a digital twin model that enables unmanned vessel operation simulation by reflecting the marine environment monitoring information and physical information of a vessel collected by the information collection unit and the unmanned vessel operation algorithms and operation status information received by the operation status information receiving unit into a virtual model; and a movement path derivation unit for deriving an unmanned vessel movement path to derive an unmanned vessel learning algorithm in the digital twin model created by the digital twin model creation unit.Including, wherein the digital twin model creation unit simulates the movement, rotation, and reaction to waves of the unmanned vessel on the sea by applying the DWP2 (Dynamic Water Physics 2) asset after completing the modeling of the unmanned vessel in the Unity environment, and simulates the physical reaction of the sea by linking the wave mesh data generated through the Crest Ocean System asset with the DWP2 asset; the movement path derivation unit learns an unmanned vessel learning algorithm in an autonomous pollution control environment or destination navigation learning environment built in the digital twin model so that the unmanned vessel detects the surrounding environment using a grid sensor and calculates a path to the destination by avoiding obstacles using the detection data from the grid sensor; the grid sensor distinguishes other vessels and obstacles around the unmanned vessel; the movement path derivation unit calculates a safe path using observation data including the Y-axis rotation value of the main vessel indicating the direction of travel of the unmanned vessel, a direction vector between the vessel and the destination expressing the direction the unmanned vessel is facing the destination as a vector, and the distance to the destination measured between the current unmanned vessel position and the destination; and regarding the unmanned vessel movement path based on the destination direction An autonomous and swarm driving management device for unmanned vessels, further comprising a reward granting unit that grants a reward based on the angle of deviation or collision with an obstacle, and grants a reward when the unmanned vessel reaches a destination, wherein the movement path derivation unit derives the movement path of the unmanned vessel by learning an unmanned vessel learning algorithm according to a scenario in which at least one unmanned vessel forms a set formation to detect a school of fish.

Citation Information

Patent Citations

  • Ship-shore collaborative inland river vessel remote control method and system

    CN114217616B

  • System and method for monitoring of maritime equipment using digital twin

    KR1020230118281A

  • Method and system for testing autonomous navigation capability of unmanned ship formation

    CN115268395A

  • Peripheral situation rendering method, navigation avoidance operation learning program, navigation avoidance operation learning system, and marine vessel

    JP2021018484A