Underwater unmanned intelligent exploration system and method based on cooperation of usv and rov

Abstract

The application provides an underwater unmanned intelligent exploration method and system based on cooperation of a USV and an ROV, and comprises the following steps: S1, a target indication information of a shore-based anti-frogman sonar system is received by a comprehensive control center; S2, the target motion elements are calculated by the comprehensive control center based on the target indication information; S3, an exploration task instruction is sent to an unmanned intelligent exploration system by the comprehensive control center; S4, the unmanned intelligent exploration system goes to a specified water area, releases an ROV task load, and adjusts the posture in advance according to the target attack direction to ensure that the sonar and optical load always face the target attack direction; real-time detection images are acquired through the sonar and optical load, and the detected images are transmitted to the comprehensive control center.

Description

Technical Field

[0001] This invention relates to the field of underwater security defense for key locations, specifically to an underwater unmanned intelligent reconnaissance system and method based on the collaboration of USV and ROV, including acoustic-optical joint intelligent identification software and system reconnaissance strategies. Background Technology

[0002] For comprehensive underwater defense of key areas, and to rapidly develop underwater surveillance and detection capabilities against intrusions by small targets such as frogmen and UUVs, shore-based underwater security systems have become the preferred option. However, these systems lack verification equipment, making it impossible to confirm the identity of suspected threats alerted by the underwater security system. Therefore, developing a detection system capable of rapidly approaching designated waters to confirm underwater targets is of practical significance.

[0003] Existing technology utilizes unmanned surface vessels (USVs) equipped with imaging sonar for threat identification and rapid response. This system was tested in Germany and Italy in 2008 and 2010 respectively, and deployed in the port of La Spezia, Italy. In this case, the imaging sonar was fixedly mounted on the hull, and the waterline and the platform's own rolling motion affected acoustic performance. This invention, however, employs a flexible connection method, using an ROV to overcome the impact of the USV's waterline and the platform's rolling motion on acoustic performance.

[0004] Patent document CN213658976U (application number: 202022745609.5) discloses an underwater sonar identification system based on an unmanned vessel, including a surface system and a land system. The surface system includes an underwater data acquisition module and a surface data transmission module installed on the unmanned vessel. The land system includes a data receiving module and a data detection module. The underwater data acquisition module acquires data based on dual-frequency identification sonar. After completing the data acquisition, it sends the relevant data signals to the data receiving module through the surface data transmission module. After receiving the data signals, the data receiving module transmits the data signals to the data detection module. The data detection module can then acquire the underwater data acquired by the dual-frequency identification sonar in real time. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the purpose of this invention is to provide an underwater unmanned intelligent exploration method and system based on the collaboration of USV and ROV.

[0006] A method for underwater unmanned intelligent exploration based on the collaboration of USV and ROV, provided by the present invention, includes:

[0007] Step S1: The integrated control center receives target indication information from the shore-based anti-frogman sonar system;

[0008] Step S2: The integrated control center calculates the target motion elements based on the target indication information;

[0009] Step S3: The integrated control center sends the exploration mission command to the unmanned intelligent exploration system;

[0010] Step S4: The unmanned intelligent reconnaissance system proceeds to the designated water area, releases the ROV mission payload, and adjusts its attitude in advance according to the direction of the target's approach to ensure that the sonar and optical payloads are always pointing in the direction of the target's approach; real-time detection images are acquired through the sonar and optical payloads, and the detected images are transmitted to the integrated control center.

[0011] Preferably, step S1 employs the following method: the shore-based anti-frogman sonar system automatically detects the acoustic echo signals emitted by the active sonar, and automatically tracks and classifies the moving target based on adjacent continuous multi-cycle data to obtain target indication information.

[0012] The target indication information includes: the target's batch number, timestamp, distance, orientation, speed, threat level, and target type.

[0013] Preferably, step S2 involves the integrated control center converting the target's position information relative to the shore-based anti-frogman sonar system into the target's position information relative to the unmanned intelligent detection system based on the relative coordinates between the shore-based anti-frogman sonar system and the unmanned intelligent detection system. Based on the relative motion speed between the underwater target and the unmanned vessel in the unmanned intelligent detection system, the center calculates the encounter time and encounter position of the underwater target and the unmanned intelligent detection system moving towards each other.

[0014] Preferably, the unmanned intelligent exploration system includes: an intelligent unmanned vessel, an underwater exploration payload, a surface mission payload, and a deployment and recovery mechanism;

[0015] The intelligent unmanned vessel serves as a platform for carrying out reconnaissance mission payloads. Based on the target threat situation and mission characteristics, it can quickly approach the designated waters and execute the mission.

[0016] The underwater exploration payload uses optical and acoustic methods to detect and verify underwater targets;

[0017] The above-water mission payload obtains the USV's own location information via GPS; and, in conjunction with the ultra-short baseline underwater terminal, obtains the ROV's underwater location information in real time; at the same time, front and rear cameras are used to observe and monitor the status of the unmanned vessel in real time.

[0018] The deployment and retrieval mechanism uses an electric winch to automatically deploy and retrieve underwater exploration payloads.

[0019] Preferably, step S4 adopts the following approach: when the unmanned intelligent exploration system approaches the designated water area, if the shore-based anti-frogman sonar system can continuously provide target information, then the target information is determined to be stable. Under the condition that the target information is stable, the dual-frequency imaging sonar of the ROV platform operates in close-range imaging mode.

[0020] When the unmanned intelligent exploration system approaches the designated water area, if the shore-based anti-frogman sonar system cannot continuously provide target information, it is determined that the target information is unstable. Under the condition of unstable target information, the ROV platform actively transmits sound wave signals to the observed water area through low-frequency long-range detection sonar and preprocesses the echo to realize the detection and search of suspicious targets. When a suspected target is found, the ROV is guided to approach and then switched to close-range imaging mode.

[0021] The close-range imaging mode dynamically adjusts the attitude and direction of the ROV platform based on the real-time calculated target position and heading information, ensuring that the high-frequency imaging sonar and optical high-definition camera always illuminate the target direction until the target is approached for evidence collection.

[0022] According to the present invention, an underwater unmanned intelligent exploration system based on the collaboration of USV and ROV includes:

[0023] Module M1: The integrated control center receives target indication information from the shore-based anti-frogman sonar system;

[0024] Module M2: The integrated control center calculates the target motion elements based on the target indication information;

[0025] Module M3: The integrated control center sends the exploration mission instructions to the unmanned intelligent exploration system;

[0026] Module M4: The unmanned intelligent reconnaissance system travels to the designated waters, releases the ROV mission payload, and adjusts its attitude in advance according to the direction of the target's approach to ensure that the sonar and optical payloads are always pointing in the direction of the target's approach; it acquires real-time detection images through the sonar and optical payloads and transmits the detected images to the integrated control center.

[0027] Preferably, the module M1 adopts: a shore-based anti-frogman sonar system that automatically detects the acoustic echo signal emitted by the active sonar, and automatically tracks and classifies the moving target based on adjacent continuous multi-cycle data to obtain target indication information;

[0028] The target indication information includes: the target's batch number, timestamp, distance, orientation, speed, threat level, and target type.

[0029] Preferably, module M2 employs the following: the integrated control center converts the target's position information relative to the shore-based anti-frogman sonar system into the target's position information relative to the unmanned intelligent detection system based on the relative coordinate positions between the shore-based anti-frogman sonar system and the unmanned intelligent detection system; and calculates the encounter time and encounter position of the underwater target and the unmanned intelligent detection system moving towards each other based on the relative motion speed between the underwater target and the unmanned vessel in the unmanned intelligent detection system.

[0030] Preferably, the unmanned intelligent exploration system includes: an intelligent unmanned vessel, an underwater exploration payload, a surface mission payload, and a deployment and recovery mechanism;

[0031] The intelligent unmanned vessel serves as a platform for carrying out reconnaissance mission payloads. Based on the target threat situation and mission characteristics, it can quickly approach the designated waters and execute the mission.

[0032] The underwater exploration payload uses optical and acoustic methods to detect and verify underwater targets;

[0033] The above-water mission payload obtains the USV's own location information via GPS; and, in conjunction with the ultra-short baseline underwater terminal, obtains the ROV's underwater location information in real time; at the same time, front and rear cameras are used to observe and monitor the status of the unmanned vessel in real time.

[0034] The deployment and retrieval mechanism uses an electric winch to automatically deploy and retrieve underwater exploration payloads.

[0035] Preferably, module M4 adopts the following approach: when the unmanned intelligent exploration system approaches the designated water area, if the shore-based anti-frogman sonar system can continuously provide target information, it is determined that the target information is stable. Under the condition that the target information is stable, the dual-frequency imaging sonar of the ROV platform operates in close-range imaging mode.

[0036] When the unmanned intelligent exploration system approaches the designated water area, if the shore-based anti-frogman sonar system cannot continuously provide target information, it is determined that the target information is unstable. Under the condition of unstable target information, the ROV platform actively transmits sound wave signals to the observed water area through low-frequency long-range detection sonar and preprocesses the echo to realize the detection and search of suspicious targets. When a suspected target is found, the ROV is guided to approach and then switched to close-range imaging mode.

[0037] The close-range imaging mode dynamically adjusts the attitude and direction of the ROV platform based on the real-time calculated target position and heading information, ensuring that the high-frequency imaging sonar and optical high-definition camera always illuminate the target direction until the target is approached for evidence collection.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] 1. This invention employs multi-frequency multi-sonar joint processing technology, including ROV dual-frequency imaging sonar and long-range low-frequency detection sonar to achieve multi-level and multi-mode exploration of long-range point target detection and close-range volume target contour imaging; at the tracking level, acoustic and optical features are processed separately using neural network technology; at the decision level, the acoustic and optical processing results are jointly processed through expert decision-making to achieve integrated acoustic-optical recognition of underwater targets, enabling the unmanned exploration system to have a multiplier effect of 1+1>2, and improving the intelligent perception of the underwater environment and the autonomous target recognition capability;

[0040] 2. This invention adopts a "mother-child" collaborative working mode combining USV and ROV, giving full play to the rapid response and high efficiency of the USV platform to achieve rapid approach. As the mother ship platform, it is responsible for communication with the shore-based integrated control center and releases the ROV child platform. Utilizing the flexibility of the multi-degree-of-freedom propeller ROV, it can maneuver and quickly adjust and correct the attitude and depth of the underwater exploration mission payload to lock onto suspicious targets, thereby overcoming the impact of the unmanned surface vessel's shallow draft and the platform's pitch and roll on acoustic performance. Attached Figure Description

[0041] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0042] Figure 1 This is a system composition diagram of the present invention.

[0043] Figure 2 This is a diagram of the software composition of the integrated control and display center.

[0044] Figure 3 This is a diagram showing the composition of the underwater exploration payload.

[0045] Figure 4 This is a typical exploration workflow of the present invention.

[0046] Figure 5 This is a block diagram illustrating the components of the present invention in conjunction with application examples.

[0047] Figure 6 This is a schematic diagram of the system composition of the present invention in conjunction with an application example.

[0048] Figure 7 This is a software implementation block diagram of the present invention combined with application examples.

[0049] Figure 8 This is a schematic diagram illustrating the exploration results of this invention in conjunction with an application example. Detailed Implementation

[0050] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0051] The purpose of this invention is to fill the gap in existing underwater security measures for key locations, which currently lack underwater verification capabilities. By developing an unmanned intelligent reconnaissance system based on acoustic-optical integration and employing a collaborative combat mode combining USV and ROV, the invention aims to achieve intelligent detection and identification of typical targets such as underwater divers, UUVs, underwater pre-positioned objects, and small-volume salvage items.

[0052] This invention utilizes the speed and maneuverability of USVs to rapidly transport ROVs to designated waters; then, the ROV is autonomously released from the USV to conduct reconnaissance missions; the USV is responsible for powering the ROV and transmitting communication data, and the ROV's reconnaissance results are exchanged with the "shore-based integrated control and display center" via the USV. After completing the mission, the USV autonomously recovers the ROV and returns.

[0053] This invention fully leverages the speed and efficiency of the USV platform, rapidly carrying mission payloads to designated waters; and utilizes the flexibility of a multi-degree-of-freedom propeller-driven ROV to maneuver and quickly adjust and correct the attitude and depth of the underwater exploration payload to lock onto suspicious targets; and uses integrated acoustic and optical video information recognition processing software to confirm the target's identity. Compared to foreign solutions that fix acoustic payloads to unmanned surface vessels, this invention overcomes the impact of the unmanned surface vessel's shallow draft and the platform's pitch and roll on acoustic performance.

[0054] Example 1

[0055] According to the present invention, an underwater unmanned intelligent exploration method based on the collaboration of USV and ROV includes:

[0056] Step S1: The integrated control center receives target indication information from the shore-based anti-frogman sonar system;

[0057] Specifically, step S1 employs the following: the shore-based anti-frogman sonar system automatically detects the acoustic echo signals emitted by the active sonar, and automatically tracks and classifies the moving target based on adjacent continuous multi-cycle data to obtain target indication information;

[0058] The target indication information includes: the target's batch number, timestamp, distance, orientation, speed, threat level, and target type.

[0059] Step S2: The integrated control center calculates the target motion elements based on the target indication information;

[0060] Specifically, step S2 involves the integrated control center converting the target's position information relative to the shore-based anti-frogman sonar system into its position information relative to the unmanned intelligent detection system based on the relative coordinates between the shore-based anti-frogman sonar system and the unmanned intelligent detection system. Based on the relative motion speed between the underwater target and the unmanned vessel in the unmanned intelligent detection system, the center calculates the encounter time and encounter position of the underwater target and the unmanned intelligent detection system moving towards each other.

[0061] Step S3: The integrated control center sends the exploration mission command to the unmanned intelligent exploration system;

[0062] The unmanned intelligent exploration system includes: an intelligent unmanned vessel, an underwater exploration payload, a surface mission payload, and a deployment and recovery mechanism;

[0063] The intelligent unmanned vessel serves as a platform for carrying out reconnaissance mission payloads. Based on the target threat situation and mission characteristics, it can quickly approach the designated waters and execute the mission.

[0064] The underwater exploration payload uses optical and acoustic methods to detect and verify underwater targets;

[0065] The payload for the maritime mission includes a GPS device, front- and rear-facing full HD night vision cameras, and an ultra-short baseline maritime terminal. Furthermore, the HD cameras have built-in microphones and speakers for voice communication.

[0066] The deployment and retrieval mechanism uses an electric winch to automatically deploy and retrieve underwater exploration payloads.

[0067] Step S4: The unmanned intelligent reconnaissance system proceeds to the designated water area, releases the ROV mission payload, and adjusts its attitude in advance according to the direction of the target's approach to ensure that the sonar and optical payloads are always pointing in the direction of the target's approach; real-time detection images are acquired through the sonar and optical payloads, and the detected images are transmitted to the integrated control center.

[0068] The designated water area is a water area centered on the location where the target may appear at the moment of encounter, with the imaging distance as its radius;

[0069] Specifically, the unmanned surface vessel (ROV) responds quickly to instructions and activates its propulsion system, traveling at maximum speed to the designated water area. It then decelerates and uses inertia to drift to the predetermined position. At this point, the ROV payload is released from the aft deck into the water via an automatic deployment device. The ROV payload adjusts its attitude in advance according to the direction of the incoming target, ensuring that the sonar and optical payloads are always pointed in that direction.

[0070] More specifically, step S4 adopts the following approach: when the unmanned intelligent exploration system approaches the designated water area, if the shore-based anti-frogman sonar system can continuously provide target information, then the target information is determined to be stable. Under the condition that the target information is stable, the dual-frequency imaging sonar of the ROV platform operates in close-range imaging mode.

[0071] When the unmanned intelligent reconnaissance system approaches the designated water area, if the shore-based anti-frogman sonar system cannot continuously provide target information, the target information is considered unstable. Under this unstable condition, the ROV platform actively transmits sound wave signals into the observed water area using low-frequency long-range sonar and preprocesses the echoes to detect and search for suspicious targets. When a suspected target is detected, the ROV is guided to approach and then switched to close-range imaging mode. The BP neural network model is trained using prior acoustic and optical data training samples from historical experiments. Once the training converges, the network weights are obtained. The trained BP neural network is then used to process the acoustic and optical data test samples in real time to obtain the target category; specifically, for example... Figure 7 As shown, (1) First, data collection is carried out on potential underwater targets by combining historical lake and sea test data, including acoustic data and optical data; (2) Data cleaning (removal) is performed on these data, and feature vector space is established according to factors such as target type, sensor type, and target status, and data annotation is performed to form a sample database; (3) Machine learning training is performed using these annotated data (neural network is selected here) to form the network parameters of this case; (4) The trained network is used to perform single-sensor target classification and recognition of optical features and acoustic features respectively; (5) The target classification and recognition results of each single sensor are time-aligned and spatial-aligned, and then the expert scoring method is used to make a decision on the final sound and light joint recognition result, and finally the target identity is confirmed.

[0072] The final audio-visual joint recognition result determined by the expert scoring method includes: (1) determining the weight coefficients of optical and acoustic results according to the working mode. In the search mode, a heavier weight coefficient is assigned to the acoustic results and a lighter weight coefficient is assigned to the optical results. (2) When in the imaging working mode, the weight coefficients of optical and acoustic results are determined according to different distance segments. At long distances (e.g., more than 20 meters), a heavier weight coefficient is assigned to the acoustic results and a lighter weight coefficient is assigned to the optical results; at short distances (within 20 meters), the weight coefficients of acoustic and optical results are evenly distributed. (3) calculating the total score of the indicators, multiplying the weight of each indicator by the corresponding result to obtain the score of that indicator. (4) deciding the final audio-visual joint recognition result based on the total score.

[0073] The close-range imaging mode dynamically adjusts the attitude and direction of the ROV platform based on the real-time calculated target position and heading information, ensuring that the high-frequency imaging sonar and optical high-definition camera always illuminate the target direction until the target is approached for evidence collection.

[0074] An underwater unmanned intelligent exploration system based on the collaboration of USV and ROV, according to the present invention, includes:

[0075] Module M1: The integrated control center receives target indication information from the shore-based anti-frogman sonar system;

[0076] Specifically, module M1 employs a shore-based anti-frogman sonar system that automatically detects the acoustic echo signals emitted by the active sonar and automatically tracks and classifies the moving target based on adjacent continuous multi-cycle data to obtain target indication information.

[0077] The target indication information includes: the target's batch number, timestamp, distance, orientation, speed, threat level, and target type.

[0078] Module M2: The integrated control center calculates the target motion elements based on the target indication information;

[0079] Specifically, module M2 employs the following: The integrated control center converts the target's position information relative to the shore-based anti-frogman sonar system into the target's position information relative to the unmanned intelligent detection system based on the relative coordinate positions between the shore-based anti-frogman sonar system and the unmanned intelligent detection system. Based on the relative motion speed between the underwater target and the unmanned vessel in the unmanned intelligent detection system, the meeting time and meeting position of the underwater target and the unmanned intelligent detection system moving towards each other are calculated.

[0080] Module M3: The integrated control center sends the exploration mission instructions to the unmanned intelligent exploration system;

[0081] The unmanned intelligent exploration system includes: an intelligent unmanned vessel, an underwater exploration payload, a surface mission payload, and a deployment and recovery mechanism;

[0082] The intelligent unmanned vessel serves as a platform for carrying out reconnaissance mission payloads. Based on the target threat situation and mission characteristics, it can quickly approach the designated waters and execute the mission.

[0083] The underwater exploration payload uses optical and acoustic methods to detect and verify underwater targets;

[0084] The payload for the maritime mission includes a GPS device, front- and rear-facing full HD night vision cameras, and an ultra-short baseline maritime terminal. Furthermore, the HD cameras have built-in microphones and speakers for voice communication.

[0085] The deployment and retrieval mechanism uses an electric winch to automatically deploy and retrieve underwater exploration payloads.

[0086] Module M4: The unmanned intelligent reconnaissance system travels to the designated waters, releases the ROV mission payload, and adjusts its attitude in advance according to the direction of the target's approach to ensure that the sonar and optical payloads are always pointing in the direction of the target's approach; it acquires real-time detection images through the sonar and optical payloads and transmits the detected images to the integrated control center.

[0087] The designated water area is a water area centered on the location where the target may appear at the moment of encounter, with the imaging distance as its radius;

[0088] Specifically, the unmanned surface vessel (ROV) responds quickly to instructions and activates its propulsion system, traveling at maximum speed to the designated water area. It then decelerates and uses inertia to drift to the predetermined position. At this point, the ROV payload is released from the aft deck into the water via an automatic deployment device. The ROV payload adjusts its attitude in advance according to the direction of the incoming target, ensuring that the sonar and optical payloads are always pointed in that direction.

[0089] More specifically, module M4 adopts the following approach: when the unmanned intelligent exploration system approaches the designated water area, if the shore-based anti-frogman sonar system can continuously provide target information, it is determined that the target information is stable. Under the condition that the target information is stable, the dual-frequency imaging sonar of the ROV platform operates in close-range imaging mode.

[0090] When the unmanned intelligent reconnaissance system approaches the designated water area, if the shore-based anti-frogman sonar system cannot continuously provide target information, the target information is considered unstable. Under this unstable condition, the ROV platform actively transmits sound wave signals into the observed water area using low-frequency long-range sonar and preprocesses the echoes to detect and search for suspicious targets. When a suspected target is detected, the ROV is guided to approach and then switched to close-range imaging mode. The BP neural network model is trained using prior acoustic and optical data training samples from historical experiments. Once the training converges, the network weights are obtained. The trained BP neural network is then used to process the acoustic and optical data test samples in real time to obtain the target category; specifically, for example... Figure 7As shown, (1) First, data collection is carried out on potential underwater targets by combining historical lake and sea test data, including acoustic data and optical data; (2) Data cleaning (removal) is performed on these data, and feature vector space is established according to factors such as target type, sensor type, and target status, and data annotation is performed to form a sample database; (3) Machine learning training is performed using these annotated data (neural network is selected here) to form the network parameters of this case; (4) The trained network is used to perform single-sensor target classification and recognition of optical features and acoustic features respectively; (5) The target classification and recognition results of each single sensor are time-aligned and spatial-aligned, and then the expert scoring method is used to make a decision on the final sound and light joint recognition result, and finally the target identity is confirmed.

[0091] The final audio-visual joint recognition result determined by the expert scoring method includes: (1) determining the weight coefficients of optical and acoustic results according to the working mode. In the search mode, a heavier weight coefficient is assigned to the acoustic results and a lighter weight coefficient is assigned to the optical results. (2) When in the imaging working mode, the weight coefficients of optical and acoustic results are determined according to different distance segments. At long distances (e.g., more than 20 meters), a heavier weight coefficient is assigned to the acoustic results and a lighter weight coefficient is assigned to the optical results; at short distances (within 20 meters), the weight coefficients of acoustic and optical results are evenly distributed. (3) calculating the total score of the indicators, multiplying the weight of each indicator by the corresponding result to obtain the score of that indicator. (4) deciding the final audio-visual joint recognition result based on the total score.

[0092] The close-range imaging mode dynamically adjusts the attitude and direction of the ROV platform based on the real-time calculated target position and heading information, ensuring that the high-frequency imaging sonar and optical high-definition camera always illuminate the target direction until the target is approached for evidence collection.

[0093] Example 2

[0094] Example 2 is a preferred example of Example 1.

[0095] The present invention provides an unmanned reconnaissance system for underwater security in key locations, comprising: a shore-based integrated control and display center, an intelligent unmanned surface vessel, an underwater reconnaissance payload, a surface mission payload and a deployment and recovery subsystem, and acoustic-optical combined intelligent identification software. The system composition diagram is shown below. Figure 1 As shown. Wherein:

[0096] The shore-based integrated control and display center serves as the information integration, visualization, and intelligent command and control hub for the unmanned platform. It integrates control and displays parameters of reconnaissance equipment; it can combine target information and nautical charts to formulate target reconnaissance strategies; and it can display real-time status information of the unmanned platform and equipment. Its hardware design primarily includes high-performance servers, multi-screen displays, touchscreens, and joysticks; such as... Figure 2 As shown, its software design scheme mainly realizes the coordination of sonar information with shore-based anti-frogman, the comprehensive situation display of multi-source information, the setting and control of unmanned verification equipment, the control of access detection equipment, the tracking, classification, threat assessment and strategy formulation of targets, and the comprehensive display and reporting of verification results.

[0097] The intelligent unmanned vessel serves as a platform for carrying out reconnaissance mission payloads. Based on the target threat situation and mission characteristics, it can quickly approach designated waters to execute missions. Simultaneously, it can establish communication with the integrated control and display center via wireless means to conduct control commands, information exchange, and data transmission.

[0098] The underwater exploration payload utilizes optical and acoustic methods to detect and verify underwater targets; it mainly includes an underwater ROV body, a long-range low-frequency detection sonar, a dual-frequency short-range imaging sonar, a high-definition optical camera, and an ultra-short baseline underwater terminal; such as Figure 3 As shown.

[0099] Specifically, optical means refer to video imaging of the observed target using a high-definition camera; acoustic means refer to acquiring the target outline using active high-frequency imaging sonar.

[0100] The payload for the surface mission shown primarily includes a GPS device, front and rear full HD night vision cameras, and an ultra-short baseline surface terminal. Additionally, the HD cameras have built-in microphones and speakers for voice communication.

[0101] The payload for maritime missions uses GPS to acquire the USV's location information and, in conjunction with an ultra-short baseline maritime terminal, to acquire the ROV's underwater location information in real time. In addition, front and rear cameras are used to observe and monitor the status of the unmanned vessel in real time (including real-time monitoring of the ROV's deployment and retrieval).

[0102] The deployment and retrieval mechanism: automatically deploys and retrieves underwater exploration payloads via an electric winch.

[0103] The aforementioned acoustic-optical joint intelligent recognition software, based on acoustic and optical observation data, mines and analyzes the significant differences in the acoustic-optical characteristics of typical underwater targets, selects reasonable and useful features to establish a feature vector space, forms a sample database, and uses intelligent processing methods to achieve autonomous detection and recognition of underwater targets in "no-human loops," enabling the unmanned exploration system to have a multiplier effect of 1+1>2, and improving the intelligent perception of the underwater environment and the autonomous recognition of targets.

[0104] Optical methods can directly reflect the appearance of a target, but the imaging range is short and a light source is required;

[0105] Acoustic imaging methods can achieve long-distance imaging; however, the imaging contours cannot directly reflect the appearance of the target.

[0106] By combining optical and acoustic methods, we can fully leverage their respective strengths and achieve complementary advantages.

[0107] Based on the operational characteristics of underwater security threats to key locations, unmanned reconnaissance systems mainly face two typical situations:

[0108] (1) If the tactical measures taken by the target to be explored are hovering or lurking, then this scenario becomes a "movement to investigate stillness" scenario, similar to "exploration of pre-placed objects". This situation is defined as guided exploration.

[0109] (2) If the target's tactical measures are to maneuver to evade or escape, then this scenario becomes a "movement-based investigation" scenario, similar to "probing a slow-moving target". This situation is defined here as autonomous search and verification.

[0110] In view of the above situation, the typical exploration workflow of the present invention is as follows: Figure 4 As shown.

[0111] Step S1: The integrated control center receives target indication information from the shore-based anti-frogman sonar system;

[0112] Step S2: The integrated control center calculates the target motion elements based on the target indication information; the integrated control center converts the target's position information (x', y') "relative to the coordinates of the shore-based anti-frogman sonar system" into position information (x, y) "relative to the coordinates of this system" according to the relative coordinates between the shore-based anti-frogman sonar system and this system, and calculates the encounter time and encounter position of the two moving towards each other based on the relative motion speed of the underwater target and the unmanned vessel in this system;

[0113] Assume the coordinate system of this system is Oxy, and the coordinate system of the shore-based anti-frogman sonar system is O'x'y'. Here, O and O' are the origin, and the coordinates of O' in Oxy are (x0, y0), with angle t from the x-axis to the x'-axis. The coordinate transformation formula is:

[0114] x=x'cost-y'sint+x0,y=x'sint+y'cost+y0.

[0115] Step S3: The integrated control center sends the exploration mission command to the unmanned intelligent exploration system;

[0116] Step S4: The unmanned intelligent reconnaissance system proceeds to the designated waters, releases its ROV mission payload, and adjusts its attitude in advance according to the direction of the incoming target to ensure that the sonar and optical payloads are always pointing in that direction. The designated waters are determined based on step S2 and combined with nautical chart information. Real-time detection images are acquired through the sonar and optical payloads and transmitted to the integrated control center.

[0117] Example 3

[0118] Example 3 is a preferred example of Example 1.

[0119] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings and examples.

[0120] An unmanned reconnaissance system for underwater security in key locations mainly comprises a shore-based integrated control and display center, an intelligent unmanned surface vessel, an underwater reconnaissance payload, a surface mission payload, and a deployment and recovery subsystem. System block diagrams and schematic diagrams are shown below. Figure 5 and Figure 6 As shown.

[0121] In practical implementation, after the key underwater security system detects a threatening target, it sends the target information and mission instructions with audible and visual alarms to the unmanned reconnaissance system in this invention. In this case, the integrated control and display center first analyzes the motion elements and lead time based on the target's location information and motion parameters, combined with the overall situation and the unmanned platform's motion parameters, to confirm the verification point. Then, the unmanned vessel carrying the mission payload, according to the integrated control instructions, quickly responds and approaches the verification point, completing actions such as releasing the underwater acoustic payload. Next, the underwater robot equipped with the acoustic payload hovers in the water, illuminating the direction of the incoming target, and performs search and detection and target identification for suspected targets, such as... Figure 8 As shown.

[0122] The acoustic-optical joint intelligent recognition software in this implementation case mainly adopts underwater typical target classification technology based on machine learning and feature engineering. The basic idea is to mine and analyze the acoustic and optical saliency differences of underwater typical targets, select reasonable and useful features to establish a feature vector space, form a sample database, build a joint classifier for expert decision-making and machine recognition, and use the underwater target feature vector of the tracked target to identify the target by mapping the feature space vector to the feature differences in the high-dimensional space, thereby solving the automatic classification and recognition capability. Figure 7 A flowchart of the processing procedure is provided.

[0123] Those skilled in the art will understand that, in addition to implementing the system, apparatus, and their modules provided by this invention in purely computer-readable program code, the same program can be implemented in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system, apparatus, and their modules provided by this invention can be considered a hardware component, and the modules included therein for implementing various programs can also be considered structures within the hardware component; alternatively, modules for implementing various functions can be considered both software programs implementing the method and structures within the hardware component.

[0124] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A method for underwater unmanned intelligent exploration based on the collaboration of USV and ROV, characterized in that, include: Step S1: The integrated control center receives target indication information from the shore-based anti-frogman sonar system; Step S2: The integrated control center calculates the target motion elements based on the target indication information; Step S3: The integrated control center sends the exploration mission command to the unmanned intelligent exploration system; Step S4: The unmanned intelligent reconnaissance system proceeds to the designated water area, releases the ROV mission payload, and adjusts its attitude in advance according to the direction of the target's approach to ensure that the sonar and optical payloads are always pointing in the direction of the target's approach; real-time detection images are acquired through the sonar and optical payloads, and the detected images are transmitted to the integrated control center; Step S4 adopts the following approach: When the unmanned intelligent exploration system approaches the designated water area, if the shore-based anti-frogman sonar system can continuously provide target information, it is determined that the target information is stable. Under the condition that the target information is stable, the dual-frequency imaging sonar of the ROV platform operates in close-range imaging mode. When the unmanned intelligent exploration system approaches the designated water area, if the shore-based anti-frogman sonar system cannot continuously provide target information, it is determined that the target information is unstable. Under the condition of unstable target information, the ROV platform actively transmits sound wave signals to the observed water area through low-frequency long-range detection sonar and preprocesses the echo to realize the detection and search of suspicious targets. When a suspected target is found, the ROV is guided to approach and then switched to close-range imaging mode. The close-range imaging mode dynamically adjusts the attitude and direction of the ROV platform based on the real-time calculated target position and heading information, ensuring that the high-frequency imaging sonar and optical high-definition camera always illuminate the target direction until the target is approached for evidence collection.

2. The underwater unmanned intelligent exploration method based on USV and ROV collaboration according to claim 1, characterized in that, Step S1 employs the following: The shore-based anti-frogman sonar system automatically detects the acoustic echo signals emitted by the active sonar, and automatically tracks and classifies the moving target based on adjacent continuous multi-cycle data to obtain target indication information. The target indication information includes: the target's batch number, timestamp, distance, orientation, speed, threat level, and target type.

3. The underwater unmanned intelligent exploration method based on USV and ROV collaboration according to claim 1, characterized in that, Step S2 involves the integrated control center converting the target's position information relative to the shore-based anti-frogman sonar system into its position information relative to the unmanned intelligent detection system based on the relative coordinates between the shore-based anti-frogman sonar system and the unmanned intelligent detection system. Based on the relative motion speed between the underwater target and the unmanned vessel in the unmanned intelligent detection system, the center calculates the encounter time and position of the underwater target and the unmanned intelligent detection system moving towards each other.

4. The underwater unmanned intelligent exploration method based on USV and ROV collaboration according to claim 1, characterized in that, The unmanned intelligent exploration system includes: an intelligent unmanned vessel, an underwater exploration payload, a surface mission payload, and a deployment and recovery mechanism; The intelligent unmanned vessel serves as a platform for carrying out reconnaissance mission payloads. Based on the target threat situation and mission characteristics, it can quickly approach the designated waters and execute the mission. The underwater exploration payload uses optical and acoustic methods to detect and verify underwater targets; The above-water mission payload obtains the USV's own location information via GPS; and, in conjunction with the ultra-short baseline underwater terminal, obtains the ROV's underwater location information in real time; at the same time, front and rear cameras are used to observe and monitor the status of the unmanned vessel in real time. The deployment and retrieval mechanism uses an electric winch to automatically deploy and retrieve underwater exploration payloads.

5. An underwater unmanned intelligent exploration system based on the collaboration of USV and ROV, characterized in that, include: Module M1: The integrated control center receives target indication information from the shore-based anti-frogman sonar system; Module M2: The integrated control center calculates the target motion elements based on the target indication information; Module M3: The integrated control center sends the exploration mission instructions to the unmanned intelligent exploration system; Module M4: The unmanned intelligent reconnaissance system travels to the designated waters, releases the ROV mission payload, and adjusts its attitude in advance according to the direction of the target's approach to ensure that the sonar and optical payloads are always pointing in the direction of the target's approach; it acquires real-time detection images through the sonar and optical payloads and transmits the detected images to the integrated control center; The module M4 adopts the following approach: when the unmanned intelligent exploration system approaches the designated water area, if the shore-based anti-frogman sonar system can continuously provide target information, it is determined that the target information is stable. Under the condition that the target information is stable, the dual-frequency imaging sonar of the ROV platform works in close-range imaging mode. When the unmanned intelligent exploration system approaches the designated water area, if the shore-based anti-frogman sonar system cannot continuously provide target information, it is determined that the target information is unstable. Under the condition of unstable target information, the ROV platform actively transmits sound wave signals to the observed water area through low-frequency long-range detection sonar and preprocesses the echo to realize the detection and search of suspicious targets. When a suspected target is found, the ROV is guided to approach and then switched to close-range imaging mode. The close-range imaging mode dynamically adjusts the attitude and direction of the ROV platform based on the real-time calculated target position and heading information, ensuring that the high-frequency imaging sonar and optical high-definition camera always illuminate the target direction until the target is approached for evidence collection.

6. The underwater unmanned intelligent exploration system based on USV and ROV collaboration according to claim 5, characterized in that, The module M1 employs a shore-based anti-frogman sonar system that automatically detects the acoustic echo signals emitted by the active sonar and automatically tracks and classifies the moving target based on adjacent continuous multi-cycle data to obtain target indication information. The target indication information includes: the target's batch number, timestamp, distance, orientation, speed, threat level, and target type.

7. The underwater unmanned intelligent exploration system based on USV and ROV collaboration according to claim 5, characterized in that, The module M2 employs the following: The integrated control center converts the target's position information relative to the shore-based anti-frogman sonar system into the target's position information relative to the unmanned intelligent detection system based on the relative coordinates between the shore-based anti-frogman sonar system and the unmanned intelligent detection system. Based on the relative motion speed between the underwater target and the unmanned vessel in the unmanned intelligent detection system, the meeting time and meeting position of the underwater target and the unmanned intelligent detection system moving towards each other are calculated.

8. The underwater unmanned intelligent exploration system based on USV and ROV collaboration according to claim 5, characterized in that, The unmanned intelligent exploration system includes: an intelligent unmanned vessel, an underwater exploration payload, a surface mission payload, and a deployment and recovery mechanism; The intelligent unmanned vessel serves as a platform for carrying out reconnaissance mission payloads. Based on the target threat situation and mission characteristics, it can quickly approach the designated waters and execute the mission. The underwater exploration payload uses optical and acoustic methods to detect and verify underwater targets; The above-water mission payload obtains the USV's own location information via GPS; and, in conjunction with the ultra-short baseline underwater terminal, obtains the ROV's underwater location information in real time; at the same time, front and rear cameras are used to observe and monitor the status of the unmanned vessel in real time. The deployment and retrieval mechanism uses an electric winch to automatically deploy and retrieve underwater exploration payloads.

Images