Water area monitoring system for cooperative operation of unmanned aerial vehicle and unmanned ship based on Mesh ad hoc network

Abstract

The invention provides a water area monitoring system for cooperative operation of an unmanned aerial vehicle and an unmanned ship based on a Mesh ad hoc network. The water area monitoring system comprises the unmanned aerial vehicle, the unmanned ship, a Mesh ad hoc network module, a cooperative control system and a shore end control system, according to the unmanned plane, wide-area patrol and image acquisition, airspace and wide-area water area information is acquired through a sensor, fixed-point acquisition and water area information acquisition are realized through an unmanned ship, and data interaction is realized by relying on nodes of a Mesh ad hoc network. Through the multi-hop relay and dynamic routing technology of the Mesh ad hoc network, autonomous and reliable communication in a scene without a fixed base station is realized, and the communication coverage range and survivability are improved; on the basis of distributed task allocation and real-time data interaction, the equipment cooperation efficiency is remarkably improved, the response delay is shortened, and efficient completion of tasks in a complex environment is guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of environmental monitoring, in particular to a water area monitoring system based on cooperation between unmanned aerial vehicles and unmanned ships in a Mesh self-organizing network. BACKGROUND

[0002] Traditional water area monitoring mainly relies on manual patrol, satellite remote sensing and fixed sensor networks, and there are great limitations in water area monitoring technology. Manual patrol is low in efficiency, limited in coverage, and restricted by bad weather and complex sea conditions, and cannot realize real-time dynamic monitoring. Satellite remote sensing is difficult to capture subtle changes in local water areas due to cloud cover and resolution limitations, and the data update cycle is long. Fixed sensor networks have high deployment costs and are difficult to maintain, and cannot adapt to dynamic changes in the water environment.

[0003] At present, there are many water quality sampling and analysis methods using unmanned aerial vehicles, mainly composed of unmanned aerial vehicles, ground control units and water sampling units. Unmanned aerial vehicles can quickly reach the sampling point, and through the sensors carried on the unmanned aerial vehicles, they can perform depth sampling as needed to obtain water samples, avoiding the need for workers to enter dangerous water areas for sampling, especially for sudden pollution incidents, which can quickly sample the core area of pollution.

[0004] With the widespread use of communication technology, the cooperative monitoring of unmanned aerial vehicles and unmanned ships has become a research hotspot. The existing technology uses unmanned aerial vehicles for wide-area patrol and unmanned ships for fixed-point sampling, and combines 4G / 5G communication to achieve data backhaul. However, in remote sea areas or mountainous water areas, insufficient base station signal coverage can easily lead to communication interruption. During the operation, the movement of unmanned aerial vehicles and unmanned ships can cause frequent changes in network nodes, and traditional star or chain network architecture cannot real-time reconstruct communication links, and the existing system mostly uses centralized task allocation, lacks distributed autonomous decision-making capability, and is difficult to cope with sudden situations in complex environments. SUMMARY

[0005] Based on the above description, the present application proposes a water area monitoring system based on cooperation between unmanned aerial vehicles and unmanned ships in a Mesh self-organizing network, which includes unmanned aerial vehicles, unmanned ships, a Mesh self-organizing network, a cooperative control system and a shore control system. The unmanned aerial vehicles, unmanned ships, cooperative control system and unmanned aerial vehicles, unmanned ships and shore control system communicate through the Mesh self-organizing network module, and through the multi-hop relay and dynamic routing technology of the Mesh self-organizing network, the communication is self-reliant and reliable in the absence of fixed base stations, improving the communication coverage and invulnerability. Relying on distributed task allocation and real-time data interaction, the cooperative efficiency of unmanned aerial vehicles and unmanned ships is significantly improved, the response delay is shortened, and the efficient completion of tasks in complex environments is ensured.

[0006] The following is the technical solution of the present invention: a water monitoring system based on a mesh self-organizing network for collaborative operation of unmanned aerial vehicles (UAVs) and unmanned surface vessels (USVs), comprising a UAV, an USV, a mesh self-organizing network module, a collaborative control system, and a shore-based control system; the UAVs and USVs communicate with the collaborative control system and with the shore-based control system through the mesh self-organizing network module; Drones are used for wide-area patrols and image acquisition. They collect airspace and wide-area water information through sensors and achieve data interaction by relying on the nodes of the Mesh self-organizing network. The drone integrates, but is not limited to, a Mesh self-organizing network module mobile terminal, an onboard computer, a flight control module, a drone drive module, a drone power actuator, a three-axis gimbal camera, a GPS + magnetic compass, an attitude sensor, a speed sensor, a voltage detection module, and distance detection radars installed in the front, back, left, right, up, and down directions of the drone. The mobile terminal of the Mesh self-organizing network module integrated in the UAV receives control commands from the collaborative control system and uploads the UAV status information. The onboard computer and flight control module form a closed-loop control logic of "perception-decision-execution" based on real-time feedback from the sensors on the UAV. The UAV drive module drives the power actuator to adjust the UAV's flight attitude and speed, and synchronously controls the gimbal camera to collect video and images and transmit them back to the shore control system and the unmanned surface vessel. The unmanned surface vessel (USV) is used for both fixed-point data collection and water area information collection. The USV integrates, but is not limited to, a Mesh self-organizing network module mobile terminal, an industrial control computer, a USV drive module, a USV power actuator, a sampling actuator, a MEMS inertial navigation system, an RTK-GPS autonomous navigation module, a voltage detection module, millimeter-wave radar, and a bow camera, constructing a comprehensive state perception system encompassing "positioning, attitude, power, and environment." The RTK-GPS autonomous navigation module provides centimeter-level latitude and longitude positioning; the MEMS inertial navigation system synchronously outputs roll angle, pitch angle, and the USV's hull speed; the voltage detection module provides real-time feedback on remaining battery power; and millimeter-wave radars in the forward, aft, left, and right directions detect obstacles on the water surface, while the bow camera supplements visual environmental information. The mobile terminal of the Mesh self-organizing network module integrated on the unmanned surface vessel (USV) receives control commands from the collaborative control system. The industrial control computer integrates the obstacle avoidance data from the RTK-GPS autonomous navigation module, MEMS inertial navigation attitude, and millimeter-wave radar to generate drive signals. It controls the USV's power actuators to adjust its speed and heading, and schedules the USV's sampling actuators to complete water sampling operations. The environmental video collected by the camera at the bow is encrypted and transmitted back to the shore control system via the Mesh self-organizing network link. The network topology of the Mesh self-organizing network module is centered on the local end of the Mesh self-organizing network module mounted on the shore control system. It connects downwards to at least Mesh self-organizing network module mobile terminal 1, Mesh self-organizing network module mobile terminal 2, Mesh self-organizing network module mobile terminal 3, and Mesh self-organizing network module mobile terminal 4. The nodes are connected to each other through Mesh wireless links to build multi-hop, redundant, decentralized connections. Among them, Mesh self-organizing network module mobile terminal 1 is associated with the UAV, Mesh self-organizing network module mobile terminal 2 is associated with the unmanned surface vessel, and the two achieve cross-platform interaction through a collaborative control system. Mesh self-organizing network module mobile terminal 3, Mesh self-organizing network module mobile terminal 4 and other devices extend the network coverage and functions.

[0007] The preferred workflow of a water monitoring system that combines drones and unmanned surface vessels is as follows: S1 mission released: The shore-based control system calls the electronic chart module to automatically generate a route suitable for collaborative operations of UAVs and unmanned surface vessels based on the monitoring targets and start and end point coordinates input by the user, and integrates them into a structured overall task that includes, but is not limited to, monitoring content, action commands and task start points. S2 Task Assignment: Based on the status information and network conditions of UAVs and unmanned surface vessels, the collaborative control system divides the structured overall task into scenario-based sub-tasks: through the local terminal of the Mesh self-organizing network module, using its multi-directional broadcast link, the scenario-based sub-tasks are distributed to Mesh self-organizing network module mobile terminal 1, Mesh self-organizing network module mobile terminal 2, Mesh self-organizing network module mobile terminal 3, Mesh self-organizing network module mobile terminal 4 and external devices. S3 Collaborative Monitoring: Unmanned surface vessels (USVs) and unmanned aerial vehicles (UAVs) conduct complementary monitoring operations based on scenario-based task assignments. UAVs collect airspace and wide-area water information through sensors, while USVs collect information at fixed points and in water areas through sensors. Real-time data interaction is achieved through a Mesh self-organizing network module mobile terminal 1 integrated on the UAV and a Mesh self-organizing network module mobile terminal 2 integrated on the USV. The information transmitted back by the UAV provides accurate positioning guidance for USV sampling, and the data collected by the USV on demand feeds back to the UAV, optimizing the pollution identification model on the UAV and forming an air-water collaborative closed loop of "aerial identification - hierarchical sampling - data feedback". S4 Dynamic Adaptation and Optimization: The Mesh self-organizing network module senses drones and unmanned surface vessels in real time and reconstructs the topology to ensure continuous communication; the collaborative control system dynamically adjusts the allocation of scenario-based sub-tasks according to the mission progress and emergencies of drones and unmanned surface vessels to shorten response latency.

[0008] Preferably, before the S1 mission is launched, environmental perception and channel preprocessing are performed. The spectrum-environment joint perception unit of the Mesh self-organizing network module is activated, and the built-in broadband spectrum analyzer scans the frequency band to identify the channel occupancy rate and interference signal strength. The channel occupancy rate threshold η is set for the sensors integrated by the UAV and the unmanned surface vessel. According to the characteristics of each sensor, if the channel occupancy rate threshold η is greater than or equal to the channel occupancy rate threshold η, it is considered an interference channel. The dynamic channel planning algorithm performs the following steps based on the environmental detection results: Based on the operating characteristics of the communication equipment and the adaptation requirements of the working environment, a high-frequency channel is allocated to the first communication node with high-altitude operation characteristics, namely the mobile terminal of the Mesh self-organizing network module integrated by the UAV; another low-frequency channel is allocated to the second communication node with water surface operation characteristics, namely the mobile terminal of the Mesh self-organizing network module integrated by the unmanned surface vessel. Based on real-time monitoring of the interference intensity of the current communication channel, when the monitored value reaches or exceeds the preset interference threshold, the channel switching logic is triggered; through the pre-configured backup channel list, it automatically switches to the target channel with interference intensity lower than the threshold.

[0009] Preferably, before receiving a task, the UAV Mesh self-organizing network module mobile terminal one and the UAV Mesh self-organizing network module mobile terminal two initiate a three-level self-check mechanism; hardware layer diagnostic verification includes, but is not limited to, Mesh self-organizing network module RF power, UAV motor speed, UAV propeller thrust, water quality sensor and stratified sampling pump; protocol layer verification includes, but is not limited to, verifying the pollution data protocol compatibility between the Mesh self-organizing network link and the collaborative control system; status layer uploading includes, but is not limited to, the remaining endurance of the UAV and UAV, and the mission payload status.

[0010] Preferably, the Mesh self-organizing network module mobile terminal one uploads the drone status information, including but not limited to type label, polluted area or pollutant coordinates; the Mesh self-organizing network module mobile terminal two uploads the unmanned surface vessel status information, including but not limited to water quality, flow rate, flow direction and pollution level. The coordinates of the polluted area or pollutants guide the unmanned surface vessel (USV) to move towards the center of the polluted area or pollutant coordinates. The parameters uploaded by the USV in real time dynamically correct the pollution type identification weight of the UAV's AI recognition model. When the USV arrives at the sampling point, it performs water stratification sampling and monitors the collected water samples in real time, uploading the data back to the shore-based control system. The collaborative control system generates control frames through the Mavlink protocol to drive the UAV to adjust its trajectory and the USV to control the sampling depth.

[0011] Preferably, when performing tasks, drones and unmanned surface vessels simultaneously perform edge data preprocessing and Mesh packet encapsulation. The onboard computer of the drone is equipped with an AI recognition model that can identify the type of pollution in the RGB images collected by the drone. The mobile terminal of the Mesh self-organizing network module uses the DCT transformation algorithm to perform edge compression with a compression rate of ≥50%, and encapsulates it into a Mesh data packet that includes, but is not limited to, pollution type and timestamp. The unmanned surface vessel (USV) receives the coordinate information of the polluted area transmitted back by the drone via the Mesh self-organizing network module mobile terminal 2. The USV autonomously navigates to the target area and uses water quality sensors and current meters to simultaneously collect water quality parameters, water flow velocity, and flow direction. The water quality parameters are first filtered for outliers using the Z-score algorithm, and then combined with the water flow velocity, a pollution diffusion prediction model is constructed using an LSTM and reinforcement learning fusion algorithm. The model outputs the pollution level and diffusion warning information for a preset time period. The association tags of water quality parameters, water flow velocity, warning information, and drone polluted area coordinate information are encapsulated together into a Mesh data packet containing, but not limited to, coordinate association identifiers and prediction confidence scores, and transmitted back to the shore-based control system.

[0012] Preferably, the data collected by UAVs and unmanned surface vessels (USVs) is encrypted using an air-water collaborative exclusive encryption mechanism before transmission. This mechanism employs the AES-128-CBC algorithm to encrypt the data at the encryption layer. In the key distribution phase, the ECDH protocol is used to dynamically generate exclusive keys for UAVs and USVs, each linked to a pollution task. At the security anchoring level, all data packets embed device digital certificates and task IDs, and the collaborative control system verifies data legitimacy by checking the certificate chain and its association with the task.

[0013] Preferably, both the UAV and the unmanned surface vessel (USV) have built-in edge processing units for the pollution scene. The edge processing units extract key features from the data collected by the UAV and USV, filter out abnormal features, and then transmit the data back in a simplified manner. Specifically, the UAV first extracts the pollution contours from the collected images of the polluted area or pollutants using the Canny edge detection algorithm, and only transmits the contour features and pollution type labels. The USV outputs the pollution level in real time based on parameters including but not limited to pH and dissolved oxygen collected by water quality sensors, and simultaneously filters out abnormal data, transmitting only the level results and abnormal labels.

[0014] Preferably, the data on the range of air pollution collected by drones and the data on the concentration of underwater pollution collected by unmanned surface vessels are integrated to generate a monitoring report that includes, but is not limited to, the type of pollution, its spatial distribution and diffusion prediction, and outputs an emergency warning for the next hour, thus realizing a closed-loop management of the entire process of monitoring-analysis-early warning.

[0015] The beneficial effects of this invention are as follows: Through multi-hop relay and dynamic routing technology in Mesh self-organizing networks, autonomous and reliable communication is achieved in scenarios without fixed base stations, improving communication coverage and resilience; relying on distributed task allocation and real-time data interaction, the collaborative efficiency between devices is significantly improved, response latency is shortened, and the efficient completion of tasks in complex environments is ensured; the mode of unmanned surface vessels (USVs) carrying drones expands the operational radius, enhances adaptability to scenarios such as remote waters and emergency monitoring, and reduces reliance on human intervention; wide-area drone patrols and fixed-point sampling by USVs form complementary "air-water" data, expanding monitoring dimensions and constructing a functionally complementary monitoring closed loop. These effects are achieved based on "a decentralized communication architecture supported by Mesh self-organizing network modules, an air-water device collaborative monitoring mechanism, and distributed task allocation and dynamic adaptation logic."

[0016] The shore-based control system in this invention serves as the core unit for human-computer interaction and global decision-making, and mainly performs the following three functions.

[0017] First, it displays the status information of unmanned surface vessels (USVs), unmanned aerial vehicles (UAVs), and the collaborative control system in real time. Specifically, it includes the position (latitude and longitude), speed, remaining battery power, and mission progress (such as sampling completion rate and aerial photography coverage) of USVs and UAVs, as well as the task allocation status of the collaborative control system (such as sub-task execution nodes and link communication quality). Secondly, structured task instructions are issued to the collaborative control system. These instructions include key parameters such as the operating area (latitude and longitude range), task type (e.g., UAV aerial photography resolution requirements, unmanned surface vessel sampling frequency), and timing constraints (e.g., action trigger time window). Third, it supports dual-mode control logic: in manual control, the operator can directly issue temporary commands through the shore-based interactive interface; in automatic control, the system drives the collaborative control system to autonomously complete the collaborative scheduling of unmanned surface vessels and unmanned aerial vehicles based on preset task parameters and real-time status information. Attached Figure Description

[0018] Figure 1 This is a diagram of the overall architecture of the present invention; Figure 2 This is a schematic diagram of the internal structure of the UAV in this invention; Figure 3 This is a schematic diagram of the internal structure of the unmanned surface vessel in this invention; Figure 4 This is a schematic diagram of the topology of the Mesh self-organizing network module in this invention; Figure 5 This is a flowchart illustrating the task allocation and collaborative workflow of the collaborative control system in this invention. Detailed Implementation

[0019] To make the technical problems solved by the present invention, the technical solutions adopted, and the technical effects achieved clearer, the technical solutions of the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Example: Figures 1 to 5 As shown, a water monitoring system based on a mesh self-organizing network for collaborative operation of unmanned aerial vehicles (UAVs) and unmanned surface vessels (USVs) includes a UAV, an USV, a mesh self-organizing network module, a collaborative control system, and a shore-based control system; the UAVs and USVs communicate with the collaborative control system and with the shore-based control system through the mesh self-organizing network module. Drones are used for wide-area patrols and image acquisition. They collect airspace and wide-area water information through sensors and achieve data interaction by relying on the nodes of the Mesh self-organizing network. The drone integrates, but is not limited to, a Mesh self-organizing network module mobile terminal, an onboard computer, a flight control module, a drone drive module, a drone power actuator, a three-axis gimbal camera, a GPS + magnetic compass, an attitude sensor, a speed sensor, a voltage detection module, and distance detection radars installed in the front, back, left, right, up, and down directions of the drone. The mobile terminal of the Mesh self-organizing network module integrated in the UAV receives control commands from the collaborative control system and uploads the UAV status information. The onboard computer and flight control module form a closed-loop control logic of "perception-decision-execution" based on real-time feedback from the sensors on the UAV. The UAV drive module drives the power actuator to adjust the UAV's flight attitude and speed, and synchronously controls the gimbal camera to collect video and images and transmit them back to the shore control system and the unmanned surface vessel. The unmanned surface vessel (USV) is used for both fixed-point data collection and water area information collection. The USV integrates, but is not limited to, a Mesh self-organizing network module mobile terminal, an industrial control computer, a USV drive module, a USV power actuator, a sampling actuator, a MEMS inertial navigation system, an RTK-GPS autonomous navigation module, a voltage detection module, millimeter-wave radar, and a bow camera, constructing a comprehensive state perception system encompassing "positioning, attitude, power, and environment." The RTK-GPS autonomous navigation module provides centimeter-level latitude and longitude positioning; the MEMS inertial navigation system synchronously outputs roll angle, pitch angle, and the USV's hull speed; the voltage detection module provides real-time feedback on remaining battery power; and millimeter-wave radars in the forward, aft, left, and right directions detect obstacles on the water surface, while the bow camera supplements visual environmental information. The mobile terminal of the Mesh self-organizing network module integrated on the unmanned surface vessel (USV) receives control commands from the collaborative control system. The industrial control computer integrates the obstacle avoidance data from the RTK-GPS autonomous navigation module, MEMS inertial navigation attitude, and millimeter-wave radar to generate drive signals. It controls the USV's power actuators to adjust its speed and heading, and schedules the USV's sampling actuators to complete water sampling operations. The environmental video collected by the camera at the bow is encrypted and transmitted back to the shore control system via the Mesh self-organizing network link. The network topology of the Mesh self-organizing module uses the local end of the Mesh self-organizing module mounted on the shore control system as the core hub, connecting downwards to at least Mesh self-organizing module mobile terminal one, Mesh self-organizing module mobile terminal two, Mesh self-organizing module mobile terminal three, and Mesh self-organizing module mobile terminal four. Each node establishes a multi-hop, redundant, decentralized connection via Mesh wireless links. Specifically, Mesh self-organizing module mobile terminal one is associated with a drone, Mesh self-organizing module mobile terminal two is associated with an unmanned surface vessel (USV), and the two achieve cross-platform interaction through a collaborative control system. Mesh self-organizing module mobile terminals three and four, along with other devices, extend network coverage and functionality. This topology leverages the dynamic self-organizing capability of the Mesh network to ensure stable communication between the shore control system and mobile nodes (USVs, drones), peripherals, and enhances resilience through multi-link redundancy design, ultimately supporting collaborative operations between USVs and drones and the expansion of system functions.

[0021] The workflow of a water monitoring system that combines drones and unmanned surface vessels is as follows: S1 mission released: The shore-based control system calls the electronic chart module to automatically generate a flight path adapted to the collaborative operation of UAVs and unmanned surface vessels based on the monitoring targets and start and end point coordinates input by the user. This path is then integrated into a structured master task that includes, but is not limited to, monitoring content, action commands, and mission start points.

[0022] Before the shore-based control system initiates the water surface pollution monitoring task, the spectrum-environment joint sensing unit of the Mesh self-organizing network module integrated in the local terminal is activated. As the coordination center for spectrum and environmental sensing, it performs dual-dimensional detection of the operating water area. It scans frequency bands through its built-in broadband spectrum analyzer to identify channel occupancy and interference signal strength. It collects parameters such as humidity and salinity by relying on external sensors of UAVs and unmanned surface vessels, and marks areas at risk of signal attenuation based on the occupancy threshold η.

[0023] For example: set the channel occupancy threshold η to be greater than or equal to 60% as an interfering channel, base station clutter power greater than or equal to -70dBm to quickly mark strong interference areas, humidity greater than or equal to 80% as a high humidity environment, and salinity greater than or equal to 35‰ as a sea salt environment.

[0024] The dynamic channel planning algorithm performs the following steps based on the environmental detection results: Based on the operational characteristics of the communication equipment and the adaptation requirements of the working environment, for example, the 5.8GHz high-frequency band channel is allocated to the first communication node with high-altitude operation characteristics, namely the mobile terminal of the Mesh self-organizing network module integrated by the UAV; the 900MHz low-frequency band channel is allocated to the second communication node with surface operation characteristics, namely the mobile terminal of the Mesh self-organizing network module integrated by the UAV; the 5.8GHz high-frequency band is allocated to the UAV, i.e., the aerial node, to reduce the attenuation of signal transmission in the salt spray environment by utilizing its high bandwidth, low latency, and high gain characteristics; the 900MHz low-frequency band channel is allocated to the second communication node with surface operation characteristics, i.e., the mobile terminal of the Mesh self-organizing network module associated with the UAV, to improve the penetration capability of surface clutter by utilizing the strong diffraction characteristics of this band.

[0025] Based on real-time monitoring of the interference intensity of the current communication channel, when the monitored value reaches or exceeds the preset interference threshold, the channel switching logic is triggered; through the pre-configured backup channel list, it automatically switches to the target channel with interference intensity lower than the threshold, ensuring the reliability of subsequent data transmission.

[0026] S2 Task Assignment: Based on the status information and network conditions of UAVs and unmanned surface vessels, the collaborative control system divides the structured overall task into scenario-based sub-tasks: through the local terminal of the Mesh self-organizing network module, using its multi-directional broadcast link, the scenario-based sub-tasks are distributed to Mesh self-organizing network module mobile terminal one, Mesh self-organizing network module mobile terminal two, Mesh self-organizing network module mobile terminal three, Mesh self-organizing network module mobile terminal four and external devices.

[0027] Based on the needs of water surface pollution monitoring, such as daily water surface inspections, and inspections for floating debris and algal bloom pollution, the shore-based control system generates scenario-based sub-tasks. Through the local terminal of the Mesh self-organizing network module, using the Mesh multi-directional broadcast link, the scenario-based sub-tasks are distributed to Mesh self-organizing network module mobile terminal one, Mesh self-organizing network module mobile terminal two, Mesh self-organizing network module mobile terminals three and four (relay / sensing nodes) and external devices, specifying the operating parameters of each node, such as UAV waypoint coordinates and UAV sampling frequency.

[0028] The drone receives aerial target identification tasks via a mobile terminal using a Mesh self-organizing network module. The unmanned surface vessel receives water surface stratification sampling tasks via the Mesh self-organizing network module mobile terminal 2.

[0029] Commands are distributed in multiple directions through the local end of the Mesh self-organizing network module. Using a multi-hop pre-connection mechanism, the reachable links between the local end and each mobile end are automatically detected before the command is issued, including direct links and relay links. Links with a signal strength greater than or equal to -70dBm are given priority to ensure that the air-water platform receives commands synchronously within 100ms, and the task allocation error is less than or equal to 500ms.

[0030] Before receiving a task, both the Mesh self-organizing network module mobile terminal 1 and the Mesh self-organizing network module mobile terminal 2 initiate a three-level self-check mechanism: hardware layer diagnostics and verification of Mesh module RF power, UAV motor speed, UAV propulsion thrust, water quality sensor and stratified sampling pump, etc.; protocol layer verification and validation of the Mesh link and the pollution data protocol compatibility of the collaborative control system; and status layer uploading of device remaining battery life, task payload status, etc.

[0031] S3 Collaborative Monitoring: Unmanned surface vessels (USVs) and unmanned aerial vehicles (UAVs) conduct complementary monitoring operations based on scenario-based task assignments. UAVs collect airspace and wide-area water information through sensors, while USVs collect information at fixed points and in water areas through sensors. Real-time data interaction is achieved through a Mesh self-organizing network module mobile terminal 1 integrated on the UAV and a Mesh self-organizing network module mobile terminal 2 integrated on the USV. The information transmitted back by the UAV provides accurate positioning guidance for USV sampling, and the data collected by the USV on demand feeds back to the UAV, optimizing the pollution identification model on the UAV and forming an air-water collaborative closed loop of "aerial identification - hierarchical sampling - data feedback". Mesh self-organizing module mobile terminal one uploads drone data, including type label, polluted area or pollutant coordinates, etc.; Mesh self-organizing module mobile terminal two uploads unmanned surface vessel data, including water quality, flow velocity, flow direction and pollution level, etc.

[0032] Based on the coordinates of the polluted area transmitted back by the drone, the unmanned surface vessel (USV) is guided to move towards the center of the polluted area, improving the spatial accuracy of water surface sampling. The pollution type identification weight of the drone's AI identification model is dynamically corrected by the water quality parameters uploaded by the USV in real time, improving the accuracy of pollution type identification. When the USV arrives at the sampling point, it performs stratified water sampling and conducts real-time detection of the collected water samples, uploading the data back to the shore-based control system.

[0033] The collaborative system generates control frames via the Mavlink protocol to drive the UAV to adjust its trajectory and the unmanned surface vessel to control its sampling depth.

[0034] S4 Dynamic Adaptation and Optimization: The Mesh self-organizing network module senses drones and unmanned surface vessels in real time and reconstructs the topology to ensure continuous communication; the collaborative control system dynamically adjusts the allocation of scenario-based sub-tasks according to the mission progress and emergencies of drones and unmanned surface vessels to shorten response latency.

[0035] When drones and unmanned surface vessels are performing missions, they simultaneously carry out edge data preprocessing and mesh packet encapsulation. The onboard computer of the drone is equipped with an AI recognition model that can identify the type of pollution in the RGB images collected by the drone. The mobile terminal of the Mesh self-organizing network module uses DCT transformation algorithm to perform edge compression with a compression rate of ≥50% and encapsulates it into a Mesh data packet containing pollution type and timestamp. The unmanned surface vessel (USV) receives the coordinate information of the polluted area transmitted back by the drone via the Mesh self-organizing network module mobile terminal 2. The USV autonomously navigates to the target area and uses water quality sensors and current meters to simultaneously collect water quality parameters, water flow velocity, and flow direction. The water quality parameters are first filtered for outliers using the Z-score algorithm, and then combined with the water flow velocity, a pollution diffusion prediction model is constructed using an LSTM and reinforcement learning fusion algorithm. The model outputs the pollution level and diffusion warning information for a preset time period. The association tags of water quality parameters, water flow velocity, warning information, and drone polluted area coordinate information are encapsulated together into a Mesh data packet containing, but not limited to, coordinate association identifiers and prediction confidence scores, and transmitted back to the shore-based control system.

[0036] Before transmission, the data collected by UAVs and unmanned surface vessels (USVs) is encrypted using a dedicated air-water collaborative encryption mechanism. This mechanism employs the AES-128-CBC algorithm at the encryption layer. In the key distribution phase, the ECDH protocol dynamically generates dedicated keys for UAVs and USVs, each linked to a specific pollution task. At the security anchoring level, all data packets embed device digital certificates and task IDs, and the collaborative control system verifies data legitimacy by checking the certificate chain and its association with the task.

[0037] Both drones and unmanned surface vessels (USVs) have built-in edge processing units for pollution scenarios. These units extract key features from the data collected by the drones and USVs, filter out abnormal features, and then transmit the data in a streamlined manner. Specifically, the drones first extract the pollution contours from the images of the polluted areas or pollutants using the Canny edge detection algorithm, and then transmit only the contour features and pollution type markers. The USVs, based on parameters including but not limited to pH and dissolved oxygen collected by water quality sensors, output the pollution level in real time through a pollution monitoring algorithm, simultaneously filter out abnormal data, and transmit only the level results and abnormal markers.

[0038] The data on the range of air pollution collected by drones and the concentration of underwater pollution collected by unmanned surface vessels are integrated to generate a monitoring report that includes, but is not limited to, pollution type, spatial distribution and diffusion prediction, and outputs an emergency warning for the next hour, realizing a closed-loop management of the entire process of monitoring-analysis-early warning.

[0039] This invention dynamically optimizes task execution based on preset monitoring performance indicators. When a new device is added, its Mesh self-organizing network module can automatically discover surrounding drones and unmanned surface vessels and complete topology integration, thereby expanding the monitoring dimensions.

[0040] This collaborative control system, designed for water surface pollution monitoring, constructs a comprehensive technical system encompassing "air-water equipment collaborative sensing—Mesh self-organizing network communication assurance—edge intelligent processing—data fusion closed loop—dynamic optimization and expansion." Through air-water collaboration involving UAV aerial identification and unmanned surface vessel underwater sampling, and relying on a mesh self-organizing network for interference-resistant communication and plug-and-play device expansion, combined with edge processing to streamline data and encryption mechanisms to ensure security, it ultimately forms a "monitoring-analysis-early warning" closed loop through air-water data fusion, and dynamically optimizes task execution based on performance indicators. Its core protection scope focuses on "the air-water collaborative pollution monitoring mechanism, the communication and expansion logic supported by the mesh network, and the data-driven dynamic optimization closed loop." Regardless of specific parameter adjustments, algorithm replacements, or device expansions, as long as the above core mechanisms are followed, they all fall within the protection scope of this invention.

Claims

1. A water area monitoring system based on a Mesh ad hoc network for cooperative operation of a UAV and an unmanned ship, characterized by, It comprises a UAV, an unmanned ship, a Mesh self-organizing network module, a cooperative control system and a shore control system; the UAV and the unmanned ship are connected to the Mesh self-organizing network through their integrated mobile communication modules and communicate with the cooperative control system and the shore control system; The UAV is used for wide-area patrol and image acquisition, and collects airspace and wide-area water information through sensors and realizes data interaction by relying on nodes of the Mesh self-organizing network; The UAV is integrated with a Mesh self-organizing network module mobile terminal one, an onboard computer, a flight control module, a UAV driving module, a UAV power execution mechanism, a three-axis gimbal camera, a GPS + magnetic compass, an attitude sensor, a speed sensor, a voltage detection module and distance detection radars installed in six directions of front, rear, left, right, top and bottom of the UAV; The integrated Mesh self-organizing network module mobile terminal one of the UAV receives control commands of the cooperative control system and uploads UAV state information, the onboard computer and the flight control module form a perception-decision-execution closed-loop control logic based on real-time feedback of sensors on the UAV, the UAV driving module drives the power execution mechanism to adjust the flight attitude and speed of the UAV, and the gimbal camera is synchronously controlled to collect videos and pictures and transmit them back to the shore control system and the unmanned ship; The unmanned ship is used for fixed-point collection and water area information collection; the unmanned ship is integrated with a Mesh self-organizing network module mobile terminal two, an industrial computer, an unmanned ship driving module, an unmanned ship power execution mechanism, a sampling execution mechanism, a MEMS inertial navigation module, an RTK-GPS autonomous navigation module, a voltage detection module, a millimeter wave radar and a bow camera, and a full-dimensional state perception system of positioning-attitude-power-environment is constructed; the RTK-GPS autonomous navigation module provides centimeter-level latitude and longitude positioning; the MEMS inertial navigation module synchronously outputs roll angle, pitch angle and body movement speed of the unmanned ship; the voltage detection module feeds back residual power in real time; the millimeter wave radars in four directions of front, rear, left and right detect water surface obstacles, and the bow camera supplements visual environmental information; The integrated Mesh self-organizing network module mobile terminal two of the unmanned ship receives control commands of the cooperative control system, the industrial computer fuses data of the RTK-GPS autonomous navigation module, the MEMS inertial navigation attitude and the millimeter wave radar, generates a driving signal, controls the unmanned ship power execution mechanism to adjust the speed and direction, and schedules the sampling execution mechanism of the unmanned ship to complete water sampling work; the environmental video collected by the bow camera is transmitted back to the shore control system through the link of the Mesh self-organizing network. The Mesh self-organizing network module adopts a decentralized network topology structure, with the local end carried by the onshore control system as the core hub, dynamically connecting at least four mobile end nodes through multi-hop and redundant wireless links to form a self-organizing communication network resistant to destruction. Among them, the local end of the Mesh self-organizing network module carried by the onshore control system is the core hub, connecting down at least Mesh self-organizing network module mobile end one, Mesh self-organizing network module mobile end two, Mesh self-organizing network module mobile end three, and Mesh self-organizing network module mobile end four, and each node is connected through multi-hop and redundant decentralized connections by Mesh wireless links. Among them, the Mesh self-organizing network module mobile end one is associated with a UAV, and the Mesh self-organizing network module mobile end two is associated with a UAV, both of which realize cross-platform interaction through a cooperative control system, and the Mesh self-organizing network module mobile end three and the Mesh self-organizing network module mobile end four and other devices extend network coverage and functions. 2.The water area monitoring system based on the cooperation between the unmanned aerial vehicle and the unmanned ship according to claim 1, wherein, The working process of the water area monitoring system for UAV and UAV cooperative operation is as follows: S1 task publishing: The onshore control system calls the electronic chart module, generates a route suitable for UAV and UAV cooperative operation according to the user input monitoring target and start and end point coordinates, and integrates it into a structured total task containing but not limited to monitoring content, action instructions and task starting point; S2 task allocation: The cooperative control system divides the structured total task into a scenario-based sub-task based on the state information and network conditions of the UAV and the UAV: through the local end of the Mesh self-organizing network module, using its multi-directional broadcast link, the scenario-based sub-task is distributed to the Mesh self-organizing network module mobile end one, the Mesh self-organizing network module mobile end two, the Mesh self-organizing network module mobile end three and the Mesh self-organizing network module mobile end four and external devices; S3 cooperative monitoring: The UAV and the UAV carry out functionally complementary monitoring operations according to the scenario-based sub-task; the UAV collects airspace and wide-area water information through sensors, and the UAV collects point sampling and water information through sensors; real-time data interaction is realized through the Mesh self-organizing network module mobile end one integrated on the UAV and the Mesh self-organizing network module mobile end two integrated on the UAV; the information returned by the UAV provides accurate positioning guidance for the sampling of the UAV, and the data collected by the UAV on demand feeds back to the UAV, optimizing the pollution identification model on the UAV, forming an air-water cooperative closed loop of "air identification-layered sampling-data mutual feedback"; S4 dynamic adaptation and optimization: The Mesh self-organizing network module senses the UAV and the UAV in real time and reconstructs the topology to ensure continuous communication; the cooperative control system dynamically adjusts the allocation of the scenario-based sub-task according to the task progress and sudden conditions of the UAV and the UAV, and shortens the response delay. 3.The water area monitoring system based on the cooperation between the unmanned aerial vehicle and the unmanned ship according to claim 2, wherein: The environment sensing and channel preprocessing are performed before the S1 task is issued, the spectrum-environment joint sensing unit on the local end of the Mesh ad hoc network module is activated, the frequency band is scanned by the built-in broadband spectrum analyzer, the channel occupancy rate and the interference signal strength are identified, the channel occupancy rate threshold η of the integrated sensor of the unmanned aerial vehicle and the unmanned ship is set, according to the characteristics of each sensor, greater than or equal to the channel occupancy rate threshold η, it is an interference channel; The dynamic channel planning algorithm performs the following steps based on the environment detection results: According to the operation characteristics of the communication equipment and the adaptation requirements of the working environment, a high-frequency channel is allocated to the first communication node with high-altitude operation characteristics, that is, the Mesh ad hoc network module mobile end one integrated with the unmanned aerial vehicle; another low-frequency channel is allocated to the second communication node with water surface operation characteristics, that is, the Mesh ad hoc network module mobile end two integrated with the unmanned ship; According to the real-time monitoring of the interference strength of the current communication channel, when the monitoring value reaches or exceeds the preset interference threshold, the channel switching logic is triggered; through the preconfigured backup channel list, the target channel with interference strength lower than the threshold is automatically switched to.

4. The water area monitoring system based on the cooperation between the unmanned aerial vehicle and the unmanned ship according to claim 1 or 2, characterized in that The unmanned aerial vehicle Mesh ad hoc network module mobile end one and the unmanned ship Mesh ad hoc network module mobile end two start a three-level self-checking mechanism before receiving the task; the hardware layer diagnosis verification includes but is not limited to the Mesh ad hoc network module radio frequency power, the unmanned aerial vehicle motor speed, the unmanned ship propeller thrust, the water quality sensor and the layered sampling pump; the protocol layer verification includes but is not limited to checking the compatibility of the Mesh ad hoc network link and the pollution data protocol of the cooperative control system; the state layer upload includes but is not limited to the remaining endurance of the unmanned aerial vehicle and the unmanned ship, the task load state.

5. The water area monitoring system based on the cooperation between the unmanned aerial vehicle and the unmanned ship according to claim 1 or 2, characterized in that The Mesh ad hoc network module mobile end one uploads the unmanned aerial vehicle state information, including but not limited to type label, pollution area or pollution coordinates; the Mesh ad hoc network module mobile end two uploads the unmanned ship state information, including but not limited to water quality, flow rate, flow direction and pollution level; Among them, the pollution area or the pollution coordinates guide the unmanned ship to move towards the center of the pollution area or the pollution coordinates, the parameters uploaded by the unmanned ship in real time dynamically correct the pollution type recognition weight of the unmanned aerial vehicle AI recognition model; when the unmanned ship reaches the sampling point to perform water layer sampling, the collected water sample is detected in real time and uploaded to the shore end control system; the cooperative control system generates a control frame through the Mavlink protocol to drive the unmanned aerial vehicle to adjust the trajectory and the unmanned ship to control the sampling depth. 6.The water area monitoring system based on the cooperation between the unmanned aerial vehicle and the unmanned ship according to claim 1 or 2, characterized in that The unmanned aerial vehicle and the unmanned ship perform edge data preprocessing and Mesh packet encapsulation simultaneously when performing the task; The AI recognition model is deployed on the airborne computer of the unmanned aerial vehicle, which can perform pollution type recognition on the RGB image collected by the unmanned aerial vehicle, and the Mesh ad hoc network module mobile end one performs edge compression using the DCT transform algorithm, the compression rate is ≥50%, and is encapsulated as a Mesh data packet containing but not limited to pollution type and timestamp. The unmanned ship receives the pollution area coordinate information returned by the unmanned aerial vehicle through the Mesh ad hoc network module mobile terminal two, and autonomously sails to the target area. The water quality sensor and flow meter are used to synchronously collect water quality parameters, water flow rate and flow direction. The water quality parameters are first screened for outliers by Z-score algorithm, and then combined with the water flow rate, a pollution diffusion prediction model is constructed by the LSTM and reinforcement learning fusion algorithm, and the pollution level and future preset period diffusion warning information are output. The water quality parameters, water flow rate, warning information and unmanned aerial vehicle pollution area coordinate information are associated with markers, and are jointly packaged as Mesh data packets containing but not limited to coordinate association markers and prediction confidence to return to the shore control system.

7. The water area monitoring system based on the cooperation between the unmanned aerial vehicle and the unmanned ship according to claim 1 or 2, characterized in that The data collected by the unmanned aerial vehicle and the unmanned ship is enabled with an air-water collaborative encryption mechanism before transmission. In the encryption layer, the AES-128-CBC algorithm is used to encrypt the data. In the key distribution link, the ECDH protocol is used to dynamically generate a special key bound to the pollution task for the unmanned aerial vehicle and the unmanned ship. In the security anchoring layer, all data packets are embedded with device digital certificates and task IDs, and the collaborative control system verifies the legality of the data by verifying the certificate chain and task association. 8.The water area monitoring system based on the cooperation between the unmanned aerial vehicle and the unmanned ship according to claim 1 or 2, characterized in that The unmanned aerial vehicle and the unmanned ship are both equipped with edge processing units for pollution scenes. The edge processing units extract key features from the data collected by the unmanned aerial vehicle and the unmanned ship, and simplify the return after abnormal feature screening. Specifically, the unmanned aerial vehicle extracts the pollution contour from the collected pollution area or pollutant image through the corresponding algorithm, and only returns the contour feature and pollution type marker. The unmanned ship collects parameters such as pH and dissolved oxygen through the water quality sensor, and outputs the pollution level in real time through the pollution monitoring algorithm, synchronously screens abnormal data, and only returns the level result and abnormal marker.

9. The water area monitoring system based on the cooperation between the unmanned aerial vehicle and the unmanned ship according to claim 1 or 2, characterized in that The data of the aerial pollution range collected by the unmanned aerial vehicle and the underwater pollution concentration collected by the unmanned ship are integrated to generate a monitoring report containing but not limited to pollution type, spatial distribution and diffusion prediction, and output an emergency warning for the next 1 hour, realizing the whole-process closed-loop management of monitoring-analysis-warning.

Images