Accurate algae prevention and control system based on air-water collaborative intelligent decision and application thereof

Abstract

The invention discloses an accurate algae prevention and control system based on air-water collaborative intelligent decision and application thereof. According to the method, 1) an algae prevention and control zero-delay response system is established: a minute-level quick response system is constructed by fusing a deep forest inversion model, a GNSS-RTK high-precision positioning module, an algal inhibition bacterium activation and addition integrated device and a dynamic decision model based on a fuzzy rule, and the problem of prevention and control response hysteresis is solved; 2) a data-driven accurate adding system: through dosage dynamic optimization and activator structure innovation, microbial inoculum adding depth self-adaption and concentration field homogenization are realized, and insufficient pesticide effect, waste and algal inhibition efficiency attenuation caused by space-time mismatch are eliminated; and 3) a full-intelligent operation system: depending on a multi-mode obstacle avoidance and remote sensing-execution-evaluation double-closed-loop architecture, achieving unattended operation of a detection-decision-policy implementation full link, and breaking through the scale deployment bottleneck and environmental adaptation limitation of traditional man-machine coupling.

Description

Technical Field

[0001] This invention relates to the field of aquatic algae control technology, specifically to a precision algae control system based on air-water collaborative intelligent decision-making and its application. Background Technology

[0002] The current technological maturity of aquatic microbial agent dosing devices exhibits an imbalance: mechanization and automation are relatively complete, while intelligent and autonomous capabilities lag far behind. Although existing patent literature has achieved automated replacement of processes such as spraying, mixing, and conveying at the mechanical execution level (e.g., CN210251895U, CN213294735U, CN112850816A), it remains at a basic stage of automation in the three core dimensions of intelligent decision-making, precise policy implementation, and unmanned operation and maintenance, failing to reach the level of autonomous capability driven by artificial intelligence.

[0003] Analysis reveals that existing technologies generally suffer from the following problems: 1. Response Lag. Under suitable conditions (warm air, strong light, high nutrients), the algal reproduction cycle can be as short as 2-30 hours, exhibiting logarithmic growth. Traditional algal control often takes several days from sampling and testing to analysis and judgment and the implementation of measures, frequently missing the control window before the critical point of algal outbreak.

[0004] 2. Blindness in policy implementation. Traditional automatic dosing equipment (such as fixed pontoons and uniform spraying boats) mostly adopts open-loop dosing with preset programs. The dosing time, dosage and location depend on experience judgment and manual preset, ignoring the non-uniformity of algal spatial distribution, resulting in low treatment efficiency, insufficient and excessive local dosage of agents, and uncontrolled spread of algal blooms.

[0005] 3. Operation and Maintenance Dependence. Traditional treatment equipment often focuses on optimizing single mechanical structures, and is mostly in remote or semi-automatic mode. Obstacle avoidance, trajectory planning, and dosage decision-making all require deep human intervention. The system lacks autonomous decision-making and fault tolerance capabilities, making unmanned operation impossible. This results in high operation and maintenance costs and makes it difficult to deploy in remote or dangerous waters. Some treatment methods have introduced unmanned surface vessel platforms, but have not achieved a closed-loop architecture of wide-area aerial perception - precise execution on the water surface - collaborative decision-making in the cloud. The dosing strategy is still based on threshold judgment rather than predictive models, which limits the system's robustness and environmental adaptability. Summary of the Invention

[0006] To address the aforementioned technical problems and shortcomings in this field, the present invention provides a precise algae control system based on air-water collaborative intelligent decision-making and its application.

[0007] To address the issue of delayed response, this invention constructs a "zero-delay" response system for algae control: real-time onboard inversion of hyperspectral data from UAVs eliminates post-processing wait times; high-precision positioning modules are used in both UAVs and unmanned vessels to shorten the time from identification to implementation; algae-inhibiting bacteria activation and application are integrated to shorten the traditional step-by-step time sequence; and a dynamic decision-making model is used for dosage application, which accelerates computation time and overcomes response bottlenecks through preset fuzzy rules.

[0008] To address the problem of blind dosing, this invention constructs a data-driven precision dosing system: the system relies on real-time sensor data from unmanned vessels and a collaborative decision-making platform to dynamically optimize the dosage, avoiding localized under- or over-dosing of the microbial agent; it also optimizes the microbial agent activator and adjusts the dosing depth to improve the utilization rate of the microbial agent.

[0009] To address the issue of operational dependence, this invention constructs a fully intelligent operation system: the unmanned vessel is equipped with a multimodal autonomous obstacle avoidance system to ensure safe unmanned navigation in complex waters; the system forms a dual closed-loop control architecture of remote sensing inversion, trajectory planning, in-situ deployment, and effect feedback, realizing deep integration of UAV hyperspectral feedforward guidance and UAV real-time feedback, replacing threshold triggering with predictive decision-making, and requiring no human intervention throughout the entire process.

[0010] The specific technical solution is as follows: In a first aspect, the present invention provides a precision algae control system based on air-water collaborative intelligent decision-making, including a UAV hyperspectral water quality inversion and hotspot marking system, an unmanned vessel precision drug delivery and autonomous trajectory control system, and a data platform collaborative decision-making system; The UAV hyperspectral water quality inversion and hotspot labeling system includes a UAV and a hyperspectral water quality inversion system and a first hotspot labeling system mounted on the UAV. The working process of the UAV hyperspectral water quality inversion and hotspot labeling system includes: the UAV inspects a designated water area along a preset route, uses the hyperspectral water quality inversion system to invert chlorophyll a concentration in real time, and when the inverted chlorophyll a concentration exceeds a preset threshold, the first hotspot labeling system records the coordinates of the risk point, generates a first four-element label containing coordinates, time, index and corresponding concentration value of the index, and transmits it to the data platform collaborative decision-making system. The unmanned surface vessel (USV) precision drug delivery and autonomous navigation control system includes the USV, a microbial agent activation and delivery device mounted on the USV, a fluorescent chlorophyll a sensor, and a second hotspot marking system. The working process of the USV precision drug delivery and autonomous navigation control system includes: the USV navigates along a planned trajectory, the fluorescent chlorophyll a sensor synchronously detects the chlorophyll a concentration in the water area, and when the detected chlorophyll a concentration is higher than a preset limit, the second hotspot marking system records the coordinates of the point exceeding the limit, generates a second four-element marker including coordinates, time, index, and concentration value, and transmits it to the data platform collaborative decision-making system. The data platform collaborative decision-making system includes a trajectory dynamic planning system and a dosage dynamic decision-making model, which are used to: plan the unmanned vessel's bacterial delivery trajectory based on the first four element markers after the UAV completes its inspection; and dynamically adjust the speed of the unmanned vessel and the dosage of the bacterial agent activation and delivery device according to concentration levels based on the second four element markers.

[0011] Furthermore, the hyperspectral water quality inversion system includes a hyperspectral water quality imager and an edge computing module.

[0012] Furthermore, the hyperspectral water quality imager has a wavelength range of 400-1000nm, a spectral resolution better than 5nm, and a spatial resolution better than 0.2m.

[0013] Furthermore, the edge computing module incorporates a lightweight deep forest inversion model. After fractional differential preprocessing and stepwise discriminant analysis feature screening of 300 bands in the 400-1000nm range, it calculates the chlorophyll a concentration in water areas in real time with a spatial resolution better than 0.2m and an average relative error of <10%. The airborne terminal processes the data in real time with a latency of <60s.

[0014] Furthermore, the first and second hotspot marking systems employ GNSS-RTK (Global Navigation Satellite System - Real-Time Kinematic) positioning modules, supporting GPS L1 / L2 and BeiDou B1 / B2 dual-frequency carrier phase differential positioning. They access the local continuously operating reference station (CORS) service via a 4G network, receive differential correction data in RTCM 3.2 format in real time, output at a frequency of 10Hz, and achieve horizontal positioning accuracy better than ±1.5cm and elevation accuracy better than ±3cm.

[0015] Furthermore, the unmanned vessel is also equipped with a multimodal autonomous obstacle avoidance system.

[0016] Furthermore, the multimodal autonomous obstacle avoidance system integrates millimeter-wave radar and forward-looking camera to build dynamic maps in real time during navigation, enabling centimeter-level avoidance of obstacles (such as aquatic plants, fishing nets, bridge piers, etc.).

[0017] Furthermore, the microbial agent activation and dosing device includes a microbial agent storage chamber, a screw conveyor, a microbial agent activator, and a flow booster.

[0018] Furthermore, the microbial agent storage compartment has a volume of 50L and is used to store the dry powder of the microbial agent.

[0019] Furthermore, a screw conveyor is placed at the bottom of the microbial agent storage chamber to transport the microbial agent dry powder to the microbial agent activator at a constant speed, with a power of 0.12kW.

[0020] Furthermore, the microbial agent activator is L-shaped to ensure that the microbial agent is added at a depth of 0.5m below the water surface.

[0021] Furthermore, the propeller is placed inside the bacterial agent activator, with a power of 0.37kW. It utilizes turbulent propeller flow to achieve efficient mixing of the bacterial agent and water, with a mixing efficiency >95% and a bacterial agent concentration gradient <0.5g / m³. 3 ·m, to achieve effective addition of microbial agents.

[0022] Furthermore, the bacterial agent is an algae-inhibiting bacterium, which is an eco-friendly bacterial agent and does not have adverse effects on aquatic plants and animals.

[0023] Furthermore, The trajectory dynamic planning system identifies points in the water area where the chlorophyll a concentration exceeds the threshold by analyzing the first four elements. It then uses a clustering algorithm to cluster these points to form key areas where unmanned surface vessels (USVs) need to deliver pesticides. Finally, it combines the distribution of key areas, the USV inspection routes, and the distribution of river systems to generate the optimal navigation path for the USV based on the simulated annealing algorithm. The system also uses a GIS map to visualize the navigation path.

[0024] Furthermore, the dose dynamic decision model takes chlorophyll a concentration and its rate of change as input, and formulates PID parameters online through preset fuzzy rules to dynamically correct the ship speed, achieving adaptive control with a response time of <5 seconds and overshoot of <10%, without the need to establish an accurate mathematical model of algal growth dynamics.

[0025] Furthermore, the data platform's collaborative decision-making system dynamically adjusts boat speed and bacterial dosage according to hydrological conditions and chlorophyll a concentration levels. For example: chlorophyll a < 10 μg / L, boat speed = 1.5 m / s; chlorophyll 10 μg / L ≤ chlorophyll a < 20 μg / L, boat speed = 1.0 m / s; chlorophyll a ≥ 20 μg / L, boat speed = 0.5 m / s (actual boat speed can be adjusted according to hydrological conditions); the bacterial dosage is simultaneously set at 0 g / m³. 3 5g / m 3 10g / m 3 Level 3 response.

[0026] Secondly, the present invention provides the application of the air-water collaborative intelligent decision-making-based algae control system described in the first aspect for the precise control of algae outbreaks in aquatic bodies.

[0027] Thirdly, the present invention provides a method for precise control of algal outbreaks based on air-water collaborative intelligent decision-making, using the algal precise control system based on air-water collaborative intelligent decision-making described in the first aspect; The method includes: The working process of the UAV hyperspectral water quality inversion and hotspot marking system includes: the UAV inspects the designated water area along the preset route, uses the hyperspectral water quality inversion system to invert the chlorophyll a concentration in real time, and when the inverted chlorophyll a concentration exceeds the preset threshold, the first hotspot marking system records the coordinates of the risk point, generates the first four-element marker containing coordinates, time, index and the corresponding concentration value of the index, and transmits it to the data platform collaborative decision-making system. The data platform collaborative decision-making system plans the unmanned vessel's bacterial delivery trajectory based on the first four elements marker after the drone completes its inspection. The unmanned vessel navigates along the planned trajectory, and the fluorescent chlorophyll a sensor simultaneously detects the chlorophyll a concentration in the water. When the detected chlorophyll a concentration is higher than the preset limit, the second hot spot marking system records the coordinates of the point exceeding the limit, generates a second four-element marker including coordinates, time, index and concentration value, and transmits it to the data platform collaborative decision-making system. The data platform collaborative decision-making system dynamically adjusts the speed of the unmanned vessel and the dosage of the bacterial agent activation and dosing device based on the concentration classification of the second and fourth element labels.

[0028] Compared with the prior art, the beneficial effects of this invention are as follows: 1) This invention establishes a "zero-delay" response system for algae control, reducing the response time from several hours in the traditional model to less than 5 minutes: the deep forest (DF) inversion model enables real-time processing on the airborne end in <60 seconds; both UAVs and unmanned surface vessels adopt GNSS-RTK modules (horizontal accuracy ±1.5cm, elevation ±3cm) to complete a rapid closed loop of hotspot locking, trajectory planning, and precise arrival; the integrated design of algae-inhibiting bacteria activation and application eliminates step-by-step time delays; the dynamic decision model tunes PID parameters online through fuzzy rules, achieving a response time of <5 seconds without the need for an accurate algae dynamics model.

[0029] 2) This invention establishes a data-driven precision dosing system: the decision-making system dynamically optimizes the dosing amount based on the real-time chlorophyll a concentration of the unmanned vessel, avoiding local under-dosing or over-dosing; the L-shaped activator achieves precise delivery 0.5m below the water surface (the active layer of algal photosynthesis), significantly improving the utilization rate of the bacterial agent.

[0030] 3) This invention establishes a fully intelligent operation system: the multimodal autonomous obstacle avoidance system integrates millimeter-wave radar and vision to achieve unattended safe navigation in complex waters; the dual closed-loop architecture of remote sensing inversion-track planning-in-situ deployment-effect evaluation replaces threshold triggering with feedforward guidance, requiring no manual intervention throughout the process, and greatly improving the system's robustness and environmental adaptability. Attached Figure Description

[0031] Figure 1 This is a diagram illustrating the overall architecture of a precision algae control system based on air-water collaborative intelligent decision-making, as described in this invention.

[0032] Figure 2 This is a schematic diagram of the structure of a microbial agent activation and dosing device according to the present invention. Detailed Implementation

[0033] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention.

[0034] See Figure 1 A precision algae control system based on air-water collaborative intelligent decision-making includes a UAV hyperspectral water quality inversion and hotspot marking system, an unmanned surface vessel precision drug delivery and autonomous trajectory control system, and a data platform collaborative decision-making system.

[0035] The UAV-based hyperspectral water quality inversion and hotspot labeling system comprises a UAV and an onboard hyperspectral water quality inversion system and a primary hotspot labeling system. The hyperspectral water quality inversion system includes a hyperspectral water quality imager and an edge computing module. The hyperspectral water quality imager has a wavelength range of 400-1000 nm, a spectral resolution better than 5 nm, and a spatial resolution better than 0.2 m. The edge computing module incorporates a lightweight deep forest inversion model. After fractional-order differential preprocessing and stepwise discriminant analysis feature selection of 300 bands within the 400-1000 nm range, it calculates the chlorophyll a concentration in the water area in real time, achieving a spatial resolution better than 0.2 m and an average relative error of <10%. The onboard processing is real-time, with a latency of <60 s. The first hotspot marking system uses a GNSS-RTK positioning module, supports GPS L1 / L2 and Beidou B1 / B2 dual-frequency carrier phase differential positioning, accesses local continuously operating reference station services through 4G network, receives differential correction data in RTCM 3.2 format in real time, outputs a frequency of 10Hz, and has a horizontal positioning accuracy better than ±1.5cm and an elevation accuracy better than ±3cm.

[0036] The unmanned surface vessel (USV) precision drug delivery and autonomous navigation control system includes the USV and a multimodal autonomous obstacle avoidance system, a fungicide activation and dispensing device, a fluorescent chlorophyll-a sensor, and a second hotspot marking system mounted on the USV. The multimodal autonomous obstacle avoidance system integrates millimeter-wave radar and a forward-looking camera to build a dynamic map in real time during navigation, achieving centimeter-level obstacle avoidance. See also Figure 2The microbial agent activation and dosing device includes a microbial agent storage tank (1), a screw conveyor (2), a microbial agent activator (3), and a propeller (4). The microbial agent storage tank (1) has a volume of 50L and is used to store the dry powder of the microbial agent. The screw conveyor (2) is placed at the bottom of the microbial agent storage tank (1) and conveys the dry powder of the microbial agent to the microbial agent activator (3) at a constant speed. Its power is 0.12kW. The microbial agent activator (3) is L-shaped to ensure that the microbial agent is added at a depth of 0.5m below the water surface. The propeller (4) is placed inside the microbial agent activator (3). Its power is 0.37kW. It uses the turbulent flow of the propeller to achieve efficient mixing of the microbial agent and water, with a mixing efficiency >95% and a microbial agent concentration gradient <0.5g / m³. 3 The system achieves effective dosing of the microbial agent. The agent is an algae-inhibiting bacterium, an eco-friendly agent that does not adversely affect aquatic plants and animals. The second hotspot marking system uses a GNSS-RTK positioning module, supporting GPS L1 / L2 and BeiDou B1 / B2 dual-frequency carrier phase differential positioning. It accesses the local continuously operating reference station service via a 4G network, receiving differential correction data in RTCM 3.2 format in real time. The output frequency is 10Hz, with a horizontal positioning accuracy better than ±1.5cm and an elevation accuracy better than ±3cm.

[0037] The data platform collaborative decision-making system includes a trajectory dynamic planning system and a dosage dynamic decision-making model. It is used to: plan the unmanned surface vessel (USV) bacterial dispensing trajectory based on the first four element markers after the UAV completes its inspection; and dynamically adjust the USV's speed and the dosage of the bacterial agent activation and dispensing device according to concentration levels based on the second four element markers. The trajectory dynamic planning system identifies points in the water area where chlorophyll a concentration exceeds a threshold by analyzing the first four element markers. It then uses a clustering algorithm to cluster these points to form key areas requiring USV dispensing. Finally, combining the distribution of key areas, the UAV inspection route, and the river system distribution, it generates the optimal navigation path for the USV based on a simulated annealing algorithm and visualizes the navigation path using a GIS map. The dosage dynamic decision-making model takes chlorophyll a concentration and its rate of change as input. It uses preset fuzzy rules to online formulate PID parameters, dynamically corrects the vessel speed, and achieves adaptive control with a response time of <5 seconds and overshoot of <10%, without requiring a precise mathematical model of algal growth dynamics. The data platform's collaborative decision-making system dynamically adjusts boat speed and bacterial dosage based on hydrological conditions and chlorophyll a concentration levels. For example: chlorophyll a < 10 μg / L, boat speed = 1.5 m / s; chlorophyll 10 μg / L ≤ chlorophyll a < 20 μg / L, boat speed = 1.0 m / s; chlorophyll a ≥ 20 μg / L, boat speed = 0.5 m / s (actual boat speed can be adjusted according to hydrological conditions); the bacterial dosage is simultaneously set at 0 g / m³. 3 5g / m 3 10g / m 3 Level 3 response.

[0038] The aforementioned air-water collaborative intelligent decision-making-based algae control system can be used for the precise control of algae outbreaks in aquatic bodies.

[0039] Combination Figure 1 A method for precise control of algal blooms based on air-water collaborative intelligent decision-making, employing the aforementioned precise algal control system based on air-water collaborative intelligent decision-making, comprising: The working process of the UAV hyperspectral water quality inversion and hotspot labeling system includes: the UAV inspects the designated water area along the preset route, uses the hyperspectral water quality inversion system to invert the chlorophyll a concentration in real time, and when the inverted chlorophyll a concentration exceeds the preset threshold (e.g., chlorophyll a>20μg / L), the first hotspot labeling system records the coordinates of the risk point, generates the first four-element label containing coordinates, time, index and corresponding concentration value of the index, and transmits it to the data platform collaborative decision-making system via NB-IoT network (Narrow Band Internet of Things). The data platform collaborative decision-making system plans the unmanned vessel's bacterial delivery trajectory based on the first four elements markers after the drone completes its inspection. The unmanned vessel navigates along the planned trajectory, and the fluorescent chlorophyll a sensor simultaneously detects the chlorophyll a concentration in the water. When the detected chlorophyll a concentration is higher than the preset limit (e.g., chlorophyll a ≥ 10 μg / L), the second hot spot marking system records the coordinates of the point exceeding the limit, generates a second four-element marker including coordinates, time, index and concentration value, and transmits it to the data platform collaborative decision-making system. The data platform's collaborative decision-making system dynamically adjusts boat speed and bacterial dosage based on hydrological conditions and chlorophyll a concentration levels. For example: chlorophyll a < 10 μg / L, boat speed = 1.5 m / s; chlorophyll 10 μg / L ≤ chlorophyll a < 20 μg / L, boat speed = 1.0 m / s; chlorophyll a ≥ 20 μg / L, boat speed = 0.5 m / s (actual boat speed can be adjusted according to hydrological conditions); the bacterial dosage is simultaneously set at 0 g / m³. 3 5g / m 3 10g / m 3 Level 3 response. The unmanned vessel completed its planned trajectory and returned along the same route.

[0040] In summary, the present invention has the following characteristics: A zero-delay response system for algae control: This system integrates a deep forest (DF) inversion model, a GNSS-RTK high-precision positioning module, an integrated device for activating and adding algae-inhibiting bacteria, and a dynamic decision-making model based on fuzzy rules to build a rapid response system that operates within minutes, thus solving the problem of delayed response to control measures.

[0041] Data-driven precision dosing system: Through dynamic dose optimization and activator structure innovation, it achieves adaptive dosing depth and uniform concentration field of bacterial agent, eliminating insufficient efficacy, waste and reduced algae suppression efficiency caused by spatiotemporal mismatch.

[0042] Fully intelligent operation system: Relying on multimodal obstacle avoidance and remote sensing-execution-evaluation dual closed-loop architecture, it achieves unmanned operation of the entire chain of detection-decision-implementation, breaking through the bottleneck of large-scale deployment and environmental adaptability of traditional human-machine coupling.

[0043] Furthermore, it should be understood that after reading the above description of the present invention, those skilled in the art can make various alterations or modifications to the present invention, and these equivalent forms also fall within the scope defined by the appended claims.

Claims

1. A precise algae control system based on air-water collaborative intelligent decision-making, characterized in that, This includes a UAV hyperspectral water quality inversion and hotspot marking system, an unmanned surface vessel precision drug delivery and autonomous trajectory control system, and a data platform collaborative decision-making system; The UAV hyperspectral water quality inversion and hotspot labeling system includes a UAV and a hyperspectral water quality inversion system and a first hotspot labeling system mounted on the UAV; The working process of the UAV hyperspectral water quality inversion and hotspot marking system includes: the UAV inspects the designated water area along the preset route, uses the hyperspectral water quality inversion system to invert the chlorophyll a concentration in real time, and when the inverted chlorophyll a concentration exceeds the preset threshold, the first hotspot marking system records the coordinates of the risk point, generates the first four-element marker containing coordinates, time, index and the corresponding concentration value of the index, and transmits it to the data platform collaborative decision-making system. The unmanned surface vessel (USV) precision drug delivery and autonomous navigation control system includes the USV, a microbial agent activation and delivery device mounted on the USV, a fluorescent chlorophyll a sensor, and a second hotspot marking system. The working process of the USV precision drug delivery and autonomous navigation control system includes: the USV navigates along a planned trajectory, the fluorescent chlorophyll a sensor synchronously detects the chlorophyll a concentration in the water area, and when the detected chlorophyll a concentration is higher than a preset limit, the second hotspot marking system records the coordinates of the point exceeding the limit, generates a second four-element marker including coordinates, time, index, and concentration value, and transmits it to the data platform collaborative decision-making system. The data platform collaborative decision-making system includes a trajectory dynamic planning system and a dosage dynamic decision-making model, which are used to: plan the unmanned vessel's bacterial delivery trajectory based on the first four element markers after the UAV completes its inspection; and dynamically adjust the speed of the unmanned vessel and the dosage of the bacterial agent activation and delivery device according to concentration levels based on the second four element markers.

2. The algae control system based on air-water collaborative intelligent decision-making according to claim 1, characterized in that, The hyperspectral water quality inversion system includes a hyperspectral water quality imager and an edge computing module; The hyperspectral water quality imager has a wavelength range of 400-1000nm, a spectral resolution better than 5nm, and a spatial resolution better than 0.2m. The edge computing module incorporates a lightweight deep forest inversion model. After fractional differential preprocessing and stepwise discriminant analysis feature screening of 300 bands in the 400-1000nm range, it calculates the chlorophyll a concentration in water areas in real time with a spatial resolution better than 0.2m and an average relative error of <10%. The airborne terminal processes data in real time with a latency of <60s.

3. The algae control system based on air-water collaborative intelligent decision-making according to claim 1, characterized in that, The first and second hotspot marking systems use GNSS-RTK positioning modules, supporting GPS L1 / L2 and BeiDou B1 / B2 dual-frequency carrier phase differential positioning. They access local continuously operating reference station services via 4G network, receive differential correction data in RTCM 3.2 format in real time, output frequency 10Hz, horizontal positioning accuracy better than ±1.5cm, and elevation accuracy better than ±3cm.

4. The algae control system based on air-water collaborative intelligent decision-making according to claim 1, characterized in that, The unmanned vessel is also equipped with a multimodal autonomous obstacle avoidance system; The multimodal autonomous obstacle avoidance system integrates millimeter-wave radar and forward-looking camera to build dynamic maps in real time during navigation, enabling centimeter-level obstacle avoidance.

5. The algae control system based on air-water collaborative intelligent decision-making according to claim 1, characterized in that, The microbial agent activation and dosing device includes a microbial agent storage chamber, a screw conveyor, a microbial agent activator, and a flow booster; The microbial agent storage compartment has a volume of 50L and is used to store dry microbial agent powder. The screw conveyor is located at the bottom of the microbial agent storage chamber and conveys the microbial agent dry powder to the microbial agent activator at a constant speed. Its power is 0.12kW. The bacterial agent activator is L-shaped to ensure that the bacterial agent is added at a depth of 0.5m below the water surface; The propeller, located inside the microbial agent activator, has a power of 0.37 kW. It utilizes turbulent propeller flow to achieve highly efficient mixing of the microbial agent and water, with a mixing efficiency >95% and a microbial agent concentration gradient <0.5 g / m³. 3 •m, to achieve effective addition of microbial agents; The bacterial agent is an algae-inhibiting bacterium, which is an eco-friendly bacterial agent and does not have adverse effects on aquatic plants and animals.

6. The algae control system based on air-water collaborative intelligent decision-making according to claim 1, characterized in that, The trajectory dynamic planning system identifies points in the water area where the chlorophyll a concentration exceeds the threshold by analyzing the first four elements. It then uses a clustering algorithm to cluster these points to form key areas where unmanned surface vessels (USVs) need to deliver pesticides. Finally, it combines the distribution of key areas, the USV inspection routes, and the distribution of river systems to generate the optimal navigation path for the USV based on the simulated annealing algorithm. The system also uses a GIS map to visualize the navigation path.

7. The algae control system based on air-water collaborative intelligent decision-making according to claim 1, characterized in that, The dose dynamic decision model takes chlorophyll a concentration and its rate of change as input, and formulates PID parameters online through preset fuzzy rules to dynamically correct the ship speed, achieving adaptive control with a response time of <5 seconds and an overshoot of <10%, without the need to establish an accurate mathematical model of algal growth dynamics.

8. The algae control system based on air-water collaborative intelligent decision-making according to claim 1, characterized in that, The data platform collaborative decision-making system dynamically adjusts boat speed and bacterial dosage according to hydrological conditions and chlorophyll a concentration.

9. The algae precision control system based on air-water collaborative intelligent decision-making as described in any one of claims 1-8 is used for the precise control of algae outbreaks in aquatic bodies.

10. A method for precise control of algal blooms based on air-water collaborative intelligent decision-making, characterized in that, The algae control system based on air-water collaborative intelligent decision-making as described in any one of claims 1-8 is adopted; The method includes: The working process of the UAV hyperspectral water quality inversion and hotspot marking system includes: the UAV inspects the designated water area along the preset route, uses the hyperspectral water quality inversion system to invert the chlorophyll a concentration in real time, and when the inverted chlorophyll a concentration exceeds the preset threshold, the first hotspot marking system records the coordinates of the risk point, generates the first four-element marker containing coordinates, time, index and the corresponding concentration value of the index, and transmits it to the data platform collaborative decision-making system. The data platform collaborative decision-making system plans the unmanned vessel's bacterial delivery trajectory based on the first four elements markers after the drone completes its inspection. The unmanned vessel navigates along the planned trajectory, and the fluorescent chlorophyll a sensor simultaneously detects the chlorophyll a concentration in the water. When the detected chlorophyll a concentration is higher than the preset limit, the second hot spot marking system records the coordinates of the point exceeding the limit, generates a second four-element marker including coordinates, time, index and concentration value, and transmits it to the data platform collaborative decision-making system. The data platform collaborative decision-making system dynamically adjusts the speed of the unmanned vessel and the dosage of the bacterial agent activation and dosing device based on the concentration classification of the second and fourth element labels.

Images