Control system and method for intelligent breeding unmanned feeding ship

Abstract

The invention relates to the technical field of aquaculture equipment, and discloses a control system and method for an intelligent culture unmanned feeding ship, and the system comprises a shore-based command system which mainly comprises a remote control module and a cloud platform and is responsible for global task distribution, data integration and remote cooperative management, and a motion control unit. The motion control system comprises a flight control, the flight control is provided with an RTK module and a receiver, and the motion control system has three navigation modes of remote control navigation, preset track fixed-point navigation and intelligent autonomous navigation. According to the control system and method for the intelligent culture unmanned feeding ship, the general water quality of a water area can be monitored in real time, so that farmers are helped to closely pay close attention to the change of various indexes of the water, take corresponding measures to adjust and control the water quality and maintain a good culture environment, and according to various water quality data, the pesticide spraying concentration is adaptively controlled; and the feeding time is reasonably controlled, so that energy waste is reduced from the source.

Description

Technical Field

[0001] The present invention relates to the technical field of aquaculture equipment, and in particular to a control system and method for an intelligent aquaculture unmanned feeding ship. Background Art

[0002] Aquaculture is a labor-intensive industry. High demand, high costs, and low efficiency have resulted in a shortage of labor, difficulty recruiting, and high labor costs, leading to significant resource waste. Traditional manual feeding in aquaculture results in high feed waste rates, and frequent water changes contribute to over 35% of electricity costs. Direct wastewater discharge exacerbates offshore eutrophication. Drone feeding, while convenient, suffers from limitations such as low payload, high energy consumption, difficult operation, and significant weather impacts. Traditional unmanned vessels are limited to water quality monitoring or single feeding operations and lack multitasking capabilities. Furthermore, inefficient path planning algorithms result in insufficient operational coverage. Traditional "bow-shaped" paths account for 20%-25% of ineffective travel, further exacerbating energy waste. While there have been breakthroughs in the intelligentization of smart aquaculture equipment, it is more convenient for indoor farming but less effective outdoors.

[0003] At present, how to reduce the negative impact of aquaculture activities on aquatic ecology, curb environmental pollution caused by drug residues and feed residues, and avoid disease outbreaks and serious water pollution are all technical issues that need to be solved urgently. Summary of the Invention

[0004] (1) Technical problems solved

[0005] In response to the shortcomings of the existing technology, the present invention provides an intelligent aquaculture unmanned feeding ship control system and method, which solves the problems of negative impact of aquaculture activities on aquatic ecology and water environment pollution caused by pesticide residues and feed residues.

[0006] (2) Technical solution

[0007] To achieve the above objectives, the present invention provides the following technical solutions: a control system and method for an intelligent aquaculture unmanned feeding ship, including a shore-based command system, wherein the shore-based command mainly includes a remote control module and a cloud platform, which is responsible for global task allocation, data integration and remote collaborative management;

[0008] Motion control unit, the motion control system includes a flight control, which is equipped with an RTK module and a receiver. The motion control system has three navigation modes: remote control navigation, preset trajectory fixed-point navigation, and intelligent autonomous navigation. The navigation mode can be freely switched according to the instructions issued by the shore-based command terminal;

[0009] The mission execution unit is composed of a single-chip microcomputer and an onboard computer. The single-chip microcomputer and the onboard computer are connected via a CAN bus. They can monitor and upload water environment data in real time, view data remotely, monitor water body data, and identify dynamic obstacles. The flight control of the motion control unit is connected to the onboard computer via the MAVlink communication protocol.

[0010] Water quality monitoring system, which uses multiple water quality sensors to collect various key water quality data in aquaculture, and transmits the data to the single-chip microcomputer through the ADC analog-to-digital converter. The single-chip microcomputer cooperates with the onboard computer to calculate the water quality parameter data using the internal algorithm and output the control signal. The single-chip microcomputer realizes adaptive control of the feeder and the sprayer according to the control signal;

[0011] The feeder is installed at the bottom of the silo on the unmanned ship, and consists of two parts: a feeding component and a pneumatic discharge component. The single-chip microcomputer controlled by the task execution unit controls the discharge and discharge speed of the feeder, and the feed discharged by the feeding component is spread to the breeding area through airflow.

[0012] Preferably, the Pixhawk flight control system integrates multiple sensors such as gyroscopes, accelerometers, magnetometers, and barometers. During navigation, the Pixhawk flight control reads and writes data from all sensors, performs attitude calculations, and controls the navigation posture through the STM32F427 main control chip. The Pixhawk flight control is connected to the propeller of the unmanned boat through a PWM line. The unmanned boat adopts a dual propulsion mode of propellers and paddle wheels, which can be switched and adjusted at will to meet the requirements of various scenarios and missions. The motion control unit has three navigation modes: remote control navigation, preset trajectory fixed-point navigation, and intelligent autonomous navigation. The navigation mode can be switched freely according to the instructions issued by the shore-based command end.

[0013] Preferably, the RTK module and receiver configured with the flight control adopt real-time dynamic measurement technology, which is a real-time differential GPS technology based on carrier phase observation. The base station receives GPS signals from satellites and also receives signals from mobile stations. By differentially processing the phases of mobile station signals and satellite signals, the dynamic relative position difference between the mobile station and the base station is calculated, and finally high-precision position information is obtained, so that the positioning error can be reduced to the centimeter level.

[0014] Preferably, the single-chip microcomputer adopts STC32G12K128 single-chip microcomputer, which is responsible for real-time acquisition of data from various sensors and underlying device drivers. The onboard computer adopts Jetson NANO. The onboard computer runs the path planning algorithm and AI decision-making model. The two work together to control the execution layer to complete various tasks. The task execution unit has intelligent decision-making functions. The task execution unit has an alarm protection mechanism, which can trigger the alarm protection mechanism in time in case of emergency, providing accurate operation basis and safety guarantee for the unmanned ship.

[0015] Preferably, the water quality sensors used in the water quality monitoring system include temperature sensors, turbidity sensors, TDS sensors and pH sensors. The single chip microcomputer changes the output PWM signal duty cycle according to the changes in water quality data, dynamically adjusts the feeding amount, and the drug spraying matches the concentration according to the water quality data to complete adaptive adjustment during the feeding process.

[0016] Preferably, the task execution unit recognizes dynamic obstacles by being equipped with a laser radar. The laser radar transmits a detection signal to the target, compares the target echo reflected from the target with the transmitted signal, and obtains the target's distance, direction, height, speed, attitude, shape and other parameters after processing, thereby realizing detection and identification of surrounding objects. The onboard computer uses an improved A* algorithm for global planning, performs optimization on a known environment map, and uses a DWA algorithm to avoid newly appearing obstacles, while adopting a dual obstacle algorithm of global and local optimization.

[0017] Preferably, an ultrasonic module is integrated at the end of the feeder. When a foreign object is detected at the tail of the hull, the single-chip microcomputer will control the feeder to stop working and perform abnormal protection until the foreign object disappears. The stirring rod is driven by the rotation of the stepper motor in the feed box to stir the sticky feed and clear the equipment blockage. In addition, the single-chip microcomputer collects the current value of the stepper motor in the feed box. When the current of the stepper motor is detected to be abnormal, the single-chip microcomputer controls the motor to reverse and release the torque. When the stall occurs again, the stirring is stopped immediately and an abnormal alarm is issued to reduce the risk of equipment damage.

[0018] Preferably, the feeding assembly of the feeder is driven by a feeding motor, and the speed of the feeding motor is used to control the feeding speed and make the feed enter the pneumatic discharge assembly evenly. The fan configured in the pneumatic discharge assembly blows the feed out and spreads it onto the water surface of the breeding area, thereby achieving accurate control of the feed discharge amount.

[0019] The present invention also provides a control method for an intelligent aquaculture unmanned feeding boat, comprising the following steps:

[0020] Step 1: The shore-based command system issues tasks to the cloud platform. The cloud platform receives the tasks issued by the shore-based command system and the status information sent by the unmanned vessel. The main cloud controller calculates the final control signal based on the current task allocation.

[0021] Step 2: Start the water quality monitoring system to monitor and upload a variety of key water quality data in real time. The single-chip microcomputer cooperates with the onboard computer to calculate the water quality parameter data using principal component analysis. The output PWM signal duty cycle is changed according to the changes in water quality data. The operation of the feeder and sprayer is regulated in real time to achieve dynamic adjustment of feed feeding amount and match the spraying concentration of the drug according to the water quality data, completing adaptive regulation.

[0022] Step 3: The motion control unit of the unmanned boat controls the steering gear and navigation motor to adjust the heading and speed according to the final control signal received, and uses the paddle wheel for low-speed drive when feeding;

[0023] Step 4: After the feeding is completed, the feeder stops and the motion control unit of the unmanned boat switches the driving mode to use the propeller to drive the boat back to the shore at high speed.

[0024] Preferably, the principal component analysis method performs data analysis on the water quality of the aquaculture environment by introducing weight factors of various water quality parameters, and realizes adaptive control of the feeder and the sprayer. Assuming that there are n evaluation objects and m comment indicators;

[0025] Using the evaluation matrix Y=(y ik ) n×m represents (i=1,2,3,...,n;j=1,2,3,...,n);

[0026] Each indicator y in the evaluation matrix ij Convert to normalized index x ij , which means the standardized calculation formula is:

[0027]

[0028] Where x ij represents the standardized value of the jth indicator of the i-th evaluation object, is the average value of the j-th indicator, is the standard deviation of the jth indicator, n is the number of evaluation objects, and m is the number of standards;

[0029] The correlation coefficient matrix can be obtained by standardization of indicators: R = (r ij ) m×m

[0030]

[0031] In the formula, r represents a single correlation coefficient value. The closer the absolute value of r is to 1, the stronger the correlation between A and B is. The closer the absolute value of r is to 0, the weaker the correlation between A and B is. k represents the summation variable. It traverses the index of the first to nth evaluation objects. x kj represents the standardized value of the jth indicator of the kth celebrity object, r ij Represents the correlation coefficient between the i-th indicator and the j-th indicator in the correlation coefficient matrix R, r ij The closer the absolute value is to 1, the stronger the correlation between index i and j is, and the closer it is to 0, the weaker the correlation is. i The comparison of (i=1,2...,m) is λ1≥λ2≥λ3...λ m ≥0, that is, the variance of the principal component, λ i The value of size is the cumulative variance contribution rate of the corresponding principal component to the original sample, and the corresponding eigenvalues ​​are u1, u2, ...u m , where u j =(u 1j ,u 2j ,...,u nj ) T , transform the standardized indicators into m new indicator variables through the feature vector;

[0032]

[0033] Where: PC1 is the first principal component, PC2 is the second principal component, PC m is the mth principal component;

[0034] After determining the principal components, weight factors are given to different principal components and mapped to the duty cycle of the feeding motor PWM signal to complete adaptive feeding. After principal component analysis, combined with water quality sensor data, adaptive adjustment of the spraying system's agent concentration is achieved.

[0035] (3) Beneficial effects

[0036] Compared with the existing technology, the present invention provides a control system and method for an intelligent aquaculture unmanned feeding ship, which has the following beneficial effects:

[0037] 1. By processing data from various sensors, the approximate water quality of the water area can be monitored in real time. This will help farmers pay close attention to changes in various water indicators and take corresponding measures to adjust and control water quality, maintain a good breeding environment, and adaptively control the concentration of spraying according to various water quality data, reasonably control the feeding time, and reduce energy waste from the source.

[0038] 2. The improved A* path planning algorithm is used to effectively reduce invalid paths during the operation of the unmanned boat. The low-speed, high-torque paddle wheel is used for operation. Compared with the feeding by drone, the energy consumption of a single operation is only 35%, and the load capacity is much greater than that of the drone, which greatly reduces the number of operations and saves energy.

[0039] 3. It has intelligent protection measures and can respond efficiently to abnormal situations, reduce the risk of equipment damage, effectively increase its service life and reduce maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is the overall system diagram of the unmanned boat in the intelligent aquaculture unmanned feeding boat control system proposed by the present invention;

[0041] Figure 2 This is a diagram of the unmanned vessel architecture in the control system for an intelligent aquaculture unmanned feeding vessel proposed by the present invention;

[0042] Figure 3 This is a hardware connection diagram of the unmanned boat in the control system for an intelligent aquaculture unmanned feeding boat proposed by the present invention;

[0043] Figure 4 This is a cascade PID flow chart of the control system for an intelligent aquaculture unmanned feeding ship proposed by the present invention;

[0044] Figure 5 This is a flow chart of the A* path planning algorithm in the control system of an intelligent aquaculture unmanned feeding ship proposed by the present invention;

[0045] Figure 6 This is a simulation diagram of the improved A* algorithm trajectory in the control system of an intelligent aquaculture unmanned feeding ship proposed by the present invention;

[0046] Figure 7 This is a simulation diagram showing a test comparison between the improved A* algorithm and the traditional A* algorithm in an intelligent aquaculture unmanned feeding ship control system proposed by the present invention;

[0047] Figure 8 This is a trajectory diagram of the upper computer of the unmanned boat fixed-point cruising in the intelligent aquaculture unmanned feeding boat control system proposed by the present invention;

[0048] Figure 9 This is a data diagram of the unmanned boat on the cloud platform in the control system for an intelligent aquaculture unmanned feeding boat proposed by the present invention;

[0049] Figure 10 This is a structural diagram of the installation of a feeder and an unmanned boat feed silo in a control system for an intelligent aquaculture unmanned feeding boat proposed by the present invention;

[0050] Figure 11This is a structural diagram of a feeder in a control system for an intelligent aquaculture unmanned feeding ship proposed by the present invention;

[0051] Figure 12 This invention proposes an intelligent aquaculture unmanned feeding ship control system Figure 11 Schematic diagram of the internal structure of the feeder;

[0052] Figure 13 This is a schematic diagram of the structure of the feeder's feed assembly in the intelligent aquaculture unmanned feeding ship control system proposed by the present invention. Figure 1 ;

[0053] Figure 14 This is a schematic diagram of the structure of the feeder's feed assembly in the intelligent aquaculture unmanned feeding ship control system proposed by the present invention. Figure 2 ;

[0054] Figure 15 This is a schematic diagram of the structure of the feeder's feed assembly in the intelligent aquaculture unmanned feeding ship control system proposed by the present invention. Figure 3 ;

[0055] Figure 16 This is a structural schematic diagram of the circular base and retaining ring of the feeder in the intelligent aquaculture unmanned feeding ship control system proposed by the present invention.

[0056] In the figure: 1. Pneumatic conveying pipe; 2. Lower shell; 3. Upper shell; 4. Screw rod; 5. Feed spiral plate; 6. Rotor; 7. Ducted fan; 8. Reducer motor; 9. Notch; 10. Round base; 11. Distribution spiral plate; 12. Retaining ring; 13. Baffle plate; 14. Push plate; 15. Retaining ring; 16. Turbine spiral plate; 17. Turbine channel; 18. Intermediate body; 19. Feed channel; 20. Turbine. DETAILED DESCRIPTION

[0057] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0058] Example 1: Refer to the attached Figure 1-9A control system and method for an intelligent aquaculture unmanned feeding vessel includes a shore-based command system. The shore-based command mainly includes a remote control module and a cloud platform, which are responsible for global task allocation, data integration, and remote collaborative management. The shore-based command system uses a 4G network to remotely control the unmanned vessel. The cloud platform receives real-time water quality, location, and equipment status information sent back by the unmanned vessel, performs remote control and issues commands, and dynamically adjusts operation strategies to ensure efficient and safe task execution.

[0059] The motion control unit includes a Pixhawk flight controller equipped with an RTK module and receiver. The Pixhawk flight controller integrates multiple sensors, including a gyroscope, accelerometer, magnetometer, and barometer. During flight, the Pixhawk's internal STM32F427 controller reads and writes data from all sensors, calculates attitude, and controls attitude. The Pixhawk flight controller uses a Mahony-based explicit complement filter for attitude calculation by default. The Mahony algorithm uses a PID feedback controller to compensate for gyroscope errors by providing feedback. The PX4 uses an improved AHRS system, an extension of the traditional IMU system (gyroscope and accelerometer). Magnetometer data and GPS data are incorporated into cascaded PID control to correct the attitude calculation data.

[0060] The Pixhawk flight controller also has a backup attitude solution algorithm, which is the Kalman-based EKF (extended Kalman filter).

[0061] During navigation, the Pixhawk flight control uses the STM32F427 main control chip to read and write sensor data, perform attitude calculation and control. The Pixhawk flight control is connected to the propeller of the unmanned boat through the PWM line. The unmanned boat adopts a dual propulsion method of propeller and paddle wheel. When performing tasks, the paddle wheel is used to sail at a lower speed, and the propeller is used for high-speed propulsion when returning. This ensures full coverage of feeding and spraying tasks in complex water environments, and can be switched and adjusted at will to meet the needs of various scenarios and tasks, and reasonably optimize the energy utilization structure. The motion control unit has three navigation modes: remote control navigation, preset trajectory fixed-point navigation and intelligent autonomous navigation. According to the instructions issued by the shore-based command end, the navigation mode can be freely switched.

[0062] The RTK module and receiver configured in the flight control system use real-time dynamic measurement technology. This is a real-time differential GPS technology based on carrier phase observation. The base station receives GPS signals from satellites and also receives signals from the rover. By performing differential processing on the phases of the rover signal and the satellite signal, the dynamic relative position difference between the rover and the base station is calculated, ultimately obtaining high-precision position information, reducing the positioning error to the centimeter level.

[0063] The mission execution unit uses a laser radar to identify dynamic obstacles. The laser radar transmits a detection signal (laser beam) to the target and compares the target echo received from the target with the transmitted signal. After processing, the target's distance, direction, height, speed, attitude, shape and other parameters can be obtained to detect and identify surrounding objects. The unmanned ship is equipped with a Jetson NANO onboard computer, whose GPU uses NVIDIA Maxwell TM architecture, equipped with 128 NVIDIA Core 0.5TFLOPS (FP16), CPU uses quad core The A57 MPCore processor efficiently runs multiple algorithms and AI-powered decision-making. The onboard computer uses an improved A* algorithm for global planning, optimizes on a known environment map, and employs the DWA algorithm for obstacle avoidance of newly emerging obstacles. This algorithm, combined with a dual obstacle avoidance algorithm combining global and local optimization, effectively addresses issues such as uneven obstacle avoidance routes, long invalid paths, and poor real-time performance of the obstacle avoidance algorithm.

[0064] In simulation tests, this technology uses an improved A* algorithm, which has lower turning times and steering angles than the traditional A* algorithm, effectively reducing the energy lost when the unmanned boat turns. At the same time, it generates a "spiral-backtracking" compound path, reducing invalid travel by 20% compared to traditional methods.

[0065] The task execution unit is composed of a single-chip microcomputer and an onboard computer. The single-chip microcomputer adopts STC32G12K128, which is responsible for real-time acquisition of data from various sensors and underlying device drivers. The onboard computer adopts NVIDIA Jetson NANO. The onboard computer runs the path planning algorithm and AI decision-making model. The two work together to control the execution layer to complete various tasks. The task execution unit has intelligent decision-making functions and an alarm protection mechanism. In case of emergency, the alarm protection mechanism can be triggered in time to provide accurate operation basis and safety guarantee for the unmanned ship. The single-chip microcomputer and the onboard computer are connected through the CAN bus, which can monitor and upload water environment data, remote data viewing, water body data monitoring and dynamic obstacle identification in real time. The flight control of the motion control unit is connected to the onboard computer through the MAVlink communication protocol; the single-chip microcomputer is connected to the feeding motor, stirring motor and fan through the PWM line.

[0066] The water quality monitoring system uses a variety of water quality sensors to collect a variety of key water quality data in aquaculture, including temperature sensors, turbidity sensors, TDS sensors, and pH sensors. The data is then transmitted to the microcontroller through the ADC analog-to-digital converter. The microcontroller, in collaboration with the onboard computer, uses internal algorithms to calculate water quality parameter data and output control signals. The microcontroller changes the output PWM signal duty cycle based on changes in water quality data, dynamically adjusting the feeding amount. The drug spraying matches the concentration according to the water quality data, completing adaptive regulation during the feeding process.

[0067] The feeder is installed at the bottom of the silo on the unmanned ship. It consists of two parts: a feeding component and a pneumatic discharge component. The microcontroller controlled by the task execution unit controls the discharge and discharge speed of the feeder, and spreads the feed discharged by the feeding component to the breeding area through airflow. An ultrasonic module is integrated at the end of the feeder. When foreign matter is detected at the tail of the hull, the microcontroller will control the feeder to stop working and perform abnormal protection until the foreign matter disappears. The stirring rod is driven by the rotation of the stepper motor in the feed box to stir the sticky feed and clear the equipment blockage. In addition, the microcontroller collects the current value of the stepper motor in the feed box. When the current of the stepper motor is detected to be abnormal, the microcontroller controls the motor to reverse and release the torque. When the stall occurs again, the stirring is stopped immediately and an abnormal alarm is issued to reduce the risk of equipment damage.

[0068] The feeding assembly of the feeder is driven by a feeding motor. The speed of the feeding motor is used to control the feeding speed and ensure that the feed enters the pneumatic discharge assembly evenly. The fan configured in the pneumatic discharge assembly blows the feed out to the water surface in the breeding area, thereby accurately controlling the discharge amount of the feed.

[0069] When in use, the present invention adopts an improved A* algorithm, introduces energy consumption weight factors and coverage density parameters, and generates a "spiral-backtracking" compound path, which reduces invalid strokes by 20% and energy consumption per unit area by 24% compared with traditional methods.

[0070] To achieve multi-source sensor data fusion, this control system proposes an integrated architecture of "environmental perception-decision control-multi-task execution", breaking through the bottleneck of single function of aquaculture equipment. It can simultaneously realize multiple functions such as water quality monitoring, bait throwing, drug spraying, intelligent obstacle avoidance, remote data viewing, etc. Users can remotely monitor task status, understand the aquaculture water environment in real time, detect water quality problems in time, and issue early warnings and handle abnormal conditions.

[0071] This control system also has intelligent protection and adaptive adjustment. It implements a stall protection mechanism through real-time current monitoring and reverse torque release, reducing the risk of equipment damage. The feeding system dynamically adjusts the bait amount based on the water temperature and biological aggregation model. The drug spraying matches the concentration according to water quality data, reducing drug abuse by more than 25%, avoiding the generation of aquaculture waste at the source, solving the problem of traditional blind breeding, and effectively reducing feed waste rate.

[0072] This technical solution can be widely used in aquaculture, including but not limited to the following scenarios: 1. Freshwater pond fish farming, 2. Offshore fish cage farming, 3. Salt-alkali land ecological restoration aquaculture, and 4. Off-season greenhouse farming. This platform is suitable for diverse environments and embodies its core advantages of full-scenario adaptability, multi-task collaboration, and green energy conservation. By replacing manual labor with intelligent technology and enabling data-driven precision operations, it can reduce overall aquaculture costs by 30% to 40%, increase production by 15% to 25%, and simultaneously reduce carbon emissions and ecological disturbances, providing a standardized solution for the sustainable development of the aquaculture industry.

[0073] Example 2: Based on Example 1, the difference is that;

[0074] Refer to the attached Figure 10-16 In order to cooperate with the feeding work of the unmanned boat, this technical solution has redesigned a feeder. The feeder controls the feeding speed by the speed of the feeding motor and ensures that the feed enters the pneumatic discharge component evenly. The fan configured in the pneumatic discharge component blows the feed out and spreads it onto the water surface of the breeding area. The specific technical solution is as follows:

[0075] The feeding assembly includes an upper shell 3 and a lower shell 2. A circular base 10 is fixedly connected to the upper shell 3. A retaining ring 15 is fixedly connected to the upper end of the circular base 10. A storage tank is formed between the retaining ring 15 and the upper end of the circular base 10. Figure 13 As shown, the storage trough has a certain height through the retaining ring 15. When the upper feed enters, the feed is stored in the storage trough. A plurality of notches 9 are provided at the edge of the circular base 10. The notches 9 and the inner side of the upper shell 3 form a feeding channel. A cylindrical rotor 6 is sleeved in the upper shell 3. The upper end of the rotor 6 is fixedly connected to a spiral rod 4. The side wall of the rotor 6 is fixedly connected to a plurality of feeding spiral plates 5 and a plurality of distribution spiral plates 11. The distribution spiral plate 11 is located below the feeding spiral plate 5. When the feeder needs to work, the rotor 6 is driven by the feeding motor to start working. At this time, the upper spiral rod 4 can be extended into the bottom of the silo of the unmanned boat to stir the feed at any time to prevent the feed from accumulating and being unable to be discharged. Figure 12As shown, when the upper feeding spiral plate 5 rotates clockwise with the rotor 6, the feeding spiral plate 5 can be used to play the role of spiral feeding. When the feed passes through the gaps of multiple feeding spiral plates 5 and falls onto the distribution spiral plate 11, the distribution spiral plate 11 arranged opposite to the feeding spiral plate 5 can play the role of distribution, so that the feed slides into the feeding assembly. In addition, since multiple feeding spiral plates 5 are provided and have overlapping parts, it is difficult for a large amount of feed to enter the bottom of the feeding spiral plate 5 when the feeder is not working. At this time, the feeding spiral plate 5 can prevent the feed from sliding into the feeding assembly.

[0076] The side wall of the rotor 6 is sleeved with a retaining ring 12, and a plurality of baffles 13 are fixedly connected between the retaining ring 12 and the rotor 6. The plurality of baffles 13 divide the space between the retaining ring 12 and the rotor 6 into a plurality of feed channels 19. A guide plate 20 is fixedly connected to the feed channel. The upper and lower ends of the guide plate 20 are provided with a guide portion. The feed channel 19 is a fan-shaped structure, and the fixed guide plate 20 can divide the feed channel 19 again. At this time, when the feed enters the feed channel 19, the fallen feed is accumulated in the storage trough, and then the push plate 14 is provided at the lower end of the baffle 13. The push plate 14 is located in the storage trough and is connected to the circular base 1 0, the lower shell 2 is funnel-shaped and fixed to the lower end of the upper shell 3 by bolts, the lower end of the lower shell 2 is fixedly connected to the pneumatic conveying pipe 1, the lower end of the pneumatic conveying pipe 1 is fixedly connected to the reduction motor 8 (i.e., the feeding motor), the output end of the reduction motor 8 is fixedly connected to the rotating shaft, the upper end of the rotating shaft passes through the circular base 10 and is fixedly connected to the lower end of the rotor 6, the reduction motor 8 can drive the rotor 6 to rotate through the rotating shaft, and the feeding spiral plate 5, the distribution spiral plate 11 and the feeding channel installed on the rotor 6 can be operated during rotation. In addition, the push plate 14 provided can push the feed in the storage trough out through the material channel and fall into the pneumatic conveying pipe 1;

[0077] The pneumatic discharge assembly includes a ducted fan 7 (i.e., a fan) and a spoiler assembly. The ducted fan 7 is fixed at one end of the pneumatic conveying pipe 1, and the spoiler assembly is fixed in the pneumatic conveying pipe 1. The spoiler assembly includes an intermediate body 18. A plurality of spoiler spiral plates 16 are fixedly connected between the side wall of the intermediate body 18 and the pneumatic conveying pipe 1. A plurality of spiral spoiler channels 17 are formed between the plurality of spoiler spiral plates 16 and the pneumatic conveying pipe 1. The airflow generated by the ducted fan 7 can blow out the falling feed and the feed accumulated in the pneumatic conveying pipe 1, and after changing the flow direction through the spiral channel 17, the feed is evenly spread onto the water surface at its original location.

[0078] Compared with the traditional feeder with a dial plate, the electromagnetic control valve and the feed guide pipe are omitted, and the feed in the feed guide pipe will not spill onto the feed plate when the feeding is stopped. Through the improvement of the feeder this time, the speed of the feeding motor is used to achieve controlled feeding. When the feeding motor stops rotating, the feed stops randomly. By controlling the speed of the ducted fan 7, the feed can be quickly stopped, and the stop is rapid. The corresponding control system can better control the feeding plan according to the changes in water quality, making the control effect more precise.

[0079] Example 3:

[0080] The present invention provides a control method for an intelligent aquaculture unmanned feeding boat, comprising the following steps:

[0081] Step 1: The shore-based command system issues tasks to the cloud platform. The cloud platform receives the tasks issued by the shore-based command system and the status information sent by the unmanned vessel. The main cloud controller calculates the final control signal based on the current task allocation.

[0082] Step 2: Start the water quality monitoring system to monitor and upload a variety of key water quality data in real time. The single-chip microcomputer cooperates with the onboard computer to calculate the water quality parameter data using principal component analysis. The output PWM signal duty cycle is changed according to the changes in water quality data. The operation of the feeder and sprayer is regulated in real time to achieve dynamic adjustment of feed feeding amount and match the spraying concentration of the drug according to the water quality data, completing adaptive regulation.

[0083] The principal component analysis method analyzes the water quality of the aquaculture environment by introducing weight factors of various water quality parameters, and realizes adaptive control of the feeder and sprayer. It is assumed that there are n evaluation objects and m comment indicators.

[0084] Using the evaluation matrix Y=(y ik ) n×m represents i=1,2,3,...,n;j=1,2,3,...,n;

[0085] Each indicator y in the evaluation matrix ij Convert to normalized index x ij , which means the standardized calculation formula is:

[0086]

[0087] Where x ij represents the standardized value of the jth indicator of the i-th evaluation object, is the average value of the j-th indicator, is the standard deviation of the jth indicator, n is the number of evaluation objects, and m is the number of standards;

[0088] The correlation coefficient matrix can be obtained by standardization of indicators: R = (r ij ) m×m

[0089]

[0090] In the formula, r represents a single correlation coefficient value. The closer the absolute value of r is to 1, the stronger the correlation between A and B is. The closer the absolute value of r is to 0, the weaker the correlation between A and B is. k represents the summation variable. It traverses the index of the first to nth evaluation objects. x kj represents the standardized value of the jth indicator of the kth celebrity object, r ij Represents the correlation coefficient between the i-th indicator and the j-th indicator in the correlation coefficient matrix R, r ij The closer the absolute value is to 1, the stronger the correlation between index i and j is, and the closer it is to 0, the weaker the correlation is. i The comparison for i=1,2...,m is λ1≥λ2≥λ3...λ m ≥0, that is, the variance of the principal component, λ i The value of size is the cumulative variance contribution rate of the corresponding principal component to the original sample, and the corresponding eigenvalues ​​are u1, u2, ...u m , where u j =(u 1j ,u 2j ,...,u nj ) T , transform the standardized indicators into m new indicator variables through the feature vector;

[0091]

[0092] Where: PC1 is the first principal component, PC2 is the second principal component, PC m is the mth principal component;

[0093] After determining the principal components, weight factors are assigned to different principal components and mapped to the duty cycle of the feeding motor PWM signal to complete adaptive feeding. After principal component analysis, combined with water quality sensor data, adaptive adjustment of the spraying system's drug concentration is achieved;

[0094] Step 3: The motion control unit of the unmanned boat controls the steering gear and navigation motor to adjust the heading and speed according to the final control signal received, and uses the paddle wheel for low-speed drive when feeding;

[0095] Step 4: After the feeding is completed, the feeder stops and the motion control unit of the unmanned boat switches the driving mode to use the propeller to drive the boat back to the shore at high speed.

[0096] Through this method, the feeding system dynamically adjusts the amount of bait based on the water temperature and bioaggregation model, and the drug spraying matches the concentration according to the water quality data, reducing drug abuse, avoiding the generation of aquaculture waste from the source, solving the traditional blind breeding problem, and effectively reducing the feed waste rate.

[0097] It should be noted that the term "comprises" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article, or apparatus that includes a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0098] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent aquaculture unmanned feeding ship control system, characterized in that: include: The shore-based command system mainly includes a remote control module and a cloud platform, responsible for global task allocation, data integration and remote collaborative management; Motion control unit, the motion control system includes a flight control, which is equipped with an RTK module and a receiver. The motion control system has three navigation modes: remote control navigation, preset trajectory fixed-point navigation, and intelligent autonomous navigation. The navigation mode can be freely switched according to the instructions issued by the shore-based command terminal; The mission execution unit is composed of a single-chip microcomputer and an onboard computer. The single-chip microcomputer and the onboard computer are connected via a CAN bus. They can monitor and upload water environment data in real time, view data remotely, monitor water body data, and identify dynamic obstacles. The flight control of the motion control unit is connected to the onboard computer via the MAVlink communication protocol. Water quality monitoring system, which uses multiple water quality sensors to collect various key water quality data in aquaculture, and transmits the data to the single-chip microcomputer through the ADC analog-to-digital converter. The single-chip microcomputer cooperates with the onboard computer to calculate the water quality parameter data using the internal algorithm and output the control signal. The single-chip microcomputer realizes adaptive control of the feeder and the sprayer according to the control signal; The feeder is installed at the bottom of the silo on the unmanned ship, and consists of two parts: a feeding component and a pneumatic discharge component. The single-chip microcomputer controlled by the task execution unit controls the discharge and discharge speed of the feeder, and the feed discharged by the feeding component is spread to the breeding area through airflow.

2. The intelligent aquaculture unmanned feeding ship control system according to claim 1 is characterized by: The Pixhawk flight control system integrates multiple sensors including a gyroscope, accelerometer, magnetometer, and barometer. During navigation, the Pixhawk flight control uses the STM32F427 main control chip to read and write data from all sensors, calculate attitude, and control the situation. The Pixhawk flight control is connected to the propeller of the unmanned boat via a PWM line. The unmanned boat uses a dual propulsion method of propellers and paddle wheels, which can be switched and adjusted at will to meet the needs of various scenarios and missions. The motion control unit has three navigation modes: remote control navigation, preset trajectory fixed-point navigation, and intelligent autonomous navigation. The navigation mode can be freely switched according to the instructions issued by the shore-based command terminal.

3. The intelligent aquaculture unmanned feeding ship control system according to claim 1 is characterized by: The flight control is equipped with an RTK module and receiver that adopts real-time dynamic measurement technology. It is a real-time differential GPS technology based on carrier phase observation. The base station receives GPS signals from satellites and also receives signals from mobile stations. By differentially processing the phases of mobile station signals and satellite signals, the dynamic relative position difference between the mobile station and the base station is calculated, and finally high-precision position information is obtained, so that the positioning error can be reduced to the centimeter level.

4. The intelligent aquaculture unmanned feeding ship control system according to claim 1 is characterized by: The single-chip microcontroller adopts STC32G12K128, which is responsible for real-time acquisition of data from various sensors and underlying device drivers. The onboard computer adopts Jetson NANO, which runs the path planning algorithm and AI decision-making model. The two work together to control the execution layer to complete various tasks. The task execution unit has intelligent decision-making functions. The task execution unit has an alarm protection mechanism, which can be triggered in time in an emergency to provide accurate operation basis and safety guarantee for the unmanned ship.

5. The intelligent aquaculture unmanned feeding ship control system according to claim 1 is characterized by: The water quality sensors used in the water quality monitoring system include temperature sensors, turbidity sensors, TDS sensors and pH sensors. The microcontroller changes the output PWM signal duty cycle according to changes in water quality data, dynamically adjusts the feeding amount, and the drug spraying matches the concentration according to the water quality data to complete adaptive adjustment during the feeding process.

6. The intelligent aquaculture unmanned feeding ship control system according to claim 1 is characterized by: The task execution unit identifies dynamic obstacles by being equipped with a laser radar. The laser radar transmits a detection signal to the target and compares the target echo reflected from the target with the transmitted signal. After processing, the target's distance, direction, altitude, speed, attitude, shape and other parameters can be obtained to detect and identify surrounding objects. The onboard computer uses an improved A* algorithm for global planning, searches for the best on a known environment map, and uses the DWA algorithm to avoid newly appearing obstacles. At the same time, a dual obstacle algorithm of global and local optimization is used.

7. The intelligent aquaculture unmanned feeding ship control system according to claim 1 is characterized by: An ultrasonic module is integrated at the end of the feeder. When foreign objects are detected at the tail of the hull, the single-chip microcomputer will control the feeder to stop working and perform abnormal protection until the foreign objects disappear. The stirring rod is driven by the rotation of the stepper motor in the feed box to stir the sticky feed and clear the equipment blockage. In addition, the single-chip microcomputer collects the current value of the stepper motor in the feed box. When the current of the stepper motor is detected to be abnormal, the single-chip microcomputer controls the motor to reverse and release the torque. When the stall occurs again, the stirring is stopped immediately and an abnormal alarm is issued to reduce the risk of equipment damage.

8. The intelligent aquaculture unmanned feeding ship control system according to claim 1 is characterized by: The feeding assembly of the feeder is driven by a feeding motor, and the speed of the feeding motor is used to control the feeding speed and ensure that the feed enters the pneumatic discharge assembly evenly. The fan configured in the pneumatic discharge assembly blows the feed out and spreads it onto the water surface of the breeding area, thereby accurately controlling the discharge amount of the feed.

9. A feeding control method using the intelligent aquaculture unmanned feeding boat control system according to any one of claims 1 to 8, characterized in that: The following steps are included: Step 1: The shore-based command system issues tasks to the cloud platform. The cloud platform receives the tasks issued by the shore-based command system and the status information sent by the unmanned vessel. The main cloud controller calculates the final control signal based on the current task allocation. Step 2: Start the water quality monitoring system to monitor and upload a variety of key water quality data in real time. The single-chip microcomputer cooperates with the onboard computer to calculate the water quality parameter data using principal component analysis. The output PWM signal duty cycle is changed according to the changes in water quality data. The operation of the feeder and sprayer is regulated in real time to achieve dynamic adjustment of feed feeding amount and match the spraying concentration of the drug according to the water quality data, completing adaptive regulation. Step 3: The motion control unit of the unmanned boat controls the steering gear and navigation motor to adjust the heading and speed according to the final control signal received, and uses the paddle wheel for low-speed drive when feeding; Step 4: After the feeding is completed, the feeder stops and the motion control unit of the unmanned boat switches the driving mode to use the propeller to drive the boat back to the shore at high speed.

10. The control method for an unmanned feeding ship for intelligent aquaculture according to claim 9, characterized in that: The principal component analysis method introduces weight factors of various water quality parameters to analyze the water quality of the aquaculture environment and realizes adaptive control of the feeder and the sprayer. It is assumed that there are n evaluation objects and m comment indicators. Using the evaluation matrix Y=(y ij ) n×m represents (i=1,2,3,...,n;j=1,2,3,...,n); Each indicator y in the evaluation matrix ij Convert to normalized index x ij , which means the standardized calculation formula is: Where x ij represents the standardized value of the jth indicator of the i-th evaluation object, is the average value of the j-th indicator, is the standard deviation of the jth indicator, n is the number of evaluation objects, and m is the number of standards; The correlation coefficient matrix can be obtained by standardization of indicators: R = (r ij ) m×m In the formula, r represents a single correlation coefficient value. The closer the absolute value of r is to 1, the stronger the correlation between A and B is. The closer the absolute value of r is to 0, the weaker the correlation between A and B is. k represents the summation variable. It traverses the index of the first to nth evaluation objects. x kj represents the standardized value of the jth indicator of the kth celebrity object, r ij Represents the correlation coefficient between the i-th indicator and the j-th indicator in the correlation coefficient matrix R, r ij The closer the absolute value is to 1, the stronger the correlation between index i and j is, and the closer it is to 0, the weaker the correlation is. i The comparison of (i=1,2...,m) is λ1≥λ2≥λ3...λ m ≥0, that is, the variance of the principal component, λ i The value of size is the cumulative variance contribution rate of the corresponding principal component to the original sample, and the corresponding eigenvalues ​​are u1, u2, ...u m , where u j =(u 1j ,u 2j ,...,u nj ) T , transform the standardized indicators into m new indicator variables through the feature vector; Where: PC1 is the first principal component, PC2 is the second principal component, PC m is the mth principal component; After determining the principal components, weight factors are given to different principal components and mapped to the duty cycle of the feeding motor PWM signal to complete adaptive feeding. After principal component analysis, combined with water quality sensor data, adaptive adjustment of the spraying system's agent concentration is achieved.

Images