An offshore intelligent fish farm feeding system

Abstract

The present invention belongs to the technical field of intelligent feeding systems and discloses an intelligent marine fish farm feeding system, comprising multiple unmanned feeding boats, a bait warehouse, a fixed charging station, and a control system. The unmanned feeding boats are responsible for traveling between fish farming cages to feed the fish. The bait warehouse is used to store different types of feed and distribute them to the intelligent unmanned feeding boats according to the needs of target fish. The fixed charging station is used to charge the unmanned feeding boats. The control system is used to control the distribution of feed from the bait warehouse to the intelligent unmanned feeding boats and control the operation and charging of the unmanned feeding boats. The system of the present invention can achieve efficient feed utilization and maximize the savings in aquaculture costs. The automated control method directly reduces manpower input and labor intensity, improves management level, and can also achieve savings in bait costs and improve the quality of seafood.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent feeding systems, and in particular relates to an intelligent offshore fishery feeding system. Background Art

[0002] Aquaculture is a traditional and important industry. Current aquaculture processes rely primarily on manual feeding, which is not only labor-intensive but also imprecise in both feeding quantity and timing, making efficient and precise farming difficult. Smart aquaculture is currently emerging as a key development direction. By applying the Internet of Things, big data, and sophisticated control systems to aquaculture, it can significantly reduce the labor input required for offshore fish farming and enable highly automated, refined farming. Summary of the Invention

[0003] The present invention aims to provide an intelligent marine fish farm feeding system to solve the technical problems of low labor productivity, low resource utilization, high labor intensity and low accuracy in manual feeding.

[0004] To solve the above technical problems, the specific technical solution of the marine intelligent fish farm feeding system of the present invention is as follows:

[0005] An offshore intelligent fish farm feeding system includes multiple unmanned feeding boats, a bait warehouse, fixed charging piles, and a control system. The unmanned feeding boats are responsible for traveling between fish farming cages to feed the fish. The bait warehouse is used to store different types of feed and distribute them to the intelligent unmanned feeding boats according to the needs of target fish. The fixed charging piles are used to charge the unmanned feeding boats. The control system is used to control the bait warehouse to distribute feed to the intelligent unmanned feeding boats and control the operation and charging of the unmanned feeding boats. The control system remotely monitors and manages the operating status of the unmanned feeding boats, the feed storage status of the bait warehouse, and the charging status of the charging piles, and adjusts the feeding strategy based on the real-time understanding of the system operation status. Furthermore, the control system includes a fishery mapping module data and algorithm module and an output interface module, and the fishery mapping module includes a production planning module, a fishery mapping module and a central monitoring module; the production planning module formulates a feed procurement plan, a feed feeding plan, a drug feeding plan, and a production adjustment plan based on the data collected by the data and algorithm module, and generates a feeding plan based on the production plan, including the type and quantity of feed and drugs that should be fed to each type of fish cage during the period;

[0006] The fishery mapping module includes the location coordinates of the fixed charging warehouse of the unmanned feeding boat, the location coordinates of the bait warehouse when the unmanned feeding boat is feeding, and the location coordinates of the cages for breeding different types of fish; the mapping module is determined according to the cages and scale of the specific offshore fishery;

[0007] The central monitoring module is used to monitor the number and growth of various fish species, the amount and frequency of bait feeding, feed consumption, water quality monitoring and equipment operation; the data obtained by the central monitoring module is output through the data output interface module.

[0008] Furthermore, the offshore aquaculture cages are abstracted into a large rectangular block composed of multiple small rectangles in the fishery mapping module. Each large rectangular block breeds the same kind of fish. There is a fixed-distance channel between two adjacent cages for the unmanned feeding boat to travel. In the fishery mapping module, the channel between adjacent cages is abstracted into an unobstructed path for the unmanned feeding boat to search for the path and travel. The fishery mapping module maps the locations of the bait warehouse and the charging warehouse according to the actual situation of the offshore fishery. The bait warehouse and the charging warehouse are deployed on the periphery of the offshore fishery cages. An offshore fishery has multiple bait warehouses. Different bait warehouses are filled with different types of bait or medicines to load bait for the unmanned feeding boat.

[0009] Furthermore, the data and algorithm module uses an A* algorithm, and the unmanned feeding boats independently use the A* algorithm to plan a path from the bait warehouse to the target cage, so as to reach the destination in the shortest time and shortest distance; the algorithm steps are as follows:

[0010] For the offshore fishery environment including bait warehouses and target cages, the environment is represented by a topological map. In the directed connection network O=<P,C,T> In the example, P represents the set of target boxes in the network P = {p1,p2...,p n}, C represents the set of charging piles for feeding boats in the network C = {c1, c2..., c n}, T represents the set of edges in the directed connection network T = {t1, t2..., t n}, and each edge in the network is represented by an ordered pair of adjacent nodes:

[0011]

[0012] The purpose of planning multiple unmanned feeding boats is to find the path set with the lowest target time and path cost to complete the feeding in the presence of charging piles. The time taken by the unmanned feeding boat in the bait warehouse is the time to reach the loading point, feeding time, charging time and obstacle avoidance waiting time, which are set as and Ω k , so the scheduling target of all unmanned feeding ships is Where k is the number of the unmanned feeding boat;

[0013] Task priority sorting: The priority of the waiting action when the unmanned feeding ship is about to collide with other unmanned feeding ships at a certain node is the highest; the priority of the unmanned feeding ship charging task is higher than the priority of other feeding tasks assigned to the unmanned feeding ship; general feeding tasks are assigned with low priority; tasks with the same priority are assigned according to the FCFS method;

[0014] A single unmanned feeding boat uses the A* algorithm to plan the path from the bait warehouse to the target cage, with the completion of feeding requirements as the scheduling goal, thereby improving feeding efficiency.

[0015] Furthermore, the A* algorithm determines the path search direction by examining a certain value, and stipulates four forward directions of the unmanned feeding ship, namely forward, backward, left turn, and right turn. Its evaluation formula is f(x)=g(x)+h(x), where f(x) is the priority evaluation function, g(x) is the traversal depth or length function during path search, h(x) is a custom heuristic function, and h(x) is set to the length of the path to be traversed from the current point to the target point; g(x) is the length of the path traversed by the unmanned feeding ship, that is, the path cost spent, and f(x) is the sum of the costs. The execution steps of the A* algorithm are:

[0016] Step 1: Initialize two auxiliary sets W and V, where set W is an ordered set used to store the set of nodes to be estimated, including bait warehouses and charging piles. Set V is an unordered set used to store the set of nodes that have been estimated. Select the initial node S and put it into set W. W is regarded as a priority queue, in which the evaluation function of node v is examined. f(v) The smaller the value, the higher its priority, and it is placed at the front of the set W;

[0017] Step 2: Find in W f(v) The smallest node, if v is the destination, return the best path, otherwise delete node v from set W, insert it into set V, and loop through the four adjacent nodes of node v;

[0018] Step 3: Determine whether the node set W to be estimated is an empty set. If it is an empty set, it means that the path search fails. If it is not an empty set, return to step 2.

[0019] Furthermore, the search method of the A* algorithm is as follows: starting from the starting point, the unmanned feeding boat selects all feasible nodes for the next step each time, estimates the path time costs of all feasible nodes, and selects the node with the smallest path time cost to move.

[0020] Furthermore, the unmanned feeding boat includes path planning. Path planning refers to the process in which the unmanned feeding boat automatically completes the feeding task according to a predetermined path by planning a path. The path planning adopts a CBS algorithm.

[0021] Furthermore, the CBS decomposes the path planning problem into two sub-problems, an upper-level and a lower-level one, through a hierarchical search method. The lower-level search regards the unmanned feeding boat path planning problem as a route planning problem for multiple separate feeding boats, and obtains an effective solution by quickly solving these separate route planning problems. The upper-level search is responsible for resolving the path conflict problem between different feeding boats by creating new constraint tree nodes and adding new constraints to ensure that the path of each feeding boat can avoid conflicts. After the upper-level search finds a path conflict in the lower-level search, it will replan the specified feeding boat path to eliminate the path conflict problem.

[0022] Furthermore, the output interface module is used to transmit the algorithm's partial path planning data to each unmanned feeding boat, control the unmanned feeding boat to reach the designated location, and control the communication between the bait warehouse, fixed charging pile and the unmanned feeding boat, so that the unmanned feeding boat can complete the operation of returning to the warehouse for charging and loading bait.

[0023] Furthermore, the output interface module uses a communication protocol to transmit the path planning data to the control system of each unmanned feeding boat. In the control system, the unmanned feeding boat performs autonomous navigation according to the received path planning data, automatically drives along the preset path, and automatically performs feeding operations when it reaches the preset position. By using the output interface, the offshore intelligent fishery feeding system realizes centralized control and scheduling of the unmanned feeding boats. At the same time, the output interface module also provides guidance for the path planning and feeding strategy of the unmanned feeding boats by transmitting real-time environmental changes and fish demand data.

[0024] The present invention's intelligent marine fish farm feeding system and method offer the following advantages: By applying advanced technologies such as the Internet of Things, big data, and real-time precision measurement and control to aquaculture, the system enables automated feeding and the planning and control of unmanned feeding vessels. This system enables efficient feed utilization and minimizes aquaculture costs. The automated control method directly reduces manpower input and labor intensity, improving management while also reducing feed costs and improving seafood quality.

[0025] The system of this invention can precisely control the use of feed and medicine, reducing water pollution, effectively improving the aquatic environment, and playing a positive role in the ecological environment. Secondly, the unmanned electric feeding boat in this system can plan the shortest feeding route, saving time and costs while also achieving energy conservation and emission reduction, making a positive contribution to environmental protection and energy conservation and emission reduction. Finally, the development of this system can promote the further application of artificial intelligence in fisheries and achieve sustainable development of the aquaculture industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1This is a structural diagram of the offshore intelligent fish farm feeding system of the present invention;

[0027] Figure 2 This is a functional module block diagram of the control system of the present invention;

[0028] Figure 3 This is a system simulation diagram of the present invention. DETAILED DESCRIPTION

[0029] In order to better understand the purpose, structure and function of the present invention, the following is a further detailed description of the marine intelligent fish farm feeding system and method of the present invention in conjunction with the accompanying drawings.

[0030] like Figure 1 As shown, the present invention provides an intelligent marine fish farm feeding system, which includes multiple unmanned feeding boats, a bait warehouse, a fixed charging station, and a control system. The unmanned feeding boats are responsible for traveling between fish farming cages to feed fish and have autonomous navigation, positioning, and obstacle avoidance functions. The bait warehouse is used to store different types of feed and distribute it to the intelligent unmanned feeding boats according to the needs of the target fish. The fixed charging station is used to charge the unmanned feeding boats and can charge multiple unmanned feeding boats at the same time. It has multiple berths and docks, and the charging station also serves as a place for daily maintenance of the unmanned feeding boats and for taking shelter in bad sea conditions. The control system is used to control the distribution of feed from the bait warehouse to the intelligent unmanned feeding boats, and to control the operation and charging of the unmanned feeding boats. The control system can remotely monitor and manage the operating status of the unmanned feeding boats, the feed storage status of the bait warehouse, and the charging status of the charging station, making it easy to understand the system operation status in real time, adjust the feeding strategy in time, and ensure the efficient operation of the farm.

[0031] like Figure 2 As shown, the control system of the present invention includes a fishery mapping module, a data and algorithm module, and an output interface module.

[0032] The following is a description of each module of the system:

[0033] The fish farm mapping module includes a production planning module, a fish farm mapping module, and a central monitoring module. The production planning module: Based on the data collected by the data and algorithm module, it formulates a feed procurement plan (the amount of feed required for different fish species), formulates a feeding plan (feeding is carried out according to the needs of each fish and the prescribed feed feeding amount, such as calculating the total amount of feed based on the number of fish and weight, and then feeding according to the feeding amount and schedule for each fish). To prevent and treat fish diseases, a drug feeding plan is formulated (regularly check the health of the fish and feed drugs as needed. According to the needs and symptoms of different fish species, the appropriate type and dosage of drugs are selected). A production adjustment plan is formulated (production plans are adjusted in a timely manner according to factors such as season, number of fish, and feed demand). Finally, a feeding plan is generated based on the production plan, which should specifically describe the type and amount of feed and drugs that should be fed to the cages of each type of fish during that time period.

[0034] The fish farm mapping module includes the coordinates of the fixed charging bins for the unmanned feeding vessels, the coordinates of the bait storage bins during feeding operations, and the coordinates of the cages housing different fish species. The mapping module should be determined based on the specific cages and scale of the fish farm. In the mapping module, the cages are abstracted into a large rectangular block composed of multiple small rectangles, each housing the same species of fish, to align with the actual aquaculture practices in real-world fish farms. In reality, there is a fixed passageway between adjacent cages to facilitate navigation by unmanned feeding vessels. In the mapping module, the passageway between adjacent cages is abstracted into an unobstructed path, allowing the unmanned feeding vessels to search for and navigate the route. Furthermore, the locations of the bait storage and charging bins need to be mapped based on the actual conditions of the fish farm. Generally, these bins should be located outside the cages to facilitate feed refilling and maintenance. Typically, a fish farm should have several bait storage bins, each containing different types of bait or medicine, which can be used to load bait for the unmanned feeding vessels. The fixed charging bin is mainly used as a place for charging, maintenance and upkeep of the unmanned feeding boat. When no feeding operation is assigned, the unmanned feeding boat will remain stationary in the charging bin.

[0035] The central monitoring module monitors the number and growth of various fish species (water quality, temperature, estimated fish population, growth rate, and health status), feed amount and frequency (ensuring appropriate feeding to avoid waste and overfeeding), feed consumption (predicting the amount of bait used and resupply time for different fish species), water quality (monitoring water temperature, oxygen content, pH value, etc.), and equipment operation (monitoring whether the set feeding tasks are completed and whether the unmanned feeding boat clusters collide). Data collected by the central monitoring module is output through the data output interface module, facilitating monitoring by offshore fishery managers and the formulation and adjustment of production plans.

[0036] The data and algorithm modules use the A* algorithm.

[0037] The unmanned feeding boats independently use the A* algorithm to plan the path from the bait warehouse to the target cage, reaching the destination in the shortest time and shortest distance, thereby improving feeding efficiency.

[0038] For the offshore fish farm environment including bait warehouses and target cages, the environment can be represented by a topological map.<P,C,T> In the example, P represents the set of target boxes in the network P = {p1,p2...,p n}, C represents the set of charging piles for feeding boats in the network C = {c1, c2..., c n}, T represents the set of edges in the directed connection network T = {t1, t2..., t n}, and each edge in the network is represented by an ordered pair of adjacent nodes:

[0039]

[0040] The purpose of planning multiple unmanned feeding boats is to find a set of paths that can complete the feeding target time and the lowest path cost when charging piles are available. The time taken by the unmanned feeding boat in the bait warehouse is the time to reach the loading point, feeding time, charging time and obstacle avoidance waiting time, which are set as and Ω k Therefore, the scheduling target of all unmanned feeding ships is Where k is the number of the unmanned feeding boat.

[0041] Task priority sorting: The priority of waiting when an unmanned feeding ship is about to collide with other unmanned feeding ships at a certain node is the highest; secondly, the priority of the unmanned feeding ship charging task is higher than the priority of other feeding tasks assigned to the unmanned feeding ship; general feeding tasks are assigned according to low priority; tasks with the same priority are assigned according to the FCFS method.

[0042] A single unmanned feeding boat uses the A* algorithm to plan the path from the bait warehouse to the target cage, with the completion of feeding requirements as the scheduling goal, thereby improving feeding efficiency.

[0043] The A* algorithm is a typical heuristic search algorithm, which will avoid making the unmanned feeding ship take unnecessary or repeated paths, and ultimately find the optimal solution for the path. By examining a certain value to determine the direction of our path search, the present invention stipulates four forward directions for the unmanned feeding ship, namely forward, backward, left turn, and right turn. Its evaluation formula is f(x)=g(x)+h(x), where f(x) is the priority evaluation function, g(x) is the traversal depth or length function during path search, and h(x) is a custom heuristic function. Set h(x) to the length of the path from the current point to the target point; g(x) is the length of the path that the unmanned feeding ship has traveled, that is, the path cost, and f(x) is the sum of the costs. In the present invention, the execution steps of the A* algorithm are:

[0044] (1) Initialize two auxiliary sets W and V. Set W is an ordered set that stores the set of nodes to be estimated, including bait warehouses and charging stations. Set V is an unordered set that stores the set of nodes that have already been estimated. Select the initial node S and put it into set W. W can be regarded as a priority queue, in which the evaluation function of node v is examined. The smaller the f(v), the higher its priority, and it is placed at the front of set W.

[0045] (2) Find the node with the smallest f(v) in W. If v is the final destination, return the best path. Otherwise, delete node v from set W and insert it into set V. Then loop through the four adjacent nodes of node v.

[0046] (3) It is necessary to determine whether the node set W to be estimated is an empty set. If it is an empty set, it means that the path search has failed. If it is not an empty set, it is necessary to return to the second step.

[0047] The A* algorithm's search method is as follows: starting from the starting point, the unmanned feeding boat selects all feasible nodes for the next step each time, estimates the path time cost of all feasible nodes, and selects the node with the smallest path time cost to move. This algorithm can efficiently optimize the path, thereby improving feeding efficiency.

[0048] Path planning for an unmanned feeding vessel involves planning a route and automatically completing the feeding mission along a predetermined path. Because feeding missions typically require covering a large area of water, path planning for an unmanned feeding vessel is complex, requiring consideration of numerous factors, such as obstacles, currents, wind direction, and possible conflicts between multiple vessels.

[0049] The CBS algorithm is a centralized path planning algorithm that effectively resolves path conflicts between multiple unmanned feeding vessels. Using a hierarchical search approach, the algorithm decomposes the path planning problem into two sub-problems: an upper-level and a lower-level one. The lower-level search considers the path planning problem for unmanned feeding vessels as a route planning problem for multiple individual feeding vessels, rapidly solving these individual route planning problems to arrive at an effective solution. The upper-level search, on the other hand, resolves path conflicts between different feeding vessels by creating new constraint tree nodes and adding new constraints to ensure that each feeding vessel's path avoids conflicts. Specifically, if the upper-level search discovers a path conflict during the lower-level search, it replans the designated feeding vessel's path to eliminate the conflict.

[0050] Compared to other path planning algorithms, the CBS algorithm has the following advantages: First, it can quickly solve the route planning problem for a single feeding vessel, achieving an effective solution in a relatively short period of time; second, it can handle path conflicts between multiple feeding vessels, ensuring that each feeding vessel's path avoids conflicts; and finally, it is highly scalable and can easily handle complex feeding vessel path planning problems. The CBS algorithm is an excellent path planning algorithm, particularly suitable for the path planning problem of unmanned feeding vessels. Through a hierarchical search approach, the algorithm can quickly solve the route planning problem for a single feeding vessel and handle path conflicts between multiple feeding vessels, ensuring that each feeding vessel's path avoids conflicts, allowing multiple unmanned feeding vessels to avoid mutual interference and collisions when performing their missions.

[0051] The output interface module is a crucial component of the offshore intelligent fishery feeding system. It's primarily used to transmit the algorithm's path planning data to each unmanned feeding vessel, control its arrival at designated locations, and manage communication between the warehouse and charging station, allowing the vessels to return to the warehouse for recharging and bait loading. In the offshore intelligent fishery feeding system, the unmanned feeding vessels rely on pre-set feeding paths for autonomous navigation and feeding, and the output interface's role is to transmit path planning data to each vessel.

[0052] Specifically, the output interface typically uses communication protocols such as CAN and TCP / IP to transmit path planning data to the control system of each unmanned feeding vessel. In the control system, the unmanned feeding vessel can autonomously navigate based on the received path planning data, automatically driving along the preset path and automatically performing feeding operations when reaching the preset location. By using the output interface, the offshore intelligent fishery feeding system can achieve centralized control and scheduling of unmanned feeding vessels, improving the system's feeding efficiency and accuracy. At the same time, the output interface can also provide more precise guidance for the unmanned feeding vessel's path planning and feeding strategy by transmitting real-time data such as environmental changes and fish needs, further improving the system's aquaculture benefits and production efficiency.

[0053] So far, through the above modules, an automatic feeding system for offshore fisheries has been realized. The specific idea of the present invention is that a large-scale fish farm usually requires a certain number of unmanned feeding boats (usually less than 40), several warehouses for different types of bait, and a certain number of unmanned feeding boat charging stations. The unmanned feeding boats in this system can freely navigate between offshore cages through autonomous navigation and obstacle avoidance algorithms, and automatically feed the fish according to the preset feeding strategy and real-time environmental changes. In order to meet the dietary needs of different fish, the system is also equipped with a plurality of different types of bait warehouses, which can flexibly switch and feed the fish as needed. At the same time, in order to ensure the continuous operation and feeding efficiency of the unmanned feeding boat cluster, the system is also equipped with a plurality of unmanned feeding boat charging stations, which can provide continuous power support for the unmanned feeding boat cluster.

[0054] The process of feeding fish in an offshore cage farm by the offshore intelligent fish farm feeding system of the present invention comprises the following steps:

[0055] (a) The unmanned feeding boat departs from the charging station and obtains the required feed from the bait warehouse;

[0056] (b) The unmanned feeding boat uses the A* algorithm to plan the path from the bait warehouse to the target cage;

[0057] (c) When multiple unmanned feeding boats are performing feeding tasks, the CBS algorithm is used to prevent collisions with other unmanned feeding boats;

[0058] (d) The system automatically allocates and adjusts the feeding amount for the unmanned feeding boat according to the species and growth conditions of the fish in each cage;

[0059] (e) After completing the feeding task, the unmanned feeding boat returns to the fixed charging station for charging.

[0060] like Figure 3 FIG. 1 is a schematic diagram of the software simulation results of the present invention. The simulation results show that the solution is effective and reliable.

[0061] It will be understood that the present invention is described by way of some embodiments, and it will be appreciated by those skilled in the art that various changes or equivalent substitutions may be made to these features and embodiments without departing from the spirit and scope of the present invention. In addition, under the teachings of the present invention, these features and embodiments may be modified to adapt to specific circumstances and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are intended to be protected by the present invention.

Claims

1. An offshore intelligent fish farm feeding system, characterized in that: It includes multiple unmanned feeding boats, bait warehouses, fixed charging piles and a control system. The unmanned feeding boats are responsible for traveling between fish farming cages to feed fish. The bait warehouse is used to store different types of feed and distribute them to the intelligent unmanned feeding boats according to the needs of target fish; the fixed charging piles are used to charge the unmanned feeding boats; the control system is used to control the bait warehouse to distribute feed to the intelligent unmanned feeding boats, and control the operation and charging of the unmanned feeding boats. The control system remotely monitors and manages the operating status of the unmanned feeding boats, the feed storage status of the bait warehouse and the charging status of the charging piles, and adjusts the feeding strategy according to the real-time understanding of the system operation status; the control system includes a fishery modeling module, a data and algorithm module and an output interface module. The fishery modeling module includes a production planning module, a fishery mapping module and a central monitoring module; The production planning module formulates a feed procurement plan, a feed feeding plan, a drug feeding plan, and a production adjustment plan based on the data collected by the data and algorithm module, and generates a feeding plan based on the production plan, including the type and quantity of feed and drugs that should be fed to each type of fish cage during the period; The fishery mapping module includes the location coordinates of the fixed charging warehouse of the unmanned feeding boat, the location coordinates of the bait warehouse when the unmanned feeding boat is feeding, and the location coordinates of the cages for breeding different types of fish; the mapping module is determined according to the cages and scale of the specific offshore fishery; The central monitoring module is used to monitor the number and growth of various fish species, the amount and frequency of bait feeding, feed consumption, water quality monitoring and equipment operation; the data obtained by the central monitoring module is output through the data output interface module; The data and algorithm module adopts the A* algorithm, and the unmanned feeding boats independently use the A* algorithm to plan the path from the bait warehouse to the target cage, so as to reach the destination in the shortest time and shortest distance; The algorithm steps are as follows: For the offshore fishery environment including bait warehouses and target cages, the environment is represented by a topological map. In the directed connection network O=<P,C,T> In the example, P represents the set of target boxes in the network P = {p1,p2...,p n }, C represents the set of charging piles for feeding boats in the network C = {c1, c2..., c n }, T represents the set of edges in the directed connection network T = {t1, t2..., t n }, and each edge in the network is represented by an ordered pair of adjacent nodes: The purpose of planning multiple unmanned feeding boats is to find the path set with the lowest target time and path cost to complete the feeding in the presence of charging piles. The time taken by the unmanned feeding boat in the bait warehouse is the time to reach the loading point, feeding time, charging time and obstacle avoidance waiting time, which are set as and Ω k , so the scheduling target of all unmanned feeding ships is Where k is the number of the unmanned feeding boat; Task priority sorting: The priority of waiting when an unmanned feeding ship is about to collide with another unmanned feeding ship at a certain node is the highest; the priority of the unmanned feeding ship charging task is higher than the priority of other feeding tasks assigned to the unmanned feeding ship; general feeding tasks are assigned with low priority; The same priority is assigned according to the FCFS method; A single unmanned feeding boat uses the A* algorithm to plan the path from the bait warehouse to the target cage, with the completion of feeding requirements as the scheduling goal, thereby improving feeding efficiency.

2. The offshore intelligent fish farm feeding system according to claim 1, characterized in that: In the fishery mapping module, the offshore aquaculture cages are abstracted into a large rectangular block composed of multiple small rectangles. Each large rectangular block breeds the same kind of fish. There is a fixed-distance channel between two adjacent cages for unmanned feeding boats to travel. In the fishery mapping module, the channel between adjacent cages is abstracted into an unobstructed path for unmanned feeding boats to search for paths and travel. The fishery mapping module maps the locations of bait warehouses and charging warehouses according to the actual situation of the offshore fishery. The bait warehouses and charging warehouses are deployed on the periphery of the offshore fishery cages. An offshore fishery has multiple bait warehouses, and different bait warehouses are filled with different types of bait or medicines to load bait for unmanned feeding boats.

3. The marine intelligent fish farm feeding system according to claim 2, characterized in that: The A* algorithm determines the path search direction by examining a certain value, and stipulates four forward directions of the unmanned feeding ship, namely forward, backward, left turn, and right turn. Its evaluation formula is f(x)=g(x)+h(x), where f(x) is the priority evaluation function, g(x) is the traversal depth or length function during path search, h(x) is a custom heuristic function, and h(x) is set to the length of the path to be traversed from the current point to the target point; g(x) is the length of the path traversed by the unmanned feeding ship, that is, the path cost spent, and f(x) is the sum of the costs. The execution steps of the A* algorithm are as follows: Step 1: Initialize two auxiliary sets W and V. Set W is an ordered set, which is used to store the set of nodes to be estimated. Nodes include bait warehouses and charging stations. Set V is an unordered set, which is used to store the set of nodes that have been estimated. Select the initial node S and put it into set W. W is regarded as a priority queue. The evaluation function of node v is examined. The smaller the f(v), the higher its priority, and it is placed at the front of set W. Step 2: Find the node with the smallest f(v) in W. If v is the destination, return the best path. Otherwise, delete node v from set W and insert it into set V. Loop through the four nodes adjacent to node v. Step 3: Determine whether the node set W to be estimated is an empty set. If it is an empty set, it means that the path search fails. If it is not an empty set, return to step 2.

4. The offshore intelligent fish farm feeding system according to claim 3, characterized in that: The search method of the A* algorithm is as follows: starting from the starting point, the unmanned feeding ship selects all feasible nodes for the next step each time, estimates the path time cost of all feasible nodes, and selects the node with the smallest path time cost to move.

5. The offshore intelligent fish farm feeding system according to claim 3, characterized in that: The unmanned feeding boat includes path planning, which refers to the process in which the unmanned feeding boat automatically completes the feeding task by planning a path according to a predetermined path. The path planning adopts the CBS algorithm.

6. The offshore intelligent fish farm feeding system according to claim 5, characterized in that: The CBS decomposes the path planning problem into two sub-problems, an upper-level and a lower-level one, through a hierarchical search approach. The lower-level search treats the unmanned feeding boat path planning problem as a route planning problem for multiple separate feeding boats, and obtains an effective solution by quickly solving these separate route planning problems. The upper-level search is responsible for resolving the path conflict problem between different feeding boats by creating new constraint tree nodes and adding new constraints to ensure that the path of each feeding boat can avoid conflicts. If the upper-level search finds a path conflict in the lower-level search, it will replan the specified feeding boat path to eliminate the path conflict problem.

7. The offshore intelligent fish farm feeding system according to claim 5, characterized in that: The output interface module is used to transmit the algorithm's partial path planning data to each unmanned feeding boat, control the unmanned feeding boat to reach the designated location, and control the communication between the bait warehouse, fixed charging pile and the unmanned feeding boat, so that the unmanned feeding boat can complete the operation of returning to the warehouse for charging and loading bait.

8. The offshore intelligent fish farm feeding system according to claim 1, characterized in that: The output interface module uses a communication protocol to transmit the path planning data to the control system of each unmanned feeding vessel. In the control system, the unmanned feeding vessel performs autonomous navigation based on the received path planning data, automatically drives along a preset path, and automatically performs feeding operations when reaching a preset location. By using the output interface, the offshore intelligent fish farm feeding system realizes centralized control and scheduling of the unmanned feeding vessels. At the same time, the output interface module also provides guidance for the path planning and feeding strategy of the unmanned feeding vessels by transmitting real-time environmental changes and fish demand data.

Images