An intelligent feeding control method, system and medium for rice-frog co-culture

Abstract

The present invention relates to the technical field of rice frog breeding, and particularly relates to an intelligent feeding control method, system and medium for rice frog co-culture. The method includes: obtaining a preset feeding operation task and a scheduled start time of each feeding operation task; dividing each feeding operation task into several subtasks based on the paddy field area; when the scheduled start time is reached, using the linear programming method to generate an execution plan for each subtask in the current feeding operation task and execute it; during the execution of each subtask, when moving to the feeding start position of each subtask, the residual bait area and water accumulation area in the corresponding area of the current subtask are recognized in real time to adjust the feeding speed of the current subtask until moving to the feeding end position of the current subtask; after all subtasks are executed, return to the starting point to standby and wait for the next feeding operation task. It can improve the precision, flexibility and operation efficiency of the feeding operation, and effectively reduce feed waste.

Description

Technical Field

[0001] The present invention relates to the technical field of rice frog farming, and particularly to an intelligent feeding control method, system and medium for rice frog co-culture. Background Art

[0002] With the development of the ecological cycle agriculture model, the "rice frog" co-culture model has gradually become a typical ecological cycle agriculture model in the rice fishery integrated farming system. The "rice frog" co-culture model makes full use of the paddy field ecological space. On the one hand, the shallow water system of the paddy field and the rice plants provide a suitable shading place for the Rana nigromaculata, and the Rana nigromaculata can grow by preying on pests; on the other hand, the rice uses the predation of pests by the Rana nigromaculata to reduce pests and diseases, and at the same time exerts the fertility effect of frog feces, reducing the application of chemical fertilizers and pesticides, so as to achieve the effects of reducing pesticides and fertilizers, using one place for two purposes and harvesting twice in one place. Among them, the feeding link is one of the most important links in the "rice frog" co-culture production. Traditional feeding operations mainly rely on manual labor. Usually, feeding platforms are set around the paddy field, and artificial feeding is carried out every day according to the growth requirements of the Rana nigromaculata. This not only has a large labor intensity, high labor costs and low operation efficiency, but also has strong randomness in manual feeding, often resulting in uneven distribution of feed.

[0003] With the development of automation technology and intelligent agricultural equipment, automated feeding equipment such as track-type automatic feeding systems and small barrel-type fixed-point feeders have gradually emerged on the market. For example, Chinese Patent CN110731293A, a track-type aquaculture tank automatic feeding system and its control method. The system includes a large-capacity automatic metering and feeding bin arranged on the hardened platform of the aquaculture tank bank, and a reciprocating feeding track is connected to the bin; the reciprocating feeding track extends horizontally from the aquaculture tank bank to above the aquaculture tank, and at least one track is arranged in parallel; a track-type electric walking feeding device is arranged on the reciprocating feeding track. This track-type aquaculture tank automatic feeding system replaces manual feeding with a fixed time and quantity through the track-type electric walking feeding device and the setting of the track, and improves the efficiency of the feeding operation to a certain extent.

[0004] In the rice frog co-culture system, the feed not only needs to meet the nutritional needs of frogs, but also cannot have an adverse impact on the growth of rice. Therefore, the precise feeding of feed is crucial. However, each feeding operation of the existing automated feeding equipment is fixed-point and fixed-quantity feeding, and it is impossible to dynamically adjust the feeding strategy according to the changes in the actual distribution of the Rana nigromaculata, environmental changes, etc. It is difficult to adapt to the environmental diversity of the "rice frog" co-culture, a composite agricultural model, with a low degree of intelligence, resulting in unsatisfactory feeding effects, insufficient precision, a large amount of feed waste, and even adverse effects such as water quality deterioration in the paddy field. Summary of the Invention

[0005] In view of the above technical problems, the present invention provides an intelligent feeding control method, system and medium for rice-frog co-culture, aiming to design a feeding control method for the environmental diversity of the composite agricultural model of "rice-frog" co-culture, which can dynamically adjust the feeding strategy according to the changes in the actual distribution of black-spotted frogs and environmental changes, greatly improve the precision and flexibility of the feeding operation, improve the efficiency of the feeding operation, and effectively reduce the waste of feed.

[0006] In a first aspect, the present application provides an intelligent feeding control method for rice-frog co-culture, including the following steps:

[0007] S1. Obtain a feeding operation task table, where the feeding operation task table includes a number of feeding operation tasks and the scheduled start time of each feeding operation task;

[0008] S2. Divide each feeding operation task into several subtasks based on the paddy field area, and the parameters of each subtask include the feeding start position, the feeding end position, and the feeding speed;

[0009] S3. When the scheduled start time is reached, use the linear programming method to generate the execution plan of each subtask in the current feeding operation task based on the parameters of each subtask, and execute each subtask in the current feeding operation task according to the execution plan;

[0010] S4. During the execution of each subtask, when moving to the feeding start position of the subtask, real-time identify the residual bait area and the water accumulation area in the corresponding area of the current subtask;

[0011] S5. Adjust the feeding speed of the current subtask based on the residual bait area and the water accumulation area in the area until moving to the feeding end position of the current subtask;

[0012] S6. Repeat steps S4 - S5 until all subtasks in the current feeding operation task are completed, return to the preset starting point to standby, and wait for the scheduled start time of the next feeding operation task.

[0013] In some embodiments, in S3, using the linear programming method to generate the execution plan of each subtask in the current feeding operation task based on the parameters of each subtask includes:

[0014] Define the decision variables of the subtask, where the decision variables of the subtask include the required number of supplementary feeds and the increased walking mileage for each supplementary feed;

[0015] Based on the parameters of each subtask, determine the constraint conditions of the decision variables;

[0016] Taking the sum of the increased walking distances required for minimizing the number of supplementary feeding times in the current feeding operation task as the objective function, a linear programming model is constructed based on the decision variables, objective function, and constraint conditions of the subtasks;

[0017] Based on the parameters of each subtask, the linear programming model is solved to obtain a set of optimal decision variable values;

[0018] Based on this set of optimal decision variable values, the execution plan for each subtask in the current feeding operation task is generated.

[0019] In some embodiments, the parameters of the subtasks further include the feeding movement speed. Based on the parameters of each subtask, the constraint conditions of the decision variables are determined, including:

[0020] Based on the feeding start position, feeding end position, and feeding movement speed of each subtask, the feeding time of each subtask is calculated;

[0021] Based on the feeding time and feeding speed of each subtask, the feeding amount of each subtask is calculated;

[0022] Based on the feeding amounts of each subtask, the total feeding amount of the current feeding operation task is calculated;

[0023] Based on the total feeding amount of the current feeding operation task, the upper limit of the required number of supplementary feeding times is determined.

[0024] In some embodiments, in S5, based on the residual bait area and water accumulation area in the field area, the feeding speed of the current subtask is adjusted, including:

[0025] Based on the residual bait area, the residual bait amount in the current area is calculated;

[0026] Based on the residual bait amount in the current area, the proportional control algorithm is used to adjust the feeding speed of the current subtask to obtain the first feeding speed of the current subtask;

[0027] Based on the water accumulation area in the current area, the non - linear control algorithm is used to adjust the first feeding speed of the current subtask to obtain the second feeding speed of the current subtask, and the second feeding speed is used as the final feeding speed of the current subtask.

[0028] In some embodiments, based on the residual bait amount in the current area, using the proportional control algorithm to adjust the feeding speed of the current subtask to obtain the first feeding speed of the current subtask, including:

[0029] Based on the residual bait amount in the current area, the residual bait correction coefficient is calculated, and the residual bait correction coefficient is inversely proportional to the residual bait amount;

[0030] Multiplying the residual bait correction coefficient by the feeding speed of the current subtask, and the product obtained is the first feeding speed of the current subtask.

[0031] In some embodiments, based on the water accumulation area in the current area, a non-linear control algorithm is used to adjust the first feeding speed of the current subtask to obtain the second feeding speed of the current subtask, including:

[0032] Determine whether the water accumulation area of the current area reaches a preset water accumulation threshold. If the water accumulation area of the current area is greater than or equal to the preset water accumulation threshold, adjust the second feeding speed of the current subtask to 0. If the water accumulation area of the current area is less than the preset water accumulation threshold, perform the following steps:

[0033] Based on the water accumulation area of the current area, calculate a water accumulation correction coefficient, and the water accumulation correction coefficient is inversely proportional to the water accumulation area;

[0034] Multiply the water accumulation correction coefficient by the first feeding speed of the current subtask, and the obtained product is the second feeding speed of the current subtask.

[0035] In some embodiments, in S3, each subtask in the current feeding operation task is executed according to the execution plan, including:

[0036] Before moving to the feeding start position of each subtask, obtain weather data in real time;

[0037] Based on the weather data, evaluate whether the current weather condition will have an adverse impact on the feeding operation. If the evaluation result is that it will have an adverse impact on the feeding operation, adjust the feeding speed of the next subtask to 0. If the evaluation result is that it will not have an adverse impact on the feeding operation, keep the feeding speed of the next subtask unchanged.

[0038] In some embodiments, the following steps are further included:

[0039] Select several points on the moving path, set RFID tags for each point and calibrate the positions of the RFID tags;

[0040] During the execution of each subtask, when moving to the position of each RFID tag, correct the current positioning data to the position of the corresponding RFID tag.

[0041] In a second aspect, the present application provides an intelligent feeding control system for rice-frog co-culture, including a feeding robot, a ground track device, a feeding tower, a charging device, and a cloud controller. The feeding robot is located on the ground track device and moves along the trajectory of the ground track device. The feeding robot, the feeding tower, and the charging device are all communicatively connected to the cloud controller. The feeding robot is used to execute an intelligent feeding control method for rice-frog co-culture as described above.

[0042] In a third aspect, a computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements an intelligent feeding control method for rice-frog co-culture as described above.

[0043] The beneficial technical effects of the present invention may at least include:

[0044] 1. Compared with the existing rail-type automatic feeding system that usually can only perform fixed-point and quantitative feeding according to a preset path and feeding amount and cannot adjust the feeding amount according to the actual situation, resulting in a large amount of feed waste, the present invention adopts an intelligent feeding control method, system and medium for rice-frog co-culture. By dividing the feeding task into multiple subtasks based on the distribution of the paddy field area, using the linear programming method to generate the execution plan for each subtask, optimizing the feeding path and speed, thereby improving the efficiency of the overall feeding operation, and dynamically adjusting the feeding strategy based on the real-time identified residual bait area and water accumulation area, it can more flexibly respond to the specific conditions of different feeding sections in the paddy field area, thus better adapting to the environmental diversity of the rice-frog co-culture composite agricultural model, while realizing the precise feeding of feed, effectively reducing feed waste, improving the feeding effect, greatly improving the intelligence level of the feeding system, and contributing to the sustainable development of the rice-frog co-culture system;

[0045] 2. Combining the characteristics of the feeding robot running on the track, determining the constraint conditions of the decision variables with the parameters of each subtask, and constructing a linear programming model with the goal of minimizing the total walking mileage increased by the number of refilling times required for the current feeding operation task. Based on the linear programming model, the autonomous planning of the feeding and refilling operations in the entire feeding operation task process is realized. While ensuring the continuity of the feeding operation, the moving path and refilling strategy are optimized, and the efficiency and reliability of the overall feeding operation are greatly improved;

[0046] 3. Instead of the existing fixed-point and fixed-quantity feeding method of automated feeding equipment, a feeding method is adopted in which feed is delivered at a certain feeding moving speed and feeding speed along the feeding section between the feeding start / stop positions, that is, feeding while moving. This not only improves the overall operation efficiency, but also, since the feed is evenly distributed along the moving path, it can reduce the waste of feed caused by concentrated feeding at fixed points. At the same time, by monitoring the residual bait area in the feeding section and converting it into the amount of residual bait, the feeding speed is adjusted in real time according to the actual feeding situation of the black-spotted frogs in the current area, reducing the feeding speed in areas with a large amount of residual bait to avoid overfeeding, ensuring the accuracy of the feeding amount, and avoiding adverse effects such as water pollution that may be caused by excessive feed. Further monitor the water accumulation area in the feeding section, and adjust the feeding speed in real time according to the actual water accumulation situation in the current area. Stop feeding or reduce the feeding speed proportionally in sections with a large water accumulation area to avoid feed dissolving in water and being unable to be ingested, reducing feed waste, so as to realize accurate decision-making on the initial feeding amount by adjusting it in real time according to the on-site visual feedback of the feeding section, thereby realizing dynamic adjustment of the feeding strategy according to the changes in the actual feeding situation of frogs and environmental changes, ensuring that black-spotted frogs obtain sufficient nutrition, while reducing feed waste, realizing intelligent feeding, reducing manual monitoring and operation, and contributing to the sustainable development of the rice-frog co-culture system and promoting the progress of ecological agriculture technology.

[0047] 4. By further linking with weather data and appropriately adjusting the feeding speed of the current subtask according to the current weather conditions to adjust the feeding amount, thereby optimizing the feeding effect, ensuring the healthy growth of black-spotted frogs and the effective utilization of feed, reducing feed waste and environmental problems caused by weather reasons, and further improving the intelligence level of the feeding strategy;

[0048] 5. Combining the scene characteristics of the ground track, a positioning correction method based on the fusion of a mileage encoder and RFID tags is proposed. While effectively eliminating the cumulative error of mileage calculation of the mileage encoder, it realizes low-cost precise positioning and navigation, providing a good positioning data basis for the accuracy and efficiency of feeding.

[0049] Other features and advantages of the present invention will be disclosed in detail in the following specific embodiments and drawings. Brief Description of the Drawings

[0050] The following further describes the present invention with reference to the drawings:

[0051] Figure 1 It is a flowchart of an intelligent feeding control method for rice-frog co-culture according to an embodiment of the present invention.

[0052] Figure 2 It is a schematic structural diagram of an intelligent feeding control system for rice-frog co-culture according to an embodiment of the present invention. Detailed Description of the Invention

[0053] The technical solutions of the embodiments of the present invention will be explained and described below with reference to the accompanying drawings of the embodiments of the present invention. However, the following embodiments are only the preferred embodiments of the present invention and not all of them. Based on the embodiments in the implementation manners, other embodiments obtained by those skilled in the art without creative efforts all fall within the protection scope of the present invention.

[0054] In the following description, terms such as "inner", "outer", "upper", "lower", "left", "right", etc. indicating orientation or positional relationship are only for the convenience of describing the embodiments and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention.

[0055] Please refer to the attached Figure 1 , Figure 1 which shows a schematic flowchart of an intelligent feeding control method for rice-frog co-culture provided by an embodiment of this specification.

[0056] As Figure 1 shown, the intelligent feeding control method for rice-frog co-culture may at least include the following steps:

[0057] S1. Obtain a feeding operation task list, where the feeding operation task list includes several feeding operation tasks and the scheduled start time of each feeding operation task.

[0058] Among them, the feeding operation task list in this embodiment is input by the farmer into the control module of the feeding robot in advance through a control program. The feeding operation task list in the feeding operation task list may be the daily feeding operation task or the weekly feeding operation task. The specific corresponding period can be set according to the growth requirements of the black-spotted frogs, and this embodiment does not limit this.

[0059] S2. Divide each feeding operation task into several subtasks based on the paddy field area. The parameters of each subtask include the feeding start position, the feeding end position, and the feeding speed.

[0060] It can be understood that the ground track device is laid along the ridge ground of the paddy field area. Each time the feeding operation task is executed, the feeding robot moves on the ground track with a fixed trajectory to feed the entire paddy field area. Dividing each feeding operation task into several subtasks based on the paddy field area makes the field areas corresponding to each subtask different and the feeding strategies different.

[0061] Among them, the parameters of the subtask include the starting position and the ending position of feeding, which represent a part of the road section in the preset paddy field area of this embodiment as the feeding platform for black-spotted frogs, that is, the feeding section of the feeding robot. In this way, the black-spotted frogs are trained to feed in a fixed area, and the existing feeding method of fixed-point and quantitative feeding by automated feeding equipment is changed. Instead, a method of feeding feed at a certain feeding moving speed and feeding speed in the feeding section between the starting and ending positions of feeding is adopted, that is, feeding while moving. This can not only improve the overall operation efficiency, but also reduce the waste of feed caused by concentrated feeding at fixed points because the feed is evenly distributed along the moving path. At the same time, by adjusting the feeding moving speed and feeding speed, the feeding amount of feed can be more accurately controlled according to the actual distribution of black-spotted frogs.

[0062] In this embodiment, a certain track position is set as the positioning origin, for example, the corresponding position of a certain feeding tower on the track, denoted as D0 = 0, and it is assumed that the direction of the head of the feeding robot along the track is positive and the direction of the tail of the feeding robot is negative. Then all positions Di on the track can be expressed as the distance along the track trajectory to the positioning origin.

[0063] Among them, the feeding speed in the parameters of the subtask can indirectly show the feeding amount of the feeding robot. Specifically, the feeding robot usually has a roller for loading feed and a feeding motor. According to the feeding amount per revolution of the roller with different specifications (for example, in grams) and the rotation speed of the feeding motor (for example, in g / min), the actual feeding amount of the feeding robot during the moving time can be calculated.

[0064] S3. When the predetermined start time is reached, based on the parameters of each subtask, an execution plan for each subtask in the current feeding operation task is generated by using the linear programming method, and each subtask in the current feeding operation task is executed according to the execution plan.

[0065] Among them, the linear programming method is a mathematical method used to optimize (maximize or minimize) a certain linear objective function under given linear constraint conditions. In the feeding operation task for rice-frog co-culture in this embodiment, the linear programming method is used to find the optimal feeding strategy to maximize the operation efficiency.

[0066] Furthermore, in this embodiment, in S3, generating an execution plan for each subtask in the current feeding operation task by using the linear programming method based on the parameters of each subtask includes:

[0067] S301. Define the decision variables of the subtask. The decision variables of the subtask include the required number of feeding replenishments and the increased walking mileage for each feeding replenishment.

[0068] It can be understood that the drum of the feeding robot can only load a certain amount of feed, which may not be sufficient to support the execution of each subtask in the entire feeding operation task. When the feeding robot is executing the feeding process according to the subtask execution plan of the feeding operation task or at the preset starting point, once the remaining feed in the drum of the feeding robot is lower than the set threshold, the feeding robot needs to enter the automatic feeding replenishment mode and move to the feeding replenishment tower for feed replenishment. Therefore, in the process of executing each subtask in the current feeding operation task, when to replenish feed and which feeding replenishment tower to move to for feed replenishment to maximize the operation efficiency of the entire feeding operation task are what need to be considered in this embodiment.

[0069] Specifically, in this embodiment, for each subtask, the decision variables include the number of feeding replenishment times j, the number y of the target feeding replenishment tower corresponding to each feeding replenishment, and the walking mileage increased by the j-th feeding replenishment. Among them, the target feeding replenishment tower is the feeding replenishment tower with the shortest distance, and the walking mileage increased by the j-th feeding replenishment can be expressed as:

[0070]

[0071] Among them, represents the feeding termination position of the feeding robot reaching the F-th subtask, represents the position of the feeding replenishment tower y (i.e., the target feeding replenishment tower) with the shortest distance from the F-th subtask, that is, the mileage passed through when going to the target feeding replenishment tower for feed replenishment and returning to the feeding termination position of the subtask after completing a certain subtask.

[0072] S302. Based on the parameters of each subtask, determine the constraint conditions of the decision variables.

[0073] Specifically, in this embodiment, the parameters of the subtask further include the feeding moving speed. Then, based on the parameters of each subtask, determine the constraint conditions of the decision variables, including:

[0074] Based on the feeding starting position, feeding termination position, and feeding moving speed of each subtask, calculate the feeding time of each subtask;

[0075] Based on the feeding time and feeding speed of each subtask, calculate the feeding amount of each subtask;

[0076] Based on the feeding amount of each subtask, calculate the total feeding amount of the current feeding operation task;

[0077] Based on the total feeding amount of the current feeding operation task, determine the upper limit of the required number of feeding replenishment times.

[0078] It can be understood that the implementation method of "based on the parameters of each subtask, determine the constraint conditions of the decision variables" in this embodiment is:

[0079] First, since the feeding method in this embodiment is to feed the feed at a certain feeding moving speed and feeding speed on the feeding section between the feeding start / stop positions, therefore, by dividing the distance of the feeding section between the feeding start position and the feeding end position by the feeding moving speed, the quotient obtained is the feeding time;

[0080] Second, the feeding robot is usually equipped with a drum for loading feed and a feeding motor. By dividing the feeding speed by the feeding amount q (for example, in grams) per revolution of the drum of different specifications, the feeding speed can be converted into the rotational speed of the feeding motor (for example, in g / min). Then, multiplying the feeding time by the rotational speed of the feeding motor, the product obtained is the actual feeding amount of the feeding robot in the current subtask;

[0081] Next, by adding up the feeding amounts of each subtask in the current feeding operation task, the total feeding amount required for the current feeding operation task can be obtained , where n represents the total number of subtasks divided in the current feeding operation task, represents the feeding amount of the nth subtask.

[0082] Finally, the constraint conditions for the decision variables are determined as:

[0083]

[0084] Among them, represents the full bin feed amount in the drum of the feeding robot, represents the ceiling function, represents the number of subtasks completed after the jth replenishment, then the serial number of the feeding subtask completed after the jth replenishment is , F represents the cumulative number of subtasks completed before the jth replenishment, , represents the sum of the feeding amounts of the subtasks completed after the jth replenishment.

[0085] S303. Taking the sum of the walking distances increased by minimizing the number of replenishment times required for the current feeding operation task as the objective function, a linear programming model is constructed based on the decision variables, objective function, and constraint conditions of the subtasks.

[0086] Among them, the objective function of the linear programming model can be expressed as: , where x is a positive integer and is the same as the number of replenishment times required for the current feeding operation task.

[0087] S304. Solve the linear programming model based on the parameters of each subtask to obtain a set of optimal decision variable values.

[0088] Solve the objective function of the linear programming model under the constraints using a linear programming algorithm (such as the simplex method or the interior point method, etc.). The solution result will give the optimal number of supplementary feeding times j for the current feeding operation task, the target feeding tower y corresponding to each supplementary feeding, and the number of subtasks completed after each supplementary feeding. as a set of optimal decision variable values.

[0089] S305. Based on this set of optimal decision variable values, generate the execution plan for each subtask in the current feeding operation task.

[0090] It can be understood that the execution plan for each subtask in the current feeding operation task includes the total number of subtasks n, the optimal number of supplementary feeding times, which subtasks need to be supplemented before execution, the target feeding tower number y corresponding to each supplementary feeding, and the movement path plan for the overall feeding operation task, etc., so as to ensure the optimization of the efficiency of the overall feeding operation.

[0091] Combined with the characteristics of the feeding robot running on the track, determine the constraint conditions of the decision variables based on the parameters of each subtask, and construct a linear programming model with the objective of minimizing the total walking mileage increased by the number of supplementary feeding times required for the current feeding operation task. Based on the linear programming model, realize the autonomous planning of feeding and supplementary feeding operations during the entire feeding operation task, optimize the movement path and supplementary feeding strategy while ensuring the continuity of the feeding operation, and greatly improve the efficiency and reliability of the overall feeding operation.

[0092] S4. During the execution of each subtask, when moving to the feeding start position of the subtask, real-time identify the residual bait area and the water accumulation area in the corresponding area of the current subtask.

[0093] Among them, "residual bait" refers to the feed left on the feeding section that has not been eaten by the black-spotted frogs during the feeding process. It can be understood that identifying the residual bait area is to evaluate the feeding situation of the animals. Because when obtaining the initially set feeding amount in the feeding operation task, the estimated feeding amount of the animals may be too high, resulting in excessive feed delivery. Moreover, the appetite of the black-spotted frogs is easily affected by various factors, such as environmental changes, health conditions, mood fluctuations, etc., which may cause them not to eat all the feed, resulting in residual bait.

[0094] It can be understood that identifying the water accumulation area is to evaluate the edibility of the feed. Because the feed is easily soluble in water, the increase in the water accumulation area will cause more feed to dissolve in the water, which will directly affect the edibility of the feed, because the dissolved feed will not be eaten by the black-spotted frogs, resulting in feed waste.

[0095] S5. Based on the residual bait area and the water accumulation area in the area, adjust the feeding speed of the current subtask until moving to the feeding end position of the current subtask.

[0096] Further, in this embodiment, in S5, adjusting the feeding speed of the current subtask based on the residual bait area and the water accumulation area in the region includes:

[0097] S501, calculating the amount of residual bait in the current region based on the residual bait area.

[0098] It can be understood that since this embodiment adopts a feeding method of feeding the feed at a certain feeding moving speed and feeding speed in the feeding section between the feeding start / stop positions, rather than the fixed-point and fixed-quantity feeding method of the prior art, the feed will not be stacked at a certain fixed point, but is dispersed in the feeding section in granular form, and the volume of a single feed is very small. Therefore, the recognized residual bait area can be converted into residual bait weight.

[0099] Exemplarily, the implementation manner of calculating the amount of residual bait in the current region based on the residual bait area can be:

[0100] 1. Calculate the volume of a single feed particle: Usually, the particles of black-spotted frog feed are regular in shape, so geometric formulas can be used to calculate the volume. For example, for spherical particles, the volume V = (4 / 3)πr³, where r is the radius of a single feed particle.

[0101] 2. Estimate the number of particles per unit area: Count the number of particles N on the known area A', then the number of particles per unit area n = N / A'.

[0102] 3. Convert the recognized residual bait area into residual bait weight: The residual bait weight W = residual bait area × number of particles per unit area n × volume V of a single particle × density ρ of the feed particle.

[0103] It should be noted that this is only an example of calculating the amount of residual bait in the current region based on the residual bait area. The calculation method in this example depends on the estimation of the number of particles, volume, and density. There may be some errors in actual operation, and more complex measurement methods can also be used to improve the accuracy of the estimation. This embodiment does not make any limitations in this regard.

[0104] S502, adjusting the feeding speed of the current subtask using a proportional control algorithm based on the amount of residual bait in the current region to obtain the first feeding speed of the current subtask.

[0105] Specifically, in this embodiment, adjusting the feeding speed of the current subtask using a proportional control algorithm based on the amount of residual bait in the current region to obtain the first feeding speed of the current subtask includes:

[0106] Calculating a residual bait correction coefficient based on the amount of residual bait in the current region, and the residual bait correction coefficient is inversely proportional to the amount of residual bait;

[0107] Multiply the residual bait correction coefficient by the feeding speed of the current subtask, and the product obtained is the first feeding speed of the current subtask.

[0108] It can be understood that the residual bait correction coefficient is inversely proportional to the amount of residual bait, indicating that the more the amount of residual bait, the smaller the residual bait correction coefficient, and the slower the first feeding speed of the current subtask, thus reducing the actual feeding amount of the current subtask.

[0109] Exemplarily, the residual bait correction coefficient can be expressed as:

[0110]

[0111] where represents the amount of residual bait in the current area, k represents the field of view parameter of the feeding robot. When the amount of residual bait is 0, the residual bait correction coefficient is 1. The field of view parameter (Field of View, FOV) is a parameter that describes the scene range that an imaging system (such as a camera, telescope, microscope, radar, etc.) or the human eye can observe. It defines the maximum monitoring area that can be observed through this imaging system or the scene range that can be captured.

[0112] S503. Based on the water accumulation area in the current area, use a non-linear control algorithm to adjust the first feeding speed of the current subtask to obtain the second feeding speed of the current subtask, and use the second feeding speed as the final feeding speed of the current subtask.

[0113] Among them, the non-linear control algorithm in this embodiment can be expressed as a proportional control algorithm with threshold cut-off, that is, this adjustment strategy combines the characteristics of proportional control and threshold control, so it belongs to a type of non-linear control algorithm. In industrial control systems, this type of control algorithm is sometimes called "two-position control" or "bang-bang control" because it has only two states: fully open (in this case, normal proportional reduction) and fully closed (feeding speed is 0). The principle of the non-linear control algorithm is that when a certain condition is reached (that is, the water accumulation area reaches the threshold), the control action (that is, the feeding speed) will change discontinuously.

[0114] Specifically, in this embodiment, based on the water accumulation area in the current area, using a non-linear control algorithm to adjust the first feeding speed of the current subtask to obtain the second feeding speed of the current subtask includes:

[0115] Judge whether the water accumulation area of the current area reaches the preset water accumulation threshold. If the water accumulation area of the current area is greater than or equal to the preset water accumulation threshold, adjust the second feeding speed of the current subtask to 0. If the water accumulation area of the current area is less than the preset water accumulation threshold, perform the following steps:

[0116] Based on the waterlogging area of the current area, a waterlogging correction coefficient is calculated, and the waterlogging correction coefficient is inversely proportional to the waterlogging area;

[0117] Multiply the waterlogging correction coefficient by the first feeding speed of the current subtask, and the obtained product is the second feeding speed of the current subtask.

[0118] It can be understood that the feed is easily dissolved in water. The increase in the waterlogging area will cause more feed to dissolve in water, which will directly affect the edibility of the feed, because the dissolved feed will not be eaten by the black-spotted frogs, resulting in feed waste. Therefore, in this embodiment, a non-linear control strategy for waterlogging is designed. Once the waterlogging area reaches or exceeds the preset waterlogging threshold, regardless of the initial set feeding amount of the current subtask, the feeding speed is directly set to 0, that is, no feeding is carried out; and when the waterlogging area does not exceed the threshold, the dissolution speed of the feed may be relatively slow, and the feeding speed of the black-spotted frogs is sufficient to eat most of the undissolved feed. However, in order to reduce the waste caused by feed dissolution, when the waterlogging area increases but does not reach the dangerous level of the preset waterlogging threshold, the feeding speed is reduced in geometric progression, which can reduce the amount of feed put into the water, thereby reducing the impact of feed dissolution on feeding behavior.

[0119] Exemplarily, the waterlogging correction coefficient can be expressed as:

[0120]

[0121] where represents the waterlogging area in the current area, b represents the preset waterlogging threshold, k represents the field-of-view parameter of the feeding robot, and when the waterlogging area is 0, the waterlogging correction coefficient is 1.

[0122] The environment in the rice-frog co-culture system is variable, including the growth of rice, the activity habits of frogs, the depth and fluidity of water, etc. These factors will affect the distribution of Rana nigromaculata in the paddy field and the feeding efficiency. Therefore, in this embodiment, the residual bait area of the feeding section is monitored and converted into the amount of residual bait, and the feeding speed is adjusted in real time according to the actual feeding situation of Rana nigromaculata in the current area, reducing the feeding speed in the area with a large amount of residual bait to avoid overfeeding, ensuring the accuracy of the feeding amount, and avoiding adverse effects such as water pollution that may be caused by excessive feed. Further, the water accumulation area of the feeding section is monitored, and the feeding speed is adjusted in real time according to the actual water accumulation situation in the current area. Stop feeding or reduce the feeding speed proportionally in the section with a large water accumulation area to avoid the feed being dissolved in water and unable to be eaten, reducing feed waste, so as to realize the accurate decision-making of the feeding amount by adjusting the initial feeding amount in real time according to the on-site visual feedback of the feeding section, thereby realizing the dynamic adjustment of the feeding strategy according to the change of the actual feeding situation of frogs and the environmental change, ensuring that Rana nigromaculata obtains sufficient nutrition, while reducing feed waste, realizing intelligent feeding, reducing manual monitoring and operation, contributing to the sustainable development of the rice-frog co-culture system, and promoting the progress of ecological agriculture technology.

[0123] S6. Repeat steps S4 - S5 until all subtasks in the current feeding operation task are completed, return to the preset starting point to standby, and wait for the scheduled start time of the next feeding operation task.

[0124] Existing orbital automatic feeding systems usually can only perform fixed-point and quantitative feeding according to preset paths and feeding amounts, and cannot adjust the feeding amount according to the actual situation, resulting in a large amount of feed waste. In this embodiment, the feeding task is divided into multiple subtasks based on the distribution of the paddy field area, and a linear programming method is used to generate the execution plan for each subtask, optimizing the feeding path and speed, thereby improving the efficiency of the overall feeding operation. And based on the dynamically recognized residual bait area and water accumulation area, the feeding strategy is adjusted, which can more flexibly respond to the specific conditions of different feeding sections in the paddy field area, thus better adapting to the environmental diversity of the composite agricultural model of rice-frog co-culture. At the same time, accurate feed placement is realized, effectively reducing feed waste, improving the feeding effect, and greatly improving the intelligence level of the feeding system, contributing to the sustainable development of the rice-frog co-culture system.

[0125] In an embodiment of this specification, in S3, executing each subtask in the current feeding operation task according to the execution plan includes:

[0126] Before moving to the feeding start position of each subtask, obtain weather data in real time;

[0127] Evaluate whether the current weather conditions will have an adverse impact on the feeding operation based on weather data. If the evaluation result is that it will have an adverse impact on the feeding operation, adjust the feeding speed of the next subtask to 0. If the evaluation result is that it will not have an adverse impact on the feeding operation, keep the feeding speed of the next subtask unchanged.

[0128] Among them, the next subtask is the subtask that is about to move to the feeding starting position to start the feeding operation.

[0129] In this embodiment, the weather conditions that will have an adverse impact on the feeding operation include but are not limited to the following, and this embodiment does not make a limit on this:

[0130] 1. Excessive rainfall will cause waterlogging in the feeding area, affecting the activities and feeding behaviors of black-spotted frogs. Heavy rainfall may wash away the feed, causing waste.

[0131] 2. High temperature may cause the feed to deteriorate faster and be unsuitable for feeding.

[0132] 3. Strong winds may blow the feed away, affecting the accuracy of feeding.

[0133] 4. Black-spotted frogs may be reluctant to move under strong light, thus affecting the feeding effect.

[0134] 5. The temperature and humidity changes brought about by seasonal changes may affect the growth cycle and feeding habits of black-spotted frogs.

[0135] This embodiment is further linked with weather data. According to the current weather conditions, appropriately adjust the feeding speed of the current subtask to adjust the feeding amount, so as to optimize the feeding effect, ensure the healthy growth of black-spotted frogs and the effective utilization of feed, reduce feed waste and environmental problems caused by weather reasons, and further improve the intelligence level of the feeding strategy.

[0136] In the existing rail-type automatic feeding system, the positioning data of the feeding robot is crucial for ensuring the accuracy and efficiency of feeding. However, complex positioning methods usually require higher-precision sensors and more complex hardware devices, increasing the installation and maintenance costs of the feeding system and also increasing the complexity of the system, which may lead to a reduction in its reliability and further affect the normal feeding operation of the feeding robot. For this reason, in an embodiment of this specification, the following steps are further included:

[0137] Select several points on the moving path, set RFID tags for each point and calibrate the positions of each RFID tag;

[0138] During the execution of each subtask, when moving to the position of each RFID tag, correct the current positioning data to the position of the corresponding RFID tag.

[0139] Among them, the RFID (Radio-Frequency Identification) tag, namely the radio frequency identification tag, is a technology that uses radio waves to achieve data communication in order to identify specific targets and read and write relevant data.

[0140] Specifically, the implementation method of this embodiment is as follows:

[0141] First, the selection of several points on the moving path can be fixed facility points such as the feeding tower, the charging device, or the columns arranged at equal intervals along the track. This embodiment does not limit this.

[0142] Secondly, measure and calibrate the track distances of each RFID tag relative to the positioning origin along the track trajectory as the positions of each RFID tag.

[0143] Then, when the feeding robot moves, the central control system obtains the mileage encoder information of the track driving wheels at a preset sampling interval, calculates the rolling mileage of the track driving wheels through the mileage encoder information, and determines the positioning data of the feeding robot on the track. Among them, the mileage encoder (also known as the incremental encoder or rotary encoder) is a low-cost sensor used to measure the angle or position of the rotating shaft and is a commonly used position feedback device.

[0144] Finally, whenever the feeding robot moves to the position of each RFID tag, after being detected by the RFID reader, the current positioning data of the feeding robot is corrected to the position of the corresponding RFID tag.

[0145] It can be understood that this embodiment combines the scene characteristics of the ground track and proposes a positioning correction method based on the fusion of the mileage encoder and the RFID tag. While effectively eliminating the cumulative error of the mileage calculation of the mileage encoder, it realizes low-cost precise positioning and navigation, providing a good positioning data basis for the accuracy and efficiency of feeding.

[0146] Please refer to the appendix Figure 2 , Figure 2 which is a schematic structural diagram of an intelligent feeding control system for rice-frog co-culture provided by an embodiment of this specification.

[0147] As Figure 2 shown, the intelligent feeding control system for rice-frog co-culture can at least include a feeding robot 1, a ground track device 2, a feeding tower 3, a charging device 4, and a cloud controller ( Figure 2 not shown in the figure). The feeding robot 1 is located on the ground track device 2 and moves along the trajectory of the ground track device 2. The feeding robot 1, the feeding tower 3, and the charging device 4 are all communicatively connected to the cloud controller. The feeding robot 1 is used to execute an intelligent feeding control method for rice-frog co-culture provided by the foregoing embodiment.

[0148] To implement an intelligent feeding control method for rice frog co-culture provided in the foregoing embodiments, the feeding robot 1 in this embodiment is correspondingly provided with hardware devices such as a visual recognition module and an odometer encoder, which will not be elaborated herein.

[0149] It can be understood that the technical concept of the feeding robot in the intelligent feeding control system for rice frog co-culture provided in this embodiment is similar to that of an intelligent feeding control method for rice frog co-culture provided in the foregoing embodiments, which will not be elaborated herein.

[0150] The walking control operation, feeding control operation of the feeding robot 1, feeding control operation between the feeding robot 1 and the feeding tower 3, charging control operation between the feeding robot 1 and the charging device 4, etc. in the intelligent feeding control system for rice frog co-culture provided in this embodiment are all similar to the control operations of the existing rail-type automatic feeding system, which will not be elaborated herein.

[0151] Another embodiment of this specification provides a computer-readable storage medium, in which instructions are stored. When they run on a computer or a processor, the computer or the processor is made to execute one or more steps in the foregoing embodiments of an intelligent feeding control method for rice frog co-culture. If the respective component modules of the above-mentioned electronic device are implemented in the form of software functional units and used as independent downstream task predictions, they can be stored in the computer-readable storage medium.

[0152] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of this specification are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired manner (such as coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or a wireless manner (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that integrates one or more available media. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a Digital Versatile Disc (DVD)), or a semiconductor medium (for example, a Solid State Disk (SSD)), etc.

[0153] As described above, only the preferred embodiments disclosed in this application and the description of the applied technical principles are provided. Those skilled in the art should understand that the scope of protection involved in this disclosure is not limited to the technical solutions formed by the specific combination of the above technical features. It should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the technical solutions formed by mutually replacing the above features with the (but not limited to) technical features with similar functions disclosed in this disclosure.

[0154] In addition, although the operations are depicted in a particular order, this should not be construed as requiring that the operations be performed in the particular order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, although a number of specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the disclosure. Certain features described in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, the various features described in the context of a single embodiment can also be implemented separately or in any suitable sub-combination in multiple embodiments.

Claims

1. An intelligent feeding control method for rice-frog co-culture, characterized in that: The following steps are involved: S1, obtaining a feeding operation task table, wherein the feeding operation task table includes a plurality of feeding operation tasks and a scheduled start time for each feeding operation task; S2, dividing each feeding task into several subtasks based on the rice field area, and the parameters of each subtask include a feeding starting position, a feeding ending position and a feeding speed; S3, when the scheduled start time is reached, a linear programming method is used to generate an execution plan for each subtask in the current feeding operation task based on the parameters of each subtask, and each subtask in the current feeding operation task is executed according to the execution plan; S4, in the process of executing each subtask, when moving to the feeding starting position of the subtask, real-time identification of the residual bait area and the accumulated water area in the corresponding area of ​​the current subtask; S5, adjusting the feeding speed of the current subtask based on the residual bait area and the accumulated water area in the area until the feeding end position of the current subtask is moved; S6, repeating steps S4-S5 until all subtasks in the current feeding task are completed, returning to the preset starting point and waiting for the scheduled start time of the next feeding task; Among them, in S5, the feeding speed of the current subtask is adjusted based on the residual bait area and the water accumulation area in the area, including: Based on the residual bait area, the residual bait amount in the current area is calculated; Based on the amount of residual bait in the current area, a proportional control algorithm is used to adjust the feeding speed of the current subtask to obtain the first feeding speed of the current subtask; Determine whether the water accumulation area in the current area has reached the preset water accumulation threshold. If the water accumulation area in the current area is greater than or equal to the preset water accumulation threshold, adjust the second feeding speed of the current subtask to 0. If the water accumulation area in the current area is less than the preset water accumulation threshold, perform the following steps: Based on the waterlogging area of ​​the current area, a waterlogging correction coefficient is calculated, where the waterlogging correction coefficient is inversely proportional to the waterlogging area; Multiply the water accumulation correction coefficient by the first feeding speed of the current subtask, and the product is the second feeding speed of the current subtask; The second feeding speed is used as the final feeding speed of the current subtask.

2. The intelligent feeding control method for rice-frog co-cultivation as claimed in claim 1, characterized in that: In S3, a linear programming method is used to generate an execution plan for each subtask in the current feeding task based on the parameters of each subtask, including: Defining decision variables of the subtask, wherein the decision variables of the subtask include the required number of refills and the additional walking mileage for each refill; Based on the parameters of each subtask, determine the constraints of the decision variables; Taking minimizing the sum of the additional walking distance required for the current feeding task as the objective function, a linear programming model is constructed based on the decision variables, objective functions and constraints of the subtasks. Solve the linear programming model based on the parameters of each subtask to obtain a set of optimal decision variable values; Based on the set of optimal decision variable values, an execution plan for each subtask in the current feeding operation task is generated.

3. The intelligent feeding control method for rice-frog co-cultivation as claimed in claim 2, characterized in that: The parameters of the subtasks also include the feeding movement speed. Based on the parameters of each subtask, the constraint conditions of the decision variables are determined, including: Based on the feeding starting position, feeding ending position and feeding moving speed of each subtask, the feeding time of each subtask is calculated; Based on the feeding time and feeding speed of each subtask, the feeding amount of each subtask is calculated; Based on the feeding amount of each subtask, the total feeding amount of the current feeding task is calculated; Based on the total feeding amount of the current feeding task, determine the upper limit of the required feeding times.

4. The intelligent feeding control method for rice-frog co-cultivation as claimed in claim 1, characterized in that: Based on the amount of residual bait in the current area, the proportional control algorithm is used to adjust the feeding speed of the current subtask to obtain the first feeding speed of the current subtask, including: Based on the amount of residual bait in the current area, a residual bait correction coefficient is calculated, and the residual bait correction coefficient is inversely proportional to the residual bait amount; Multiply the residual bait correction coefficient by the feeding speed of the current subtask, and the product obtained is the first feeding speed of the current subtask.

5. The intelligent feeding control method for rice-frog co-cultivation as claimed in claim 1, characterized in that: In S3, each subtask in the current feeding task is executed according to the execution plan, including: Before moving to the starting feeding position of each subtask, obtain weather data in real time; Based on the weather data, it is evaluated whether the current weather conditions will have an adverse impact on the feeding operation. If the evaluation result is that it will have an adverse impact on the feeding operation, the feeding speed of the next subtask will be adjusted to 0. If the evaluation result is that it will not have an adverse impact on the feeding operation, the feeding speed of the next subtask will remain unchanged.

6. The intelligent feeding control method for rice-frog co-culture according to claim 1, characterized in that: The following steps are also included: Select several points on the moving path, set an RFID tag for each point and calibrate the position of each RFID tag; In the process of executing each subtask, when moving to the position of each RFID tag, the current positioning data is corrected to the position of the corresponding RFID tag.

7. An intelligent feeding control system for rice-frog co-culture, characterized in that: The invention comprises a feeding robot, a ground track device, a feeding tower, a charging device and a cloud controller, wherein the feeding robot is located on the ground track device and moves along the track of the ground track device, the feeding robot, the feeding tower and the charging device are all communicatively connected with the cloud controller, and the feeding robot is used to execute the intelligent feeding control method for rice-frog co-culture as described in any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, an intelligent feeding control method for rice-frog co-culture as described in any one of claims 1 to 6 is implemented.

Images