An intelligent animal husbandry method and system for preventing overgrazing of grassland

Abstract

The present invention discloses an intelligent animal husbandry method and system for preventing overgrazing of grasslands, the method comprising obtaining forage information on grasslands; dividing the grasslands into a plurality of areas according to the forage information on the grasslands and configuring area information for each area; obtaining environmental parameters within a preset time, inputting the forage information and the environmental parameters into a pre-constructed forage growth prediction model to obtain a forage growth prediction result corresponding to each area, the forage growth prediction result including forage increment; obtaining animal information to be grazed, the animal information including the number of animals, inputting the animal information, the area information and the forage growth prediction result corresponding to each area into a pre-constructed intelligent animal husbandry model, obtaining animal husbandry planning results, the animal husbandry planning results including the number of animals corresponding to each area and forage reduction, the forage reduction corresponding to each area being less than the sum of the corresponding forage amount and forage increment. The present invention can prevent overgrazing of grasslands and unmanned grazing.

Description

Technical Field

[0001] The present invention relates to the fields of artificial intelligence and the Internet of Things, and in particular to an intelligent animal husbandry method and system for preventing overgrazing on grasslands. Background Art

[0002] As a traditional industry, animal husbandry plays an important role in providing dairy products, meat products and leather products to human society.

[0003] As the herd grows, the demand for manual grazing increases, and traditional grazing methods can no longer meet the requirements of modern, efficient and unmanned grazing.

[0004] The disclosure of the above background technology content is only used to assist in understanding the inventive concept and technical solution of the present invention. It does not necessarily belong to the prior art of this patent application, nor does it necessarily provide technical guidance. In the absence of clear evidence that the above content has been disclosed before the filing date of this patent application, the above background technology should not be used to evaluate the novelty and creativity of the present application. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent animal husbandry method and system for preventing overgrazing of grasslands, which can prevent overgrazing of grasslands and eliminate human grazing.

[0006] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0007] An intelligent livestock farming method for preventing grassland overgrazing comprises the following steps:

[0008] Acquire forage grass information on the grassland, wherein the forage grass information includes the type of forage grass, the area, distribution direction and growth status of each forage grass;

[0009] Divide the grassland into several areas according to the forage information on the grassland and configure area information for each area, wherein the area information includes the number, area, direction of the area and the amount of forage currently corresponding to the area;

[0010] Acquire environmental parameters within a preset time, input the forage information and the environmental parameters into a pre-built forage growth prediction model to obtain a forage growth prediction result corresponding to each of the areas, wherein the forage growth prediction result includes a forage increment;

[0011] The animal information to be grazed is obtained, wherein the animal information includes the number of animals, and the animal information, the area information and the grass growth prediction result corresponding to each area are input into a pre-built intelligent animal husbandry model to obtain the animal husbandry planning result, wherein the animal husbandry planning result includes the number of livestock and the grass reduction corresponding to each area, and the grass reduction corresponding to each area is less than the sum of the corresponding grass amount and the grass increase.

[0012] Further, based on any one of the above-mentioned technical solutions or a combination of multiple technical solutions, the preset period is divided into multiple continuous time periods according to the target number of animal husbandry in the preset period, and each of the time periods has a corresponding relationship with the number of animal husbandry in the preset period, and the corresponding relationship is: n The secondary livestock corresponds to the n The time period, n The time period starts from n -1 livestock start time to n The start time of the second livestock raising;

[0013] Using the forage growth prediction model to obtain the forage growth prediction results corresponding to each of the areas in each of the time periods;

[0014] Based on the animal information, the area information and the grass growth prediction results corresponding to each area in each time period, the intelligent animal husbandry model is used to perform prediction planning for each animal husbandry within a preset period and obtain the animal husbandry planning results corresponding to each animal husbandry.

[0015] Further, based on any one of the technical solutions or a combination of multiple technical solutions described above, the intelligent animal husbandry model is used to predict and plan each animal husbandry within a preset period and obtain the animal husbandry planning results corresponding to each animal husbandry, including obtaining the amount of forage corresponding to each area before each animal husbandry, and determining the animal husbandry planning results corresponding to each area according to the forage amount, and the animal husbandry planning results include the number of animal husbandry corresponding to each area, wherein:

[0016] The amount of forage corresponding to each of the areas before the first animal husbandry is the initial amount of forage corresponding to each area and the increase in forage in the time period before the first animal husbandry;

[0017] Not the first livestock n The amount of grass corresponding to each area before the second animal husbandry is calculated by the following formula:

[0018]

[0019] in, n ≥2, For the n The first animal husbandry iThe amount of forage corresponding to each of the above areas, For the n The period before the second livestock raising i The forage increment corresponding to each of the areas, For the n - 1 livestock i The forage reduction corresponding to each of the areas is in a corresponding relationship with the number of livestock.

[0020] Further, based on any one of the technical solutions or a combination of multiple technical solutions described above, the animal information also includes the type of animal and a unit amount of forage consumption, the unit amount of forage consumption is the amount of forage that an animal needs to consume in a livestock farming process, and the unit amount of forage consumption has a corresponding relationship with the type of the animal;

[0021] The forage reduction corresponding to one area is the sum of the forage consumption of multiple animals corresponding to the area, and the forage consumption of one animal is the sum of the unit amounts of forage consumption corresponding to each animal contained in the area.

[0022] Further, based on any one of the technical solutions or a combination of multiple technical solutions described above, the animal information also includes the weight of the animal and a unit amount of forage consumption, the unit amount of forage consumption is the amount of forage that an animal needs to consume in a livestock farming process, and the unit amount of forage consumption has a corresponding relationship with the weight of the animal;

[0023] The forage reduction corresponding to one area is the sum of the unit amounts of forage consumption of each animal corresponding to the area.

[0024] Further, based on any one of the technical solutions or a combination of multiple technical solutions described above, the animal information also includes the type, weight and unit amount of forage consumption of the animal, the unit amount of forage consumption is the amount of forage that an animal needs to consume in one animal husbandry process, and the unit amount of forage consumption has a corresponding relationship with the type and weight of the animal;

[0025] The forage reduction corresponding to one area is the sum of the forage consumption of multiple animals corresponding to the area, and the forage consumption of one animal is the sum of the unit amounts of forage consumption corresponding to each animal contained in the area.

[0026] Furthermore, based on any one of the technical solutions or a combination of multiple technical solutions mentioned above, the animal husbandry planning result also includes the animal husbandry time corresponding to each animal husbandry.

[0027] Furthermore, any one of the above-mentioned technical solutions or a combination of multiple technical solutions further comprises evaluating the growth state of the forage grass, and the better the growth state of the forage grass is, the higher the evaluation value is;

[0028] Based on the animal husbandry planning result, an evaluation value of the growth state of the grass after animal husbandry according to the animal husbandry planning result is predicted, and the predicted evaluation value is not lower than a first threshold.

[0029] Furthermore, based on any one of the technical solutions or a combination of multiple technical solutions mentioned above, it also includes predicting an evaluation value of the growth status of the grass after livestock farming according to the livestock farming planning results based on the livestock farming planning results, and the predicted evaluation value is not higher than a second threshold, and the second threshold is greater than the first threshold.

[0030] Further, based on any one of the above-mentioned technical solutions or a combination of multiple technical solutions, the forage growth prediction model is constructed in the following manner, including:

[0031] Pre-establish a basic prediction model based on artificial neural network;

[0032] Acquire a learning sample set of forage growth data, the learning sample set including the forage information and environmental parameters corresponding to multiple time points, the forage information and environmental parameters corresponding to the same time point are configured with a matching first label;

[0033] The basic prediction model is trained multiple times using a learning sample set of grass growth data and a preset first loss function to obtain an optimized grass growth prediction model. Based on the optimized grass growth prediction model, the environmental parameters within a preset time and the grass information at the initial moment are used as input to obtain the grass information after the preset time.

[0034] Furthermore, based on any one of the technical solutions or a combination of multiple technical solutions mentioned above, the forage growth prediction model also includes a forage quantity calculation module, and the forage quantity calculation module is configured to calculate the forage quantity based on the forage information, and the forage quantity is the product of the forage quantity per unit area and the forage area.

[0035] Further, based on any one of the above-mentioned technical solutions or a combination of multiple technical solutions, the intelligent livestock model is constructed in the following manner, including:

[0036] Pre-establish a basic intelligent animal husbandry model based on artificial neural network;

[0037] Acquire a learning sample set of animal husbandry data, the learning sample set including the animal information, the area information and the forage information corresponding to each animal husbandry, the animal information, the area information and the forage information corresponding to one animal husbandry are configured with a matching second label;

[0038] The basic intelligent animal husbandry model is trained multiple times using a learning sample set of animal husbandry data and a preset second loss function to obtain an optimized intelligent animal husbandry model.

[0039] Furthermore, based on any one of the technical solutions or a combination of multiple technical solutions mentioned above, the forage information on the grassland is obtained in the following manner, including: collecting grassland images by a drone equipped with an image acquisition device, and processing the grassland images using image processing technology to obtain the forage information.

[0040] Further, any one of the above-mentioned technical solutions or a combination of multiple technical solutions further includes executing the animal husbandry planning result using an intelligent animal husbandry drive device;

[0041] The intelligent livestock driving device includes one or more electronic collars for animals to be grazed. A grazing instruction is sent to each electronic collar according to the livestock planning result. The grazing instruction includes target area information. In response to receiving the grazing instruction, the electronic collar drives the animal wearing it to the target area.

[0042] According to another aspect of the present invention, the present invention provides an intelligent animal husbandry system for preventing overgrazing of grasslands, comprising an image acquisition device, a processor and an intelligent animal husbandry driving device, wherein:

[0043] The image acquisition device is configured to acquire grassland images and transmit the images to the processor;

[0044] The processor is configured to obtain grass information on the grassland according to the grassland image, the grass information including grass type, area, distribution direction and growth status of each grass; and divide the grassland into a plurality of areas according to the grass information on the grassland and configure area information for each area, the area information including the number, area, direction and current corresponding grass amount of the area;

[0045] The processor also includes a forage growth prediction module and an intelligent livestock module, wherein:

[0046] The forage growth prediction module has a pre-stored trained forage growth prediction model, and the forage information and pre-acquired environmental parameters are input into the forage growth prediction model to obtain a forage growth prediction result corresponding to each of the areas;

[0047] The intelligent animal husbandry module pre-stores a trained intelligent animal husbandry model, and the animal information to be grazed, the area information and the grass growth prediction result corresponding to each area are input into the intelligent animal husbandry model to obtain the animal husbandry planning result, which includes the number of livestock corresponding to each area;

[0048] The intelligent livestock driving device is configured to receive and execute the livestock planning result.

[0049] Furthermore, based on any one of the technical solutions or a combination of multiple technical solutions mentioned above, an intelligent animal husbandry system for preventing overgrazing of grasslands is provided, wherein the intelligent animal husbandry drive device includes one or more electronic collars for animals to be worn by the animals to be grazed, and grazing instructions are sent to each electronic collar according to the animal husbandry planning results, wherein the grazing instructions include target area information, and the electronic collar drives the animal wearing it to the target area in response to receiving the grazing instruction, and drives the animal wearing it back after each animal husbandry is completed.

[0050] The beneficial effects brought by the technical solution provided by the present invention are as follows:

[0051] a. The present invention determines the amount of grass in each area through the grass information on the grassland, and obtains the grass growth prediction result including the grass increment based on the grass information and environmental information. Based on the fact that the grass reduction corresponding to the grazing in each area is less than the sum of the grass amount and the grass increment, the animal husbandry specification results of each area are intelligently planned in advance, which can realize unmanned animal husbandry management efficiently and intelligently, greatly saving the manpower cost of animal husbandry, and can prevent the grass in each area of ​​the grassland from being over-grazed by animals due to excessive grazing, which can effectively prevent the occurrence of overgrazing on the grassland;

[0052] b. The present invention evaluates the growth state of the forage, and predicts the evaluation value of the growth state of the forage after the forage is raised according to the forage planning result based on the forage planning result, and the predicted evaluation value is not lower than a first threshold value, and the predicted evaluation value is not higher than a second threshold value, and the second threshold value is greater than the first threshold value, so that grazing can be planned more reasonably, so that the forage in each area will not be excessively eaten due to overgrazing, and will not be missed due to human negligence;

[0053] c. The present invention determines the unit amount of grass consumption of each animal based on the animal species and weight, and can obtain accurate grass reduction in combination with the number, thereby making it possible to plan grazing results for each area more reasonably and reliably. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1A flow chart of a first intelligent grazing method provided for an exemplary embodiment of the present invention;

[0056] Figure 2 A flow chart of an intelligent grazing method for multiple grazing planning is provided for an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0058] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product or equipment that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0059] In one embodiment of the present invention, an intelligent animal husbandry method for preventing grassland overgrazing is provided, see Figure 1 , the intelligent animal husbandry method comprises the following steps:

[0060] Acquire forage grass information on the grassland, wherein the forage grass information includes the type of forage grass, the area, distribution direction and growth status of each forage grass;

[0061] Divide the grassland into several areas according to the forage information on the grassland and configure area information for each area, wherein the area information includes the number, area, direction of the area and the amount of forage currently corresponding to the area;

[0062] Acquire environmental parameters within a preset time, the environmental parameters including geographic location, soil information, light, temperature, rainfall, humidity, etc., input the forage information and the environmental parameters into a pre-built forage growth prediction model to obtain a forage growth prediction result corresponding to each of the areas, the forage growth prediction result including forage increment;

[0063] The animal information to be grazed is obtained, the animal information includes the number of animals, the animal information, the area information and the grass growth prediction result corresponding to each area are input into the pre-built intelligent animal husbandry model, and the animal husbandry planning result is obtained, the animal husbandry planning result includes the number of animals and grass reduction corresponding to each area, and the grass reduction corresponding to each area is less than the sum of the corresponding grass amount and the grass increment. Among them, the grass reduction of a area is the total grass consumption of all animals to be grazed allocated to the area.

[0064] After obtaining the animal husbandry planning results, the corresponding number of animals to be grazed can be driven to the corresponding areas directly according to the animal husbandry planning results. Specifically, the animal husbandry planning results can be executed using an intelligent animal husbandry drive device. The intelligent animal husbandry drive device includes one or more electronic collars for the animals to be grazed, and sends grazing instructions to each electronic collar according to the animal husbandry planning results. The grazing instructions include target area information, and the electronic collar drives the animal wearing it to the target area in response to receiving the grazing instructions. It should be noted that the animals to be grazed in the present application are all grazing objects that have been pre-trained and equipped with the electronic collars, so that the animals to be grazed will move to the corresponding directions / areas after being trained to obtain different driving actions of the electronic collars.

[0065] Preferably, the animal husbandry planning result also includes the animal husbandry duration corresponding to each animal husbandry. After each grazing is completed, the grazing instruction also includes the animal husbandry duration. The electronic collar is configured with a timer and a drive module. After the electronic collar receives the grazing instruction, the timer starts timing and the drive module starts to drive the animal to the target area. When the timing duration reaches the animal husbandry duration, the timer sends a return signal to the drive module, and the drive module drives the animal wearing it to return in response to receiving the return signal. In this way, unmanned animal husbandry management can be realized efficiently and intelligently, which greatly saves the manpower cost of animal husbandry and improves the artificial intelligence in the field of animal husbandry. And it can intelligently control the grass in each area of ​​the grassland so that it is not over-eaten by grazing animals, which can effectively prevent the occurrence of overgrazing on the grassland.

[0066] The grassland information is obtained by collecting grassland images through a drone equipped with an image acquisition device, and processing the grassland images using image processing technology. The grassland images are processed and calculated to obtain the grass type, the area of ​​each type of grass, the distribution direction of each type of grass, and the growth status of grass in each area of ​​the grassland.

[0067] In one embodiment of the present invention, see Figure 2The intelligent animal husbandry method for preventing overgrazing of grasslands can predict and plan multiple animal husbandry within a preset period at one time, and then each grazing within the preset period only needs to be executed according to the corresponding animal husbandry planning results each time. In this way, labor costs and physical costs can be further saved.

[0068] In this embodiment, if Figure 2 The method flow chart shown in the figure executes the intelligent animal husbandry method, and divides the preset period into a plurality of continuous time periods according to the target animal husbandry times within the preset period, and each of the time periods has a corresponding relationship with the animal husbandry times within the preset period, and the corresponding relationship is: n The secondary livestock corresponds to the n The time period, n The time period starts from n -1 livestock start time to n The start time of the second livestock farming.

[0069] The forage growth prediction model is used to obtain the forage growth prediction results corresponding to each of the areas in each of the time periods, and the forage growth prediction results at least include the forage increment.

[0070] Based on the animal information, the area information and the grass growth prediction results corresponding to each of the areas within each of the time periods, the intelligent animal husbandry model is used to perform prediction planning for each animal husbandry within a preset period and obtain animal husbandry planning results corresponding to each animal husbandry, including obtaining the amount of grass corresponding to each of the areas before each animal husbandry, and determining the animal husbandry planning results corresponding to each area based on the amount of grass and the grass growth prediction results within the time period corresponding to the animal husbandry, and the animal husbandry planning results include the number of livestock corresponding to each area.

[0071] In this embodiment, the amount of grass corresponding to each area before the first livestock grazing is the initial amount of grass corresponding to each area and the grass increment in the time period before the first livestock grazing. n The amount of grass corresponding to each area before the second animal husbandry is calculated by the following formula:

[0072]

[0073] in, n ≥2, For the n The first animal husbandry i The amount of forage corresponding to each of the above areas, For the n The period before the second livestock raising i The forage increment corresponding to each of the areas, For the n - 1 livestocki The forage reduction corresponding to each of the areas is in a corresponding relationship with the number of livestock.

[0074] The intelligent animal husbandry method for preventing overgrazing of grasslands provided by this embodiment can perform artificial intelligence planning for animal husbandry within a preset time of one week, one month or even half a year at a time. The environmental data can be captured based on network data. For long-term animal husbandry planning, since the accuracy of environmental parameters is lower in the future, in order to ensure the reliability and accuracy of the planning results, environmental data can be captured every week or every day or even several times a day, and the data can be continuously updated iteratively, and then iteratively obtained. More accurate and reliable animal husbandry planning results. For example, the initial forage information on the grassland is obtained through aerial photography, and the grassland is divided into several areas according to the forage information and the area information is configured for each area. Then, the animal husbandry planning results of each area corresponding to all animal husbandry times within half a year can be planned. However, each subsequent time based on the newly captured environmental data, the animal husbandry planning results corresponding to each animal husbandry time can be continuously updated iteratively. Therefore, even for long-term animal husbandry planning, the reliability of its planning results can be guaranteed.

[0075] In order to accurately predict the grass reduction corresponding to each livestock area each time, it is necessary to accurately calculate the total grass consumption of the grazing animals planned in each area, which is explained below through three specific embodiments.

[0076] In one embodiment of the present invention, there is only one type of animal to be grazed, and the animal information is determined to further include the weight of the animal and the unit amount of forage consumption. The unit amount of forage consumption is the amount of forage that an animal needs to consume during one animal husbandry process. The unit amount of forage consumption has a corresponding relationship with the weight of the animal, that is, q k =f ( W k ),in, q k For the k The unit amount of grass consumed by each animal, W k For the k The weight of the animal, f ()express q k and W k The grass reduction corresponding to a patch is the sum of the grass consumption units of each animal in the patch, that is, , where m is the n - 1 livestock i The number of animals obtained by planning the area.

[0077] In another embodiment of the present invention, there are multiple animal species to be grazed, and the effect of the weight of the same animal on the grass consumption is not considered. In this embodiment, the animal information also includes the type of animal and the unit amount of grass consumption. The unit amount of grass consumption is the amount of grass that an animal needs to consume in a livestock farming process, and the unit amount of grass consumption has a corresponding relationship with the type of animal. The grass reduction corresponding to a section is the sum of the grass consumption of multiple animals corresponding to the section, and the grass consumption of one animal is the sum of the unit amounts of grass consumption corresponding to each animal contained therein.

[0078] In another embodiment of the present invention, there are multiple animal species to be grazed, and the effect of the weight of the same animal on the grass consumption is considered. In this embodiment, the animal information also includes the animal species, weight and unit amount of grass consumption, the unit amount of grass consumption is the amount of grass that an animal needs to consume in one animal husbandry process, and the unit amount of grass consumption has a corresponding relationship with the animal species and weight. The grass reduction corresponding to one area is the sum of the grass consumption of multiple animals corresponding to the area, and the grass consumption of one animal is the sum of the unit amounts of grass consumption corresponding to each animal contained therein.

[0079] In one embodiment of the present invention, in order to further protect grasslands and prevent overgrazing, a more reasonable animal husbandry planning method is proposed. In this embodiment, the intelligent animal husbandry method for preventing overgrazing of grasslands also includes evaluating the growth state of the forage, and the better the growth state of the forage, the higher the evaluation value. Specifically, the growth state of the forage can be evaluated by scoring according to the growth state of the forage. For example, for any kind of forage, it is evaluated as 100 points when it grows to the most vigorous state, and if the forage is not visible on the surface, its growth state is evaluated as 0 points. The growth state of the forage can be evaluated as a corresponding score within 0-100 points, and the growth state of the forage can be obtained by image acquisition and processing. Of course, other evaluation methods can also be used to evaluate the growth state of the forage, for example, the evaluation level can be divided according to the growth height of the forage and the color of the forage. The present invention does not limit the scope of protection of this application by the specific evaluation method of the growth state of the forage.

[0080] In this embodiment, the evaluation value of the growth state of the grass after livestock raising according to the livestock planning result is predicted based on the livestock planning result, and the predicted evaluation value is not lower than the first threshold. And / or, the predicted evaluation value is not higher than the second threshold, and the second threshold is greater than the first threshold. For example, if the first threshold is 20 points and the second threshold is 60 points, then through this planning method, the growth state of the grass in each area of ​​the grassland can be controlled between 20 points and 60 points, so that more reasonable grazing can be achieved, and the grass in each area will not be over-grazed due to overgrazing, and will not be missed due to human negligence.

[0081] In one embodiment of the present invention, the forage growth prediction model is constructed in the following manner, including:

[0082] Pre-establish a basic prediction model based on artificial neural network;

[0083] Acquire a learning sample set of forage growth data, the learning sample set including the forage information and environmental parameters corresponding to multiple time points, the forage information and environmental parameters corresponding to the same time point are configured with a matching first label;

[0084] The basic prediction model is trained multiple times using a learning sample set of grass growth data and a preset first loss function to obtain an optimized grass growth prediction model. Based on the optimized grass growth prediction model, the environmental parameters within a preset time and the grass information at the initial moment are used as input to obtain the grass information after the preset time.

[0085] In this embodiment, the forage growth prediction model further includes a forage quantity calculation module, and the forage quantity calculation module is configured to calculate the forage quantity according to the forage information, and the forage quantity is the product of the forage quantity per unit area and the forage area.

[0086] In one embodiment of the present invention, the intelligent livestock model is constructed in the following manner, including:

[0087] Pre-establish a basic intelligent animal husbandry model based on artificial neural network;

[0088] Acquire a learning sample set of animal husbandry data, the learning sample set including the animal information, the area information and the forage information corresponding to each animal husbandry, the animal information, the area information and the forage information corresponding to one animal husbandry are configured with a matching second label;

[0089] The basic intelligent animal husbandry model is trained multiple times using a learning sample set of animal husbandry data and a preset second loss function to obtain an optimized intelligent animal husbandry model.

[0090] It should be noted that it is a better choice to obtain the intelligent animal husbandry model by training the artificial neural network learning model. Another way is to establish the intelligent animal husbandry model by mathematical means, that is, the intelligent animal husbandry model is a mathematical model, wherein the independent variables are the animal information with grazing, the amount of forage in each area and the increment of forage, etc., the animal information includes animal species, weight and unit consumption of forage, etc., and the variables are the animal husbandry planning results of each area, including the number of grazing animals planned for each area. Through this mathematical model, the forage planning results of each area can be generally achieved. It should be noted that the logic in the animal husbandry planning process is that for each area on the grassland, the sum of its forage amount and forage increment is first determined, combined with the minimum amount of forage remaining after each animal husbandry set for it, and the difference between the sum of the forage amount and the forage increment and the minimum amount of forage is calculated. The difference is the forage reduction planned for the area this time. According to the relationship between the forage reduction and the animal species, weight and number, it is easy to get the animal husbandry planning results.

[0091] In one embodiment of the present invention, an intelligent animal husbandry system for preventing overgrazing of grassland is proposed, comprising an image acquisition device, a processor and an intelligent animal husbandry driving device, wherein:

[0092] The image acquisition device is configured to acquire grassland images and transmit the images to the processor;

[0093] The processor is configured to obtain grass information on the grassland according to the grassland image, the grass information including grass type, area, distribution direction and growth status of each grass; and divide the grassland into a plurality of areas according to the grass information on the grassland and configure area information for each area, the area information including the number, area, direction and current corresponding grass amount of the area;

[0094] The processor also includes a forage growth prediction module and an intelligent livestock module, wherein:

[0095] The forage growth prediction module has a pre-stored trained forage growth prediction model, and the forage information and pre-acquired environmental parameters are input into the forage growth prediction model to obtain a forage growth prediction result corresponding to each of the areas;

[0096] The intelligent animal husbandry module pre-stores a trained intelligent animal husbandry model, and the animal information to be grazed, the area information and the grass growth prediction result corresponding to each area are input into the intelligent animal husbandry model to obtain the animal husbandry planning result, which includes the number of livestock corresponding to each area;

[0097] The intelligent livestock driving device is configured to receive and execute the livestock planning result.

[0098] In this embodiment, the intelligent livestock driving device includes one or more electronic collars for animals to be grazed. A grazing instruction is sent to each electronic collar according to the livestock planning result. The grazing instruction includes target area information. The electronic collar drives the animal wearing it to the target area in response to receiving the grazing instruction, and drives the animal wearing it back after each livestock grazing is completed.

[0099] It should be noted that the embodiment of the intelligent animal husbandry system for preventing overgrazing of grasslands and the embodiment of the intelligent animal husbandry method for preventing overgrazing of grasslands belong to the same inventive concept, and the entire content of the embodiment of the intelligent animal husbandry method for preventing overgrazing of grasslands is incorporated into the embodiment of the intelligent animal husbandry system for preventing overgrazing of grasslands by reference.

[0100] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0101] The above is only a specific implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. An intelligent animal husbandry method for preventing overgrazing of grasslands, characterized in that: The following steps are involved: Acquire forage grass information on the grassland, wherein the forage grass information includes the type of forage grass, the area, distribution direction and growth status of each forage grass; Divide the grassland into several areas according to the forage information on the grassland and configure area information for each area, wherein the area information includes the number, area, direction of the area and the amount of forage currently corresponding to the area; Acquire environmental parameters within a preset time, input the forage information and the environmental parameters into a pre-built forage growth prediction model to obtain a forage growth prediction result corresponding to each of the areas, wherein the forage growth prediction result includes a forage increment; Obtaining information about animals to be grazed, the animal information including the number of animals, inputting the animal information, the area information and the grass growth prediction result corresponding to each area into a pre-built intelligent animal husbandry model to obtain animal husbandry planning results, the animal husbandry planning results including the number of animals and grass reduction corresponding to each area, and the grass reduction corresponding to each area is less than the sum of the corresponding grass amount and the grass increment; The preset period is divided into a plurality of continuous time periods according to the target number of animal husbandry in the preset period, and each of the time periods has a corresponding relationship with the number of animal husbandry in the preset period, and the corresponding relationship is: n The secondary livestock corresponds to the n The time period, n The time period starts from n -1 livestock start time to n The start time of the second livestock raising; Using the forage growth prediction model to obtain the forage growth prediction results corresponding to each of the areas in each of the time periods; Based on the animal information, the area information and the grass growth prediction results corresponding to each area in each time period, the intelligent animal husbandry model is used to predict and plan each animal husbandry within a preset period and obtain the animal husbandry planning results corresponding to each animal husbandry; After obtaining the animal husbandry planning results of each area corresponding to all animal husbandry times within a preset period, the animal husbandry planning results corresponding to each animal husbandry time are continuously updated iteratively based on the newly captured environmental data each time; The forage growth prediction model is constructed in the following manner, including: Pre-establish a basic prediction model based on artificial neural network; Acquire a learning sample set of forage growth data, the learning sample set including the forage information and environmental parameters corresponding to multiple time points, the forage information and environmental parameters corresponding to the same time point are configured with a matching first label; The basic prediction model is trained multiple times using a learning sample set of grass growth data and a preset first loss function to obtain an optimized grass growth prediction model. Based on the optimized grass growth prediction model, the environmental parameters within a preset time and the grass information at the initial moment are used as input to obtain the grass information after the preset time. The intelligent livestock model is constructed in the following manner, including: Pre-establish a basic intelligent animal husbandry model based on artificial neural network; Acquire a learning sample set of animal husbandry data, the learning sample set including the animal information, the area information and the forage information corresponding to each animal husbandry, the animal information, the area information and the forage information corresponding to one animal husbandry are configured with a matching second label; The basic intelligent animal husbandry model is trained multiple times using a learning sample set of animal husbandry data and a preset second loss function to obtain an optimized intelligent animal husbandry model.

2. The intelligent animal husbandry method for preventing grassland overgrazing according to claim 1, characterized in that: The intelligent animal husbandry model is used to predict and plan each animal husbandry within a preset period and obtain the animal husbandry planning results corresponding to each animal husbandry, including obtaining the amount of forage corresponding to each area before each animal husbandry, and determining the animal husbandry planning results corresponding to each area according to the amount of forage, and the animal husbandry planning results include the number of animal husbandry corresponding to each area, wherein: The amount of forage corresponding to each of the areas before the first animal husbandry is the initial amount of forage corresponding to each area and the increase in forage in the time period before the first animal husbandry; Not the first livestock n The amount of grass corresponding to each area before the second animal husbandry is calculated by the following formula: ； in, n ≥2, No. n The first animal husbandry i The amount of forage corresponding to each of the above areas, For the n The period before the second livestock raising i The forage increment corresponding to each of the areas, For the n - 1 livestock i The forage reduction corresponding to each of the areas is in a corresponding relationship with the number of livestock.

3. The intelligent animal husbandry method for preventing grassland overgrazing according to claim 2, characterized in that: The animal information also includes the type of animal and the unit amount of forage consumption, wherein the unit amount of forage consumption is the amount of forage that an animal needs to consume during one animal husbandry process, and the unit amount of forage consumption has a corresponding relationship with the type of animal; The forage reduction corresponding to one area is the sum of the forage consumption of multiple animals corresponding to the area, and the forage consumption of one animal is the sum of the unit amounts of forage consumption corresponding to each animal contained in the area.

4. The intelligent animal husbandry method for preventing grassland overgrazing according to claim 2, characterized in that: The animal information also includes the weight of the animal and the unit amount of forage consumption, wherein the unit amount of forage consumption is the amount of forage that an animal needs to consume during one animal husbandry process, and the unit amount of forage consumption has a corresponding relationship with the weight of the animal; The forage reduction corresponding to one area is the sum of the unit amounts of forage consumption of each animal corresponding to the area.

5. The intelligent animal husbandry method for preventing grassland overgrazing according to claim 2, characterized in that: The animal information also includes the type, weight and unit amount of forage consumption of the animal, where the unit amount of forage consumption is the amount of forage that an animal needs to consume during one animal husbandry process, and the unit amount of forage consumption has a corresponding relationship with the type and weight of the animal; The forage reduction corresponding to one area is the sum of the forage consumption of multiple animals corresponding to the area, and the forage consumption of one animal is the sum of the unit amounts of forage consumption corresponding to each animal contained in the area.

6. The intelligent animal husbandry method for preventing grassland overgrazing according to claim 1, characterized in that: The animal husbandry planning result also includes the animal husbandry duration corresponding to each animal husbandry.

7. The intelligent animal husbandry method for preventing grassland overgrazing according to claim 1, characterized in that: The method further comprises evaluating the growth state of the forage grass, wherein the better the growth state of the forage grass is, the higher the evaluation value thereof is; Based on the animal husbandry planning result, an evaluation value of the growth state of the grass after animal husbandry according to the animal husbandry planning result is predicted, and the predicted evaluation value is not lower than a first threshold.

8. The intelligent animal husbandry method for preventing grassland overgrazing according to claim 7, characterized in that: It also includes predicting the evaluation value of the growth status of the grass after animal husbandry according to the animal husbandry planning results based on the animal husbandry planning results, and the predicted evaluation value is not higher than a second threshold, and the second threshold is greater than the first threshold.

9. The intelligent animal husbandry method for preventing grassland overgrazing according to claim 1, characterized in that: The forage growth prediction model also includes a forage quantity calculation module, which is configured to calculate the forage quantity according to the forage information, and the forage quantity is the product of the forage quantity per unit area and the forage area.

10. The intelligent animal husbandry method for preventing grassland overgrazing according to claim 1, characterized in that: The forage information on the grassland is obtained in the following manner, including: collecting grassland images by using a drone equipped with an image acquisition device, and obtaining the forage information by processing the grassland images using image processing technology.

11. An intelligent animal husbandry system for preventing overgrazing of grasslands, characterized in that: The intelligent animal husbandry method for preventing grassland overgrazing according to any one of claims 1 to 10 comprises an image acquisition device, a processor and an intelligent animal husbandry driving device, wherein: The image acquisition device is configured to acquire grassland images and transmit the images to the processor; The processor is configured to obtain grass information on the grassland according to the grassland image, the grass information including grass type, area, distribution direction and growth status of each grass; and divide the grassland into a plurality of areas according to the grass information on the grassland and configure area information for each area, the area information including the number, area, direction and current corresponding grass amount of the area; The processor also includes a forage growth prediction module and an intelligent livestock module, wherein: The forage growth prediction module has a pre-stored trained forage growth prediction model, and the forage information and pre-acquired environmental parameters are input into the forage growth prediction model to obtain a forage growth prediction result corresponding to each of the areas; The intelligent animal husbandry module pre-stores a trained intelligent animal husbandry model, and the animal information to be grazed, the area information and the grass growth prediction result corresponding to each area are input into the intelligent animal husbandry model to obtain the animal husbandry planning result, which includes the number of livestock corresponding to each area; The intelligent livestock driving device is configured to receive and execute the livestock planning result.

Images