Optimized regulation and control method for grassland livestock breeding

Abstract

The invention discloses an optimal regulation and control method for grassland livestock breeding, and the method comprises the steps: collecting the remote sensing data of a grassland through a satellite, and collecting the ground data of the grassland through a ground sensor; fusing the remote sensing data and the ground data, and constructing a three-dimensional growth model of the grassland; acquiring physiological indexes of the livestock through an information acquisition sensor of the livestock; constructing a grazing planning model based on the three-dimensional growth model and the physiological indexes; and determining a grazing path of the livestock according to the grazing planning model. Therefore, the influence of various factors on grassland animal husbandry can be comprehensively considered, and the accuracy of risk management and control is improved.

Description

Technical Field

[0001] The present application relates to the field of animal husbandry and computer technology, and in particular to an optimization and control method for grassland animal husbandry. Background Art

[0002] The optimization and regulation methods of grassland animal husbandry mainly refer to the use of scientific and reasonable management means to regulate the effects of grass growth, soil nutrients, weather changes, animal physiological status and water source distribution on grassland ecosystems and livestock breeding benefits.

[0003] However, in actual operations, herders are generally required to make adjustments based on their own experience, and are unable to comprehensively consider the impact of multiple factors on grassland animal husbandry, resulting in low accuracy of risk control. Summary of the Invention

[0004] The technical problem to be solved by this application is how to provide an optimization and regulation method for grassland animal husbandry, which can comprehensively consider the impact of various factors on grassland animal husbandry and improve the accuracy of risk management.

[0005] To solve the above problems, in a first aspect, the present application provides a method for optimizing and controlling grassland animal husbandry, the method comprising: collecting remote sensing data of grassland through satellites, and collecting ground data of the grassland through ground sensors; fusing the remote sensing data and the ground data to construct a three-dimensional growth model of the grassland; obtaining the physiological indicators of the livestock through livestock information collection sensors; constructing a grazing planning model based on the three-dimensional growth model and the physiological indicators; and determining the grazing path of the livestock according to the grazing planning model.

[0006] In an optional embodiment, the grazing planning model uses a constraint satisfaction algorithm to determine the grazing path of the livestock; wherein the grazing path satisfies at least one of the following constraints: ecological carrying capacity constraint; terrain safety constraint; livestock movement cost constraint; disease isolation constraint.

[0007] In an optional embodiment, the grassland is divided into a plurality of grid units, and the ecological carrying capacity constraint satisfies the following equation:

[0008]

[0009] Wherein, i and j are the grid unit coordinates of the grassland, i is the row coordinate and j is the column coordinate; is the number of times the grid unit (i, j) is effectively utilized during the monitoring period; T recovery (i, j) is the ecological restoration period of grid cell (i, j); K is the number of time segments in the monitoring period; and t is the time slice index (t = 1, 2, ..., K).

[0010] In an optional embodiment, the livestock movement cost constraint satisfies the following equation:

[0011]

[0012] Among them, E move is the basic exercise energy consumption, R terrain is the terrain resistance correction parameter, C env is the environmental stress correction parameter, E max,daily is the maximum tolerable daily energy consumption of livestock, and a is the path segment index (a=1, 2,…, A).

[0013] In an optional embodiment, the terrain safety constraint condition includes: marking areas with a slope greater than 25 degrees as grazing-prohibited areas.

[0014] In an optional embodiment, the disease isolation constraint condition includes: when the physiological indicators of the livestock are monitored to exceed preset values, the grid unit where the livestock is located is used as an isolation area.

[0015] In an optional embodiment, the grassland is divided into multiple areas, and the grazing path of the livestock is determined according to the grazing planning model, including: monitoring the distribution amount of grass in different areas through the grazing planning model, and calculating the optimal grazing density of each area based on the distribution amount; adjusting the grazing time of each area based on the growth cycle of the grass and the water source distribution density in each area; if the water source distribution density D satisfies D≤D_max, then limiting the daily grazing time of the area to less than T_min, where D represents the number of water sources per unit area, D_max is a preset water source density threshold, and T_min is the set minimum daily grazing time.

[0016] In an optional embodiment, the remote sensing data and the ground data are fused through an edge computing gateway, and the edge computing gateway performs at least one of the following operations: outlier filtering, data spatiotemporal alignment, and data feature extraction.

[0017] In an optional embodiment, after determining the grazing path of the livestock according to the grazing planning model, the method further includes: controlling the smart electric fence in the pasture to be turned on or off based on the grazing path.

[0018] In an optional embodiment, the method further includes: when it is predicted that the grass gap is greater than a preset value, starting an automatic refeeding function.

[0019] In a second aspect, the present application also provides an optimization and control device for grassland animal husbandry, including: a grassland data acquisition module, used to collect remote sensing data of the grassland through satellites, and to collect ground data of the grassland through ground sensors; a grassland model construction module, used to fuse the remote sensing data and the ground data to construct a three-dimensional growth model of the grassland; a livestock data acquisition module, used to obtain the physiological indicators of the livestock through livestock information acquisition sensors; a grazing planning model construction module, used to construct a grazing planning model based on the three-dimensional growth model and the physiological indicators; and a model analysis module, used to determine the grazing path of the livestock according to the grazing planning model.

[0020] In a third aspect, the present application further provides a computer device comprising a processor and a storage device, wherein the storage device is suitable for storing a plurality of program codes, and the program codes are suitable for being loaded and run by the processor to execute any of the methods described above.

[0021] In a fourth aspect, the present application further provides a storage medium storing a plurality of program codes, wherein the program codes are suitable for being loaded and run by a processor to execute any of the methods described above.

[0022] Compared with the prior art, the technical solution of the embodiment of the present application has the following beneficial effects:

[0023] In the optimization and control method for grassland animal husbandry in the embodiment of the present application, a big data model or a neural network model can be constructed based on the three-dimensional growth model of the grassland and the physiological indicators of the livestock to analyze the grass growth and water source information of the grassland, and allocate suitable grazing paths for the livestock so that the livestock can graze along the grazing path. In this way, the grazing path can be considered comprehensively considering a large number of different factors and the risks encountered during the grazing process can be avoided as much as possible.

[0024] Furthermore, factors such as the growth cycle of grass can be combined in the grazing planning model to plan different grazing paths in different time periods, thereby achieving sustainable growth of grass in the pasture. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A schematic flow chart of an optimization and control method for grassland animal husbandry according to an embodiment of the present application;

[0026] Figure 2 For the embodiment of this application Figure 1 A flow chart of a specific implementation of step S105;

[0027] Figure 3 This is a structural schematic diagram of an optimization and control device for grassland animal husbandry according to an embodiment of the present application. DETAILED DESCRIPTION

[0028] As mentioned in the background technology, in actual operations, herders are generally required to make adjustments based on their own experience, and are unable to comprehensively consider the impact of multiple factors on grassland animal husbandry, resulting in low accuracy of risk control.

[0029] In order to further illustrate the technical means and effects adopted by this application to achieve the predetermined application purpose, the specific implementation method and effects of the optimized regulation method for grassland animal husbandry proposed in this application are described in detail below in combination with the accompanying drawings and preferred embodiments.

[0030] According to the first aspect of this application, see Figure 1 , Figure 1 The figure is a flow chart of an optimization and control method for grassland animal husbandry, which includes the following steps S101, S102, S103, S104 and S105. Each step is described in detail below.

[0031] Step S101 : collecting remote sensing data of the grassland through satellites, and collecting ground data of the grassland through ground sensors.

[0032] Remote sensing technology can capture large-area imagery from high altitudes, including information on vegetation coverage, soil moisture, water distribution, and topographical undulations. Furthermore, satellite-mounted multispectral and thermal imaging sensors can accurately identify the height and health of pasture grasses.

[0033] Ground-based data requires IoT devices, such as soil nutrient monitoring stations, precipitation gauges, and vegetation density sensors, deployed in key areas. For example, to obtain precise data, portable environmental monitoring modules with wireless transmission can be deployed at multiple sampling points, using a geographic information system to locate each data source. This ensures diverse and complementary data sources.

[0034] Step S102: Fusing the remote sensing data and the ground data to construct a three-dimensional growth model of the grassland.

[0035] The second critical step is to integrate these two sets of heterogeneous data to create a three-dimensional grassland growth model. This stage involves converting the data at different resolutions to a consistent spatial scale, and using georeferencing and data noise reduction to improve its usability.

[0036] Specifically, deep learning architectures within the field of artificial intelligence can be used to extract complex nonlinear relationships, ultimately forming a mathematical representation of grassland ecosystem evolution using time as a dynamic parameter. This constructed three-dimensional growth model not only predicts grassland resource conditions within a specific timeframe, but also allows ranch managers to intuitively understand the changing spatiotemporal distribution patterns of the entire ecosystem.

[0037] For example, in one embodiment, it is found that vegetation in a certain plot of land has grown rapidly due to abundant rainfall in the early stage. In this area, arrangements can be made in advance to appropriately increase the frequency of livestock visits to prevent the problem of increased plant aging fiber affecting palatability.

[0038] Step S103: Acquire the physiological indicators of the livestock through livestock information collection sensors.

[0039] In addition to considering pasture information, relevant biological parameters from livestock are also needed to support overall strategy development. Modern livestock farms are increasingly using wearable vital sign monitors such as heart rate monitors and GPS collars to track individual animal behavior and internal health status as physiological indicators.

[0040] Based on the processing of a large number of raw signals with the assistance of the big data platform, results such as estimated activity energy consumption are obtained and fed back to managers to help determine whether individuals are on a normal developmental track. It can also reveal potential interference factors in social competition relationships when living in groups.

[0041] For example, suppose a flock of sheep has recently frequently moved away from the central drinking water point. Through in-depth data analysis, it can be confirmed that the reason may involve a reshuffle of internal trophic levels or that hot weather has caused foraging preferences to shift to cool and shaded areas rather than directly near the water source.

[0042] Step S104: constructing a grazing planning model based on the three-dimensional growth model and the physiological indicators.

[0043] Step S105: determining the livestock grazing path according to the grazing planning model.

[0044] Based on the three-dimensional growth model of the grassland and the physiological indicators of livestock, a big data model or neural network model can be constructed to analyze the grassland's grass growth and water source information, and to assign appropriate grazing paths for livestock to graze along. This allows for comprehensive consideration of a wide range of factors and minimizes risks encountered during grazing. Furthermore, the grazing planning model can incorporate factors such as the grass growth cycle to plan different grazing paths for different time periods, thereby achieving sustainable grass growth on the pasture.

[0045] In an optional embodiment, the grazing planning model can be constructed based on a long short-term memory (LSTM) network, whose structural design must closely integrate the characteristics of time series data (such as weather changes, grassland growth, and livestock behavior). LSTM can include: incorporating prior knowledge of grassland recovery cycles into the forget gate (for example, to determine whether to forget overload records from three days ago); integrating real-time weather warnings into the input gate as gating conditions (for example, controlling the influence weight of new rainfall data); introducing a spatial attention mechanism into the candidate state (for example, highlighting changes in the Normalized Difference Vegetation Index (NDVI) in steep slope areas); and associating livestock physiological cycle characteristics into the output gate (for example, adjusting the movement prediction of lactating sheep).

[0046] Optionally, LSTM can be designed with multi-time-scale memory units, which may include: fast memory units to handle hourly changes (such as temperature drops, etc.); slow memory units to track monthly / quarterly trends (such as inter-annual grassland degradation, etc.).

[0047] Optionally, LSTM can also include a spatiotemporal feature fusion layer, thereby combining the spatial characteristics of the grassland's three-dimensional growth model with the characteristics of the time series data. By fusing spatiotemporal features and embedding animal husbandry knowledge, the reliability of predictions in complex pasture environments is significantly improved. In actual deployment, it can be combined with an online learning mechanism to update model parameters every quarter to adapt to environmental changes.

[0048] In an optional embodiment, the grazing planning model uses a constraint satisfaction algorithm to determine the grazing path of the livestock; wherein the grazing path satisfies at least one of the following constraints: ecological carrying capacity constraint; terrain safety constraint; livestock movement cost constraint; disease isolation constraint.

[0049] The ecological carrying capacity constraint states that the total number of livestock in any unit area must not exceed the environmental support threshold, Cmax. For example, in some grasslands, Cmax = 10 livestock per square kilometer. This constraint ensures that grassland vegetation will not be irreversibly damaged by excessive trampling.

[0050] Terrain safety constraints require that grazing paths avoid dangerous areas such as high slopes or deep valleys, and a threshold angle α≤α_threshold (α is the surface inclination, i.e., the slope) can be set.

[0051] To reduce energy consumption, the cost of livestock movement is limited to a certain range, that is, d <= d_max (d is the Euclidean distance between two points, indicating the maximum movement distance between each step on the path, usually set to the capacity limit of livestock single movement, for example, the average walking capacity of cattle is 3 kilometers / day).

[0052] Finally, epidemic isolation conditions require that routes avoid areas where epidemics have been reported.

[0053] In an optional embodiment, when constructing a grazing planning model, the core variable that needs to be solved is the livestock grazing path. In this process, the path is regarded as a continuous function f(x, t) in a discrete space, where x represents the spatial position (such as different grid areas of grassland) and t is a time parameter. The value of f(x, t) represents whether a certain time and place is suitable as a path node for livestock movement. This model abstracts the basic form of the path generation problem through a space-time framework, thereby facilitating the subsequent constraint application and algorithm calculation.

[0054] When initially setting the variable range, the set of all possible grazing points, P = {p1, p2, ..., pn}, must be sampled in advance, where n represents the total sample size and pi (i = 1, 2, ..., n) corresponds to a specific latitude and longitude. For example, if a 1 square kilometer grassland is divided into grids with a spacing of 50 meters, a total range of approximately 40,000 points is generated.

[0055] Entering the search and solution phase, a backtracking method combined with a feedforward mechanism gradually generates paths that meet the requirements. If a candidate path fails to meet any of the above conditions at any point, the exploration branch is immediately pruned, thereby improving computational efficiency. Specifically, assuming that a partial path [p_s, p_a, ...] has been established from point s to target t, the next potential connection node pi is checked in turn to see if it violates any constraint rules.

[0056] The final optimization step selects the optimal path from all valid paths as the grazing path output. For example, in one embodiment, assuming there are three trajectory sequences that meet the requirements {[s,x,y,z], [s,v,w,z], [s,j,k,l]}, and their costs are Cost([p]) = 8, 7, and 6 unit distance values, respectively, the system ultimately selects the third minimum-cost path to complete the control task.

[0057] These parameters are configured with optimal reference values determined based on experimental statistical data. For ecological load limits, the Cmax value is dynamically adjusted based on the average recovery rate from historical data over many years, taking into account grassland species and seasonal variations. The maximum movement cost, d_max, is determined based on the species' actual physiological characteristics. Furthermore, multiple weighted evaluation indicators can be introduced to comprehensively assess the importance of each factor to achieve optimal overall results.

[0058] In a specific embodiment, the grassland is divided into a plurality of grid units, and the ecological carrying capacity constraint condition satisfies the following equation (1):

[0059]

[0060] Wherein, i and j are the grid unit coordinates of the grassland, i is the row coordinate and j is the column coordinate; is the number of times the grid unit (i, j) is effectively utilized during the monitoring period; T recovery (i, j) is the ecological restoration period of grid cell (i, j); K is the number of time segments in the monitoring period; and t is the time slice index (t = 1, 2, ..., K).

[0061] The grid cell division allows each grassland area to independently record grazing activity and statistically analyze the ecological recovery cycle. Rules for the use of relevant parameters within this recovery cycle are then defined, and equation constraints are used to ensure sustainable grassland use and healthy recovery. Within this framework, the dynamic characteristics of each grid cell during the monitoring cycle and their relationship to the entire ecosystem are further defined.

[0062] The first step is to spatially grid the entire grassland. By dividing the large area into smaller areas and expressing them in a matrix coordinate system (i represents the row coordinate and j represents the column coordinate), the carrying capacity of each area can be assessed separately. The key to this step is to facilitate the management and analysis of the characteristics and differences of each local area, so as to accurately implement resource management plans.

[0063] in, The range of values depends on practical factors such as local livestock demand and plant regeneration rates. Ideally, excessive grazing should be avoided, leading to excessive wear and tear on the land and preventing natural repair. For example, in an arid grassland ecosystem, given the low vegetation cover and long recovery time, the value may be limited to 3 to 5 times per year as the optimal solution, balancing ecological stability and economic benefits.

[0064] Next is to introduce T recovery (i, j) represents the number of days of waiting for ecological restoration or the length of the rotation interval required for each part. This parameter reflects the minimum time required for the land to recover from one farming and grazing period and become suitable for the next use. Its specific value distribution range is generally determined by factors such as local environmental climate and soil fertility. For alpine meadow types, assuming that they grow relatively slowly, the corresponding average T recovery (i,j) may be set to 60 to 90 days.

[0065] The parameter K also represents the logical concept of segmenting the total time domain observations into segments, dividing the year into several phases to facilitate adjusting measures based on seasonal fluctuations. The time segment index t serves as an identifier for traversing these subsegments. By breaking down the annual cycle and examining the performance of each smaller interval, more refined control strategy guidance can be obtained.

[0066] The above mathematical expression structure can better adapt to the complex and changing requirements of actual conditions. Ultimately, the goal of optimizing grassland resource allocation while ensuring that the ecological environment is not irreversibly damaged is revealed.

[0067] In one example, consider a herder on a plateau. He has a vast, relatively concentrated water source pasture and needs to plan a rational grazing schedule. First, the pasture is divided into evenly sized rectangular grids (20 x 15 = 300 sub-units) based on slope and proximity to water sources. Next, based on historical data, the optimal stocking frequency for each sub-area is estimated to be approximately 4 to 6 times, with a defined recovery period of approximately two to three months. Next, the year is divided into three phases, based on the three key growth stages, for a rolling management, monitoring, and evaluation process. Specifically, in early spring, the first round of grazing is prioritized in areas near lakes with ample water and rapid temperature rises. Areas reaching the upper limit are recorded in real time and must be removed from current use for maintenance and adjustment before the next round.

[0068] In another optional embodiment, the livestock movement cost constraint satisfies the following equation (2):

[0069]

[0070] Among them, E move is the basic exercise energy consumption, R terrain is the terrain resistance correction parameter, C env is the environmental stress correction parameter, E max,daily is the maximum tolerable daily energy consumption of livestock, and a is the path segment index (a=1, 2,…, A).

[0071] Among them, the basic motion energy consumption equation can be expressed as E move =α·d+β·d·s2+γ·Δh.

[0072] In this equation, α is the energy consumption coefficient for gait on flat ground, which is breed-specific; for example, α differs between sheep and yaks. β is the quadratic coefficient for velocity, reflecting the loss of locomotor efficiency. γ is the elevation penalty (e.g., +1.2 kJ / m for uphill running and -0.3 kJ / m for downhill running).

[0073] The terrain resistance correction can be expressed as R terrain =1+0.05×θ+0.1×SD roughness .

[0074] In this equation, θ is the slope of the terrain; SD roughness is the standard deviation of surface roughness.

[0075] Environmental stress correction C envSet based on factors such as temperature and precipitation.

[0076] In this example, the livestock movement cost constraint achieves dual guarantees for animal welfare and economic benefits by quantifying the interaction between the physiological limits of livestock movement and the terrain environment. Its core value lies in transforming fuzzy "empirical judgments" into a calculable dynamic energy budget system, providing key technical support for smart animal husbandry.

[0077] In yet another optional embodiment, the terrain safety constraint condition includes: marking areas with a slope greater than 25 degrees as grazing-prohibited areas.

[0078] This method involves the following steps: first, analyzing the terrain of the target area; second, setting a slope threshold and using this threshold to identify areas to be marked as prohibited grazing areas; then, spatially labeling these prohibited grazing areas; and finally, using the results for grassland livestock breeding planning. Each step has its own unique meaning and function. The first step involves using terrain analysis tools to analyze terrain data and obtain detailed terrain attribute data (such as elevation and slope). In the second step, a slope parameter is set, and areas deemed unsuitable for grazing are determined to be greater than or equal to a certain value. The key parameter in this formula, slope, typically ranges from 0 to 90 degrees, but in this case, the optimal setting is 25 degrees, as a slope of 25 degrees significantly increases the risk of soil erosion caused by livestock breeding. This step aims to protect ecologically sensitive areas and ensure the long-term sustainability of livestock breeding. The final two steps involve integrating the spatial data generated based on these assessments and demarcating regions for practical application.

[0079] Specifically, in one embodiment, optimizing grassland and livestock husbandry in a hilly area requires incorporating topographic constraints. For example, in the initial analysis phase, a high-precision digital terrain model is used to obtain slope values for each point in the area. Subsequently, a slope threshold greater than 25 degrees is used for slope screening. This identifies certain hilltops and steep areas as grazing-prohibited areas, excluding those that do not meet grazing standards, while retaining other, flatter areas for pasture growth and livestock movement. Furthermore, a slope-dependent function is implemented in this example to assess potential hazard levels: when a calculated slope α within a grid cell exceeds 25 degrees, the corresponding output is marked 1, indicating that the area should be designated as a grazing-prohibited area; otherwise, the output is marked 0, preserving the potential for existing use. The choice of 25 degrees as the criterion here reflects both environmental protection objectives and recommendations from animal behavioral adaptability research—ensuring that even high-intensity rainfall does not significantly cause surface soil slippage, thereby minimizing the likelihood of soil erosion.

[0080] In another optional embodiment, the disease isolation constraint condition includes: when the physiological indicators of the livestock are monitored to exceed preset values, the grid unit where the livestock is located is used as an isolation area.

[0081] This embodiment adopts reasonable zoning and effective control measures for grid units within the pasture, which can include the following steps: setting up a grid system and monitoring livestock in groups; obtaining and evaluating livestock physiological indicator data in real time; determining whether to activate the isolation mechanism based on preset thresholds; and setting the corresponding grid unit as an isolation zone under the confirmed trigger conditions.

[0082] Optionally, during the grid system setup phase, the pasture is divided into equally sized areas for more refined management. For example, the grid size can be expressed as D × D, where D is the grid side length and ranges from 10 to 50 meters, with 30 meters typically being the optimal value. This formula is used to maximize the balance between information coverage density and monitoring accuracy. The significance of this phase is to establish a spatial benchmark for tracking the location and status of each group.

[0083] Alternatively, sensors can be used to collect real-time physiological data such as individual body temperature, pulse, and feeding activity, and statistical or artificial intelligence models can be used to identify data anomalies. For example, the upper limit of the average normal body temperature for a particular breed of cattle could be set at 39.2°C. If the measured value exceeds this, the system will flag it as suspicious. The calculated error in temperature T should be kept below ±0.1°C to ensure reliability.

[0084] Optionally, an alarm is automatically triggered when the monitoring data of an individual livestock deviates significantly from expectations.

[0085] Optionally, after determining that the animals in a grid meet the aforementioned constraints, the physical area in which they reside is designated as an isolation zone. In one embodiment, suppose a dairy cow is located in grid number 30, corresponding to the coordinates (180m, 150m) in the center of the pasture. If its body temperature exceeds the set warning value, grid number 30 is immediately locked as a potential infection source treatment area, and personnel or other livestock are prohibited from entering or leaving the area until a health assessment and subsequent disinfection are completed.

[0086] These steps ensure that the entire quarantine process is quick to respond and has a clear basis, reducing the risk of disease transmission while maintaining the overall stability and production efficiency of the grassland livestock environment.

[0087] In an optional embodiment, the grassland is divided into multiple zones. The size of each zone can be determined based on geographical characteristics and management requirements. Through a reasonable regional design, accurate monitoring and scheduling can be achieved, and clear data support can be provided for subsequent steps.

[0088] See Figure 1 and Figure 2 , Figure 1 Determining the livestock grazing path according to the grazing planning model in step S105 may include:

[0089] Step S201 : monitoring the distribution of grass in different areas through the grazing planning model, and calculating the optimal grazing density of each area based on the distribution.

[0090] Stocking density is typically determined by both pasture area and grass yield, typically ranging from 0.2 to 1 head per hectare. For example, if monitoring of a 50-hectare pasture indicates that a certain area has abundant grass resources and can accommodate a higher stocking density, the optimal stocking density might be 0.8 head per hectare. This allocation can help avoid wasting grass resources or overgrazing.

[0091] Step S202: adjusting the grazing time of each area based on the growth cycle of the grass and the water source distribution density in each area.

[0092] This phase requires a careful consideration of two variables: the rate of grass recovery and the availability of water resources. For example, if the grass in a particular area has a shorter growing period and water resources are relatively scarce, the grazing period allocated to that area may need to be shortened to ensure that excessive damage is not inflicted before the grass has fully recovered.

[0093] Step S203: If the water source distribution density D satisfies D≤D_max, the daily grazing time in the area is limited to less than T_min, where D represents the number of water sources per unit area, D_max is the preset water source density threshold, and T_min is the set minimum daily grazing time.

[0094] Assuming the optimal configuration is D_max equal to 0.05 water sources per square meter and T_min set to 4 hours, this means that when water sources are scarce, the daily effective grazing time should be controlled to reduce the risk of livestock energy consumption due to distance from water sources and the risk of not getting enough water in a timely manner.

[0095] Specifically, in one embodiment, a grassland consists of several areas. If one of these areas is water-scarce and its water density falls below a threshold, the above formula will be used to set rules that will correspondingly reduce the daily grazing period to less than the minimum four hours. This ensures that livestock receive adequate nutrition even in water-scarce conditions, without weakening themselves by traveling long distances to find water. This approach also reflects a balanced approach between optimizing grassland efficiency and maintaining a healthy ecosystem.

[0096] In an optional embodiment, the remote sensing data and the ground data are fused through an edge computing gateway, and the edge computing gateway performs at least one of the following operations: outlier filtering, data spatiotemporal alignment, and data feature extraction.

[0097] Among them, the main goal of outlier filtering is to detect and remove outliers from remote sensing and ground data sets. Outliers are usually caused by measurement noise or external interference, which may destroy the overall consistency of the data or the accuracy of model predictions. To this end, statistical methods such as standard deviation or outlier detection algorithms (Outlier Detection), such as the Z-scor method, can be used to determine whether certain data points are marked as outliers. Assuming that the data follows a normal distribution, the parameter threshold is usually set to greater than or equal to 3 as the judgment criterion, indicating that a point is considered an outlier if the number of standard deviations from the mean exceeds this value. Setting the formula in this way ensures that most non-abnormal data is retained, while abnormal data is accurately filtered out.

[0098] Data temporal and spatial alignment is another core step. Since the recording time of remote sensing data and ground acquisition equipment may be out of sync or geolocation offset, the temporal and spatial information of the data must be processed uniformly. This process requires the use of time series interpolation technology and geographic coordinate mapping functions to achieve synchronization. Taking a specific interpolation formula as an example, the blank intervals on the time axis can be filled by linear interpolation or other high-order interpolation functions. For example, if the missing time data between two sampling points is filled by an interpolation algorithm, the function type must be reasonably determined, and the model that can minimize error fluctuations should be given priority. In space, the deviation of GPS coordinate points can be corrected by projection transformation to achieve the best match between the spatial coordinates of different data sources.

[0099] Data feature extraction aims to extract valuable information from cleaned and aligned data. Feature extraction can effectively reduce dimensionality and highlight the main variables. For example, Fourier analysis is used to extract periodic patterns or principal component analysis (PCA) is used to extract the direction representing the maximum variance. Assuming that the satellite remote sensing image of a grassland livestock breeding scene contains NDVI (vegetation index), feature extraction can be used to obtain the plant growth trend curve of the area and use it to evaluate the health of the grassland. In one embodiment, by modeling historical NDVI data, the specific variation range and position migration of the inter-annual growth peak can be discovered.

[0100] Specifically, in an example of optimizing grassland livestock breeding, edge computing devices access a network of soil moisture sensors on the grassland and infrared imagery captured by drones. Outlier filtering removes data segments where the sensors occasionally report erroneous values. Spatiotemporal matching is used to correct infrared image signals from inconsistent acquisition periods, ensuring that image pixels within each time period correspond to ground truth measurement station locations. Furthermore, for grassland ecological monitoring, a prediction scheme is developed by linking growth dynamic indices derived from feature extraction with herd distribution. This approach not only improves data quality and the efficiency of correlation analysis but also provides support for more precise grazing planning.

[0101] In an optional embodiment, after determining the grazing path of the livestock according to the grazing planning model, the method further includes: controlling the smart electric fence in the pasture to be turned on or off based on the grazing path.

[0102] First, after planning a grazing route, the route is used as input, combining current environmental variables (such as grassland vegetation coverage, weather conditions, and herd location) to dynamically adjust the operating logic of the smart electric fence. The core of this step is to determine the time periods and areas for the electric fence to be activated and deactivated based on real-time data and predefined rules. Specifically, the system determines whether a specific area is suitable for grazing. If so, the electric fence can be opened; otherwise, it remains closed to protect the grassland ecosystem.

[0103] In one specific example, a pasture is divided into several zones, each equipped with a remotely controlled smart electric fence. Specifically, assume that the initial coverage of a zone is 0.5 (meaning only 50% of the grassland is healthy), and it is known that this zone can achieve 70% natural vegetation regeneration after two months of grazing. Therefore, when the vegetation in the zone recovers to the target value (such as 60%), a new cattle grazing plan is allowed to start according to the planned path and the corresponding electric fence is opened. Otherwise, entry will be prohibited until the requirements are met.

[0104] This not only achieves precise management but also improves the efficiency of animal husbandry, reduces pressure on natural resources, and ultimately achieves the goal of sustainable development.

[0105] In an optional embodiment, the method further includes: when it is predicted that the grass gap is greater than a preset value, starting an automatic refeeding function.

[0106] In other words, the above method can be supplemented with an automatic refeeding function. This can be done in the following steps: predicting the grass gap; determining whether the grass gap is greater than a preset value; and activating the automatic refeeding function. The following is a detailed description of each step.

[0107] First, the forage gap is predicted by analyzing the difference between the actual grassland carrying capacity and livestock requirements. The forage gap can be calculated using the formula G = RC, where G is the forage gap, R is the daily forage requirement of the livestock in kilograms, ranging from 50 to 300 kilograms (for example, for adult cattle), with the optimal value depending on the breed and climatic conditions, and C is the current actual grassland supply, also in kilograms, ranging from 10 to 200 kilograms. This formula compares livestock needs with supply capacity to determine the size of the shortfall, which is the prerequisite for supplementary feeding.

[0108] Next, after calculating the G value, we determine whether it is greater than a preset value, T. Here, T represents the acceptable grass deficit threshold, typically set between 5 kg and 30 kg, with an optimal value of 20 kg. This range is chosen based on a balance between economic efficiency and livestock health; exceeding this value may lead to nutritional deficiencies.

[0109] Then, if the first two steps are met, the third step is to start the automatic refeeding function. That is, when the grass shortage exceeds the preset value, the mechanical feeding device is activated or the dry feed refeeding process is deployed.

[0110] For example, suppose a pasture supports 10 cows, each requiring 150 kg of grass per day. R = 1500 kg / day. Current measurements show that the grassland provides approximately 1200 kg per day, resulting in G = 1500 - 1200 = 300 kg. Furthermore, if the system sets T to 20 kg, then because G > T, mechanical feed addition is triggered to make up the difference. Specifically, in one embodiment, additional pelleted compound feed or straw-mixed feed is delivered to the fenced area for grazing.

[0111] Such a complete regulatory mechanism not only improves breeding efficiency but also ensures animal welfare.

[0112] The present invention's optimized control method for grassland livestock farming provides a comprehensive, dynamic pasture management system through multi-source data fusion and intelligent analysis. Specifically, the method addresses key technical issues encountered in grassland livestock farming through the following steps:

[0113] Question 1: How to regulate grazing density according to the grass growth cycle to solve the problem of excessive grassland trampling; Question 2: How to regulate fertilization frequency according to changes in soil nutrients to solve the problem of unstable grass quality; Question 3: How to regulate the amount of supplementary feeding according to weather trends to solve the problem of unbalanced nutrient intake of livestock; Question 4: How to regulate the rotational grazing interval according to the physiological state of the animal to solve the problem of declining livestock product production; Question 5: How to regulate the grazing area according to the distribution characteristics of water sources to solve the problem of local grassland degradation.

[0114] First, using satellite remote sensing technology and ground-based sensors to collect grassland growth and environmental data, we integrated this multi-source data to construct a three-dimensional grassland growth model. This model accurately reflects the distribution characteristics, growth cycle, and soil nutrient status of forage grasses, providing a basis for subsequent grazing density control (addressing Question 1). Based on the dynamic forage growth data provided by this model, we adjust grazing density to reduce the risk of overtrampling.

[0115] Secondly, based on the three-dimensional growth model, the soil nutrient change monitoring function is further introduced. By analyzing the nutrient requirements of forage grass and soil conditions, the fertilization frequency is precisely regulated (for problem 2), thereby ensuring the stability of forage growth quality and adapting to the self-recovery ability of the ecosystem.

[0116] Furthermore, combining weather forecast data with information on livestock physiological status, information collection sensors can be used to predict weather trends and their impact on grazing conditions. Based on weather factors, supplementary feeding amounts and ratios can be scientifically planned to meet feed supply requirements under different conditions and ensure balanced nutrition for livestock (addressing question 3).

[0117] In order to address the problem of declining livestock product production, the method also analyzes the appropriate rotational grazing interval based on animal health status data (for question 4) to ensure that animals continue to produce in the best condition and thus maximize economic benefits.

[0118] Finally, redesigning grazing area demarcation strategies based on the spatial distribution of water sources across grasslands (addressing question 5) can effectively prevent localized degradation in certain areas due to water shortages or inappropriate water management. This approach can better protect and restore grassland ecosystems and achieve sustainable development goals.

[0119] See Figure 3 The embodiment of the present application also provides an optimization and control device 30 for grassland animal husbandry, including: a grassland data acquisition module 301, which is used to collect remote sensing data of the grassland through satellites and collect ground data of the grassland through ground sensors; a grassland model construction module 302, which is used to fuse the remote sensing data and the ground data to construct a three-dimensional growth model of the grassland; a livestock data acquisition module 303, which is used to obtain the physiological indicators of the livestock through livestock information acquisition sensors; a grazing planning model construction module 304, which is used to construct a grazing planning model based on the three-dimensional growth model and the physiological indicators; a model analysis module 305, which is used to determine the grazing path of the livestock according to the grazing planning model. The grazing planning model uses a constraint satisfaction algorithm to determine the grazing path of the livestock;

[0120] In an optional embodiment, the grazing path satisfies at least one of the following constraints: ecological carrying capacity constraint; terrain safety constraint; livestock movement cost constraint; and disease isolation constraint.

[0121] In an optional embodiment, the grassland is divided into a plurality of grid units, and the ecological carrying capacity constraint satisfies the following equation:

[0122]

[0123] Wherein, i and j are the grid unit coordinates of the grassland, i is the row coordinate and j is the column coordinate; is the number of times the grid unit (i, j) is effectively utilized during the monitoring period; T recovery (i, j) is the ecological restoration period of grid cell (i, j); K is the number of time segments in the monitoring period; and t is the time slice index (t = 1, 2, ..., K).

[0124] In an optional embodiment, the livestock movement cost constraint satisfies the following equation:

[0125]

[0126] Among them, E move is the basic exercise energy consumption, R terrain is the terrain resistance correction parameter, C env is the environmental stress correction parameter, E max,daily is the maximum tolerable daily energy consumption of livestock, and a is the path segment index (a=1, 2,…, A).

[0127] In an optional embodiment, the terrain safety constraint condition includes: marking areas with a slope greater than 25 degrees as grazing-prohibited areas.

[0128] In an optional embodiment, the disease isolation constraint condition includes: when the physiological indicators of the livestock are monitored to exceed preset values, the grid unit where the livestock is located is used as an isolation area.

[0129] In an optional embodiment, the grassland is divided into multiple areas, and the model analysis module 305 may include:

[0130] a grazing density confirmation unit, configured to monitor the distribution of grass in different areas through the grazing planning model, and calculate the optimal grazing density of each area based on the distribution;

[0131] A grazing duration adjustment module, configured to adjust the grazing duration of each area based on the growth cycle of the forage grass and the water source distribution density in each area;

[0132] Among them, if the water source distribution density D satisfies D≤D_max, the daily grazing time in the area is restricted to less than T_min, where D represents the number of water sources per unit area, D_max is the preset water source density threshold, and T_min is the set minimum daily grazing time.

[0133] In an optional embodiment, the remote sensing data and the ground data are fused through an edge computing gateway, and the edge computing gateway performs at least one of the following operations: outlier filtering, data spatiotemporal alignment, and data feature extraction.

[0134] In an optional embodiment, the optimization and control device 30 for grassland animal husbandry may further include: a fence control module for controlling the opening or closing of the smart electric fence in the grassland based on the grazing path.

[0135] In an optional embodiment, the grassland livestock breeding optimization and control device 30 may further include: an automatic refeeding module, which is used to start the automatic refeeding function when it is predicted that the grass gap is greater than a preset value.

[0136] For more information about the working principle and working method of the optimization and control device 30 for grassland animal husbandry, please refer to the Figure 1 and Figure 2 The relevant description of the optimization and regulation methods for grassland animal husbandry will not be repeated here.

[0137] The present application also provides a storage medium, specifically a computer-readable storage medium, for storing a computer program that, when executed by a computer or processor, performs any of the steps of the above-described method for optimizing and controlling grassland livestock farming. The computer-readable storage medium may include non-volatile or non-transitory memory, and may also include an optical disk, a mechanical hard disk, a solid-state drive, and the like.

[0138] An embodiment of the present application also provides a computer device, which may include a computer device including a memory and a processor, wherein the memory stores a computer program, and when the program is executed by the processor, it can implement the steps of any one of the above-mentioned methods for optimizing and regulating grassland animal husbandry.

[0139] In the embodiments of the present application, the processor may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component, and may implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of the present application. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of the present application may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0140] In an embodiment of the present application, the memory may be a non-volatile memory, such as a hard disk drive (HDD) or a solid-state drive (SSD), or a volatile memory (volatile memory), such as a random-access memory (RAM). The memory is any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory in the embodiment of the present application may also be a circuit or any other device that can implement a storage function, for storing computer programs and / or data.

[0141] The optimization and control method for grassland animal husbandry provided in the embodiment of the present application 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 is loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user device or other programmable device. The computer program can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. 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 data center that includes one or more available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital video disc (DVD)), or a semiconductor medium (eg, an SSD).

[0142] Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include these modifications and variations.

[0143] It should be noted that for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

[0144] In the above embodiments, the description of each embodiment has its own emphasis. Any multiple embodiments can be used in combination. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0145] During implementation, each step of the above method can be completed by an integrated logic circuit of hardware in a processor or by instructions in the form of software. The steps of the method disclosed in conjunction with the embodiments of the present application can be directly embodied as being executed by a hardware processor, or can be executed by a combination of hardware and software units in the processor. The software unit can be located in a storage medium mature in the art, such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor executes the instructions in the memory, and in combination with its hardware, completes the steps of the above method. To avoid repetition, a detailed description is not given here.

[0146] In the embodiments of the present application, the processor of the above-mentioned device may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0147] The present application also provides a computer program product comprising a non-transitory computer-readable storage medium storing a computer program, wherein the computer program is operable to cause a computer to execute some or all of the steps of any of the methods described in the above method embodiments. The computer program product may be a software installation package.

[0148] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above-mentioned units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0149] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0150] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0151] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, including a number of instructions for enabling a computer device (which can be a personal computer, server or TRP, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program code.

[0152] It should be understood that the term "and / or" in this document simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document indicates that the related objects are in an "or" relationship.

[0153] The term "plurality" used in the embodiments of the present application refers to two or more.

[0154] The first, second, etc. descriptions appearing in the embodiments of this application are only for illustration and distinction of the description objects. There is no order, nor does it indicate any special limitation on the number of devices in the embodiments of this application, and cannot constitute any limitation on the embodiments of this application.

[0155] The "connection" appearing in the embodiments of the present application refers to various connection methods such as direct connection or indirect connection to achieve communication between devices, and the embodiments of the present application do not impose any limitations on this.

[0156] The above is a detailed introduction to the embodiments of the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of the present application. At the same time, for those skilled in the art, according to the idea of the present application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for optimizing and controlling grassland animal husbandry, characterized in that: The method comprises: Collecting remote sensing data of the grassland through satellites and collecting ground data of the grassland through ground sensors; fusing the remote sensing data and the ground data to construct a three-dimensional growth model of the grassland; Acquiring physiological indicators of the livestock through livestock information collection sensors; constructing a grazing planning model based on the three-dimensional growth model and the physiological indicators; The grazing path of the livestock is determined according to the grazing planning model.

2. The method according to claim 1, characterized in that The grazing planning model uses a constraint satisfaction algorithm to determine the grazing path of the livestock; The grazing path satisfies at least one of the following constraints: ecological carrying capacity constraint; terrain safety constraint; livestock movement cost constraint; and disease isolation constraint.

3. The method according to claim 2, characterized in that The grassland is divided into a plurality of grid units, and the ecological carrying capacity constraint condition satisfies the following equation: Wherein, i and j are the grid unit coordinates of the grassland, i is the row coordinate and j is the column coordinate; is the number of times the grid unit (i, j) is effectively utilized during the monitoring period; T recovery (i, j) is the ecological restoration period of grid cell (i, j); K is the number of time segments in the monitoring period; and t is the time slice index (t = 1, 2, ..., K).

4. The method according to claim 2, characterized in that The livestock movement cost constraint satisfies the following equation: Among them, E move is the basic exercise energy consumption, R terrain is the terrain resistance correction parameter, C env is the environmental stress correction parameter, E max,daily is the maximum tolerable daily energy consumption of livestock, and a is the path segment index (a=1, 2,…, A).

5. The method according to claim 2, characterized in that The terrain safety constraint condition includes: marking areas with a slope greater than 25 degrees as grazing-prohibited areas.

6. The method according to claim 2, characterized in that The disease isolation constraint condition includes: when the physiological indicators of the livestock are monitored to exceed preset values, the grid unit where the livestock is located is used as an isolation area.

7. The method according to any one of claims 1 to 6, characterized in that The pasture is divided into a plurality of areas, and determining the grazing path of the livestock according to the grazing planning model includes: monitoring the distribution of grass in different areas through the grazing planning model, and calculating the optimal grazing density of each area based on the distribution; Adjusting the grazing time in each area based on the growth cycle of the forage grass and the water source distribution density in each area; If the water source distribution density D satisfies D≤D_max, the daily grazing time in the area is limited to less than T_min, where D represents the number of water sources per unit area, D_max is the preset water source density threshold, and T_min is the set minimum daily grazing time.

8. The method according to any one of claims 1 to 6, characterized in that The remote sensing data and the ground data are fused through an edge computing gateway, and the edge computing gateway performs at least one of the following operations: outlier filtering, data spatiotemporal alignment, and data feature extraction.

9. The method according to any one of claims 1 to 6, characterized in that After determining the livestock grazing path according to the grazing planning model, the method further includes: The smart electric fence in the pasture is controlled to be turned on or off based on the grazing path.

10. The method according to any one of claims 1 to 6, characterized in that The method further includes: when it is predicted that the grass gap is greater than a preset value, starting an automatic supplementary feeding function.

Images