Grazing path planning method, device, electronic device and storage medium

Abstract

The present invention provides a grazing path planning method, apparatus, electronic device and storage medium, relating to the field of remote sensing technology. The method includes: dividing a target grazing area into blocks of different forage grades based on the remote sensing image of the target grazing area; obtaining the livestock groups to be grazed corresponding to each forage grade based on the characteristic information of each livestock to be grazed; the number of blocks of any forage grade in the target grazing area is multiple; constructing an objective function corresponding to each grazing path planning objective; based on the multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning objective, obtaining the cruising order of the livestock groups to be grazed corresponding to each forage grade for the blocks of each forage grade, and further determining the grazing paths of the livestock groups to be grazed corresponding to each forage grade. The grazing path planning method, apparatus, electronic device and storage medium provided by the present invention can obtain an optimal grazing path plan more efficiently in a complex scenario.

Description

Technical Field

[0001] The present invention relates to the field of remote sensing technology, and in particular to a grazing path planning method, device, electronic equipment and storage medium. Background Art

[0002] Grassland is an important part of the terrestrial ecosystem and plays a vital role in ecological security and animal husbandry development.

[0003] However, overgrazing and aimless grazing have seriously aggravated the degradation of grasslands and weakened their ecological functions. Therefore, it is crucial to guide grazing activities in a scientific way to monitor and utilize the entire grassland ecosystem.

[0004] However, grassland ecosystems have very rich biodiversity and complex ecological relationships, and traditional grazing path planning methods are usually difficult to plan the optimal grazing path in complex scenarios, and the efficiency of traditional grazing path planning is low. Therefore, how to plan more efficiently and obtain a better grazing path is a technical problem that needs to be solved in this field. Summary of the invention

[0005] The present invention provides a grazing path planning method, device, electronic device and storage medium, which are used to solve the defects that traditional grazing path planning methods in the prior art are usually difficult to plan the optimal grazing path in complex scenarios, and the traditional grazing path planning methods are inefficient in grazing path planning, so as to achieve more efficient planning and obtain a better grazing path.

[0006] The present invention provides a grazing path planning method, comprising the following steps.

[0007] The remote sensing image of the target grazing area, the characteristic information of each livestock to be grazed and the planning target of each grazing route are obtained, wherein the characteristic information includes the livestock growth period information and health information.

[0008] Based on the remote sensing image, the target grazing area is divided into blocks of different forage grades. Based on the characteristic information of each of the livestock to be grazed, the livestock to be grazed are grouped to obtain livestock groups to be grazed corresponding to each forage grade. There are multiple blocks of any one of the forage grades in the target grazing area.

[0009] Construct an objective function corresponding to each grazing path planning target, and for each forage grade, obtain the cruising order of the livestock group to be grazed corresponding to each forage grade for the blocks of each forage grade based on a multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning target.

[0010] Determine the grazing path of the livestock group to be grazed corresponding to each pasture level based on the cruising order of the blocks of each pasture level.

[0011] According to a grazing path planning method provided by the present invention, the dividing the target grazing area into blocks of different pasture levels based on the remote sensing image includes: obtaining the target vegetation index value of the target grazing area based on the remote sensing image, where the target vegetation index includes normalized difference vegetation index, enhanced vegetation index, normalized difference red edge index, normalized difference water index, and blue light normalized difference vegetation index; obtaining the grass yield and nutrient index value of the target grazing area based on the target vegetation index value of the target grazing area; and dividing the target grazing area into blocks of different pasture levels based on the grass yield and nutrient index value of the target grazing area.

[0012] According to a grazing path planning method provided by the present invention, the obtaining the cruising order of the livestock group to be grazed corresponding to each pasture level for the blocks of each pasture level based on the multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning objective includes: determining the geometric center point of each block of each pasture level within the target grazing area as a node; determining the predefined grazing starting point as the starting node and the predefined grazing ending point as the ending node, and solving for the cruising order of the livestock group to be grazed corresponding to each pasture level for the blocks of each pasture level based on the multi-objective dynamic programming algorithm, the position information of the starting node, the ending node, and each node, and the objective function corresponding to each grazing path planning objective, to obtain candidate solutions for the cruising order of each block corresponding to each pasture level; determining the candidate solution for the cruising order of the block with the smallest corresponding objective function value among the candidate solutions for the cruising order of each block corresponding to each pasture level as the optimal solution for the cruising order of the block corresponding to each pasture level; and determining the cruising order of the livestock group to be grazed corresponding to each pasture level for the blocks of each pasture level based on the optimal solution for the cruising order of the block corresponding to each pasture level.

[0013] According to a grazing path planning method provided by the present invention, solving the cruising order of the livestock group to be grazed at each forage level for each block at each forage level based on the multi-objective dynamic programming algorithm, the starting node, the ending node, the position information of each node, and the objective function corresponding to each grazing path planning objective, and obtaining candidate solutions for the cruising order of each block corresponding to each forage level, including: after solving the cruising order of the livestock group to be grazed at each forage level for each block at each forage level based on the multi-objective dynamic programming algorithm, the starting node, the ending node, the position information of each node, and the objective function corresponding to each grazing path planning objective, and obtaining candidate solutions for the cruising order of each block corresponding to each forage level in this iteration, judging whether the objective function value corresponding to the candidate solution for the cruising order of each block corresponding to each forage level in this iteration is less than the objective function value corresponding to the candidate solution for the target cruising order of each block corresponding to each forage level, where the candidate solution for the target cruising order of each block corresponding to each forage level is the candidate solution for the cruising order of each block corresponding to each forage level obtained before this iteration and having the smallest objective function value; in the case where the objective function value corresponding to the candidate solution for the cruising order of each block corresponding to each forage level in this iteration is less than the objective function value corresponding to the candidate solution for the target cruising order of each block corresponding to each forage level, updating the candidate solution for the target cruising order of each block corresponding to each forage level to the candidate solution for the cruising order of each block corresponding to each forage level in this iteration, and in the case where the objective function value corresponding to the candidate solution for the cruising order of each block corresponding to each forage level in this iteration is not less than the objective function value corresponding to the candidate solution for the target cruising order of each block corresponding to each forage level, deleting the candidate solution for the cruising order of each block corresponding to each forage level in this iteration.

[0014] According to a grazing path planning method provided by the present invention, obtaining the grass yield and nutrient index value of the target grazing area based on the target vegetation index value of the target grazing area, including: inputting the target vegetation index value of the target grazing area into a grass yield estimation model to obtain the grass yield and nutrient index value of the target grazing area output by the grass yield estimation model; wherein, the grass yield estimation model is constructed based on a random forest machine learning model and trained based on the target vegetation index value of the sample grazing area and the grass yield and nutrient index value of the sample grazing area.

[0015] A grazing path planning method provided by the present invention, which determines the grazing paths of the livestock groups to be grazed corresponding to each pasture level based on the cruising order of the blocks of each pasture level, includes: sequentially connecting the geometric centers of the blocks of each pasture level from the grazing starting point according to the cruising order of the blocks of each pasture level corresponding to the livestock groups to be grazed, and then taking the connected path as the grazing path of the livestock groups to be grazed corresponding to each pasture level.

[0016] A grazing path planning method provided by the present invention, each of the grazing path planning objectives includes: the shortest grazing path length, the minimum total turning angle on the grazing path, avoiding the restricted grazing area, passing through the watering point, the single grazing duration of the livestock to be grazed not exceeding the duration threshold, and avoiding multiple blocks in the blocks of each pasture level in the target grazing area that are in the overloaded carrying capacity state.

[0017] The present invention also provides a grazing path planning device, including the following modules:

[0018] The data acquisition module is used to acquire the remote sensing image of the target grazing area, the characteristic information of each livestock to be grazed, and each grazing path planning objective, and the characteristic information includes livestock growth period information and health information.

[0019] The block division module is used to divide the target grazing area into blocks of different pasture levels based on the remote sensing image, group each livestock to be grazed based on the characteristic information of each livestock to be grazed, and obtain the livestock groups to be grazed corresponding to each pasture level. The number of blocks of any pasture level in the target grazing area is multiple.

[0020] The path planning module is used to construct the objective function corresponding to each grazing path planning objective. For each pasture level, based on the multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning objective, obtain the cruising order of the livestock groups to be grazed corresponding to each pasture level for the blocks of each pasture level.

[0021] The path generation module is used to obtain the grazing paths of the livestock groups to be grazed corresponding to each pasture level based on the cruising order of the blocks of each pasture level.

[0022] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements any one of the above-mentioned grazing path planning methods.

[0023] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the grazing path planning method as described in any one of the above is implemented.

[0024] The present invention also provides a computer program product, including a computer program, and when the computer program is executed by a processor, the grazing path planning method as described in any one of the above is implemented.

[0025] The grazing path planning method, device, electronic device and storage medium provided by the present invention divide the target grazing area into blocks of different forage levels based on the remote sensing image of the target grazing area, group each livestock to be grazed based on the characteristic information of each livestock to be grazed, and after obtaining the livestock groups to be grazed corresponding to each forage level, construct an objective function corresponding to each grazing path planning objective. For each forage level, based on the multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning objective, obtain the cruising order of the livestock groups to be grazed corresponding to each forage level for the blocks of each forage level, and then determine the grazing paths of the livestock groups to be grazed corresponding to each forage level based on the cruising order of the blocks of each forage level. When performing grazing path planning, factors such as the carrying capacity of the grassland and the differences in the livestock herd structure are considered, the grazing path planning process under multiple grazing path planning objectives is optimized, the optimal grazing path planning can be obtained more efficiently in complex scenarios, the problems of overgrazing and aimless grazing can be better solved, the automatic selection and autonomous planning of the optimal grazing route are realized, and a multi-path intelligent grazing control strategy is formulated for the differences in the carrying capacity of the grassland and the livestock herd structure, which can provide technical support for environmental protection and the sustainable utilization of grassland resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0027] Figure 1 It is a schematic flowchart of the grazing path planning method provided by the present invention.

[0028] Figure 2 It is a schematic diagram of the area of each forage level in the target grazing area in the grazing path planning method provided by the present invention.

[0029] Figure 3 It is a schematic structural diagram of the grazing path planning device provided by the present invention.

[0030] Figure 4It is a schematic structural diagram of the electronic device provided by the present invention. Specific Embodiments

[0031] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Apparently, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without any creative efforts shall fall within the protection scope of the present invention.

[0032] In the description of the present invention, it should be noted that unless otherwise clearly defined and limited, the terms "installed", "connected" and "coupled" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the internal communication of two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0033] In the description of the present application, the terms "first", "second", etc. are used to distinguish similar objects and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of the same category, and the number of objects is not limited. For example, the first object may be one or more. In addition, in the description of the present application, " / or" means at least one of the connected objects, and the character " / " generally means an "or" relationship between the associated objects before and after.

[0034] It should be noted that natural grasslands support the economic, social stability and ecological security of pastoral animal husbandry. At present, the level of informatization management of family ranches in pastoral areas of China is backward, lacking reasonable grazing plans and optimal utilization of resources, and there is a shortage of software and hardware technology products for grazing management. Unregulated overgrazing and unplanned grazing have exacerbated grassland degradation and decline in ecological functions. Grazing management based on scientific planning algorithms to guide the grazing activities of family ranches is crucial for protecting grassland ecological functions, maintaining the balance between grass and livestock, and sustainable utilization.

[0035] In related technologies, traditional grazing path planning methods can plan the grazing paths of herds based on algorithms such as dynamic programming.

[0036] Among them, the dynamic programming (DP for short) algorithm is an algorithmic technique that breaks down complex problems into smaller sub-problems and constructs the final solution by solving these sub-problems. Its core idea lies in saving the solutions of sub-problems to avoid redundant calculations, thereby significantly improving efficiency. Dynamic programming is particularly suitable for problems with overlapping sub-problems and optimal sub-structure properties. In dynamic programming, it is usually necessary to define a state transition equation, which describes how to construct the solution to a larger problem from the solutions of the already solved sub-problems. To this end, dynamic programming generally includes the following steps: defining the state, determining the state transition equation, initializing the boundary conditions, and iteratively calculating the final solution through the state transition equation.

[0037] For the grazing route planning task where a herd needs to visit multiple fixed grazing points and finally return to the pen, when planning the above grazing route planning task based on the dynamic programming algorithm, the above grazing route planning task can be abstracted into the classical Traveling Salesman Problem (TSP), that is, finding the shortest path that passes through all grazing points and returns to the starting point.

[0038] When solving the Traveling Salesman Problem based on the dynamic programming algorithm, state representation and sub-problem division are usually used. Specifically, a two-dimensional array can be used to represent the state. Through the state transition equation, the solution to a larger problem can be gradually constructed from smaller sub-problems.

[0039] When performing the above grazing route planning task based on the dynamic programming algorithm, first, all the grazing points to be visited and the starting point need to be marked as nodes, and then the distances between these nodes are calculated to construct the distance matrix dis. Next, the dynamic programming array dp is initialized, and the shortest path for each state is recursively calculated through the state transition equation. Finally, by traversing all possible end points, the shortest path from the starting point passing through all grazing points and returning to the starting point is found. By saving the intermediate results, dynamic programming effectively avoids redundant calculations and significantly improves the calculation efficiency. Especially in grazing path planning, by optimizing the path length, not only can the workload of herdsmen be reduced, but also the grazing time and location can be arranged more scientifically and reasonably, improving the grazing efficiency.

[0040] However, due to the very rich biodiversity and complex ecological relationships in the grassland ecosystem, in practical applications, the grazing route planning task usually includes multiple goals. For example, the herd not only needs to visit multiple fixed grazing points and finally return to the pen, but also the grazing path obtained needs to meet the requirements of the shortest path and the minimum turning angle at the same time. When performing the above multi-goal grazing path planning task based on the dynamic programming algorithm, there are significant defects.

[0041] On the one hand, dynamic programming algorithms usually focus on the optimization of a single objective and it is difficult to consider multiple objectives simultaneously. For example, although dynamic programming algorithms can effectively find solutions with the shortest path length, they cannot meet the optimization of the path turning angle. This is because the state definition and transition equations are usually only for a single objective and cannot comprehensively reflect the complex relationships between multiple objectives, resulting in the optimization result not being globally optimal.

[0042] On the other hand, due to its characteristic of traversal search, the computational efficiency of dynamic programming algorithms is relatively low. The algorithm must traverse all possible state and path combinations, which not only makes the computational amount increase exponentially but also leads to a large number of redundant and repetitive calculation problems. Especially in the multi-objective path planning task, the computational efficiency of the dynamic programming algorithm is extremely low, and it is difficult to calculate the optimal grazing path within a reasonable time, resulting in a low efficiency of the grazing path planning.

[0043] Therefore, the dynamic programming algorithm performs well in some single-objective grazing path tasks, but its limitations are very obvious in the multi-objective grazing path planning task in complex scenarios.

[0044] The Multi-Objective Dynamic Programming (MODP) algorithm is the application of the dynamic programming method in multi-objective optimization problems. As an optimization algorithm, the multi-objective dynamic programming algorithm can determine the optimal grazing planning path in the multi-objective grazing path planning task in complex scenarios. By introducing multiple objective functions and performing comprehensive optimization during the dynamic programming process, the multi-objective dynamic programming algorithm can better adapt to the complex actual application requirements.

[0045] In the grazing path planning method provided by the present invention, the traditional dynamic programming algorithm is improved so that it can consider multiple objectives simultaneously and can dynamically adjust and optimize the grazing path under multiple objectives to meet the requirements of different grassland carrying capacities and herd structures. The grazing path planning method provided by the present invention takes into account factors such as the grassland carrying capacity and the differences in herd structures, optimizes the multi-objective grazing path planning process, and realizes efficient and accurate grazing path planning. Compared with the traditional grazing path planning method, the grazing path planning method provided by the present invention can better solve the problems of overgrazing and untargeted grazing, realize the automatic selection and autonomous planning of the optimal route for intelligent grazing, formulate a multi-path intelligent grazing control strategy for the differences in grassland carrying capacity and herd structures, and can provide technical support for the protection of grassland ecological environment and the sustainable utilization of pasture resources.

[0046] The following combines Figure 1 - Figure 2 to describe the grazing path planning method of the present invention.

[0047] Figure 1It is a schematic flowchart of the grazing path planning method provided by the present invention. As Figure 1 shown, the method includes the following: Step 101, obtain the remote sensing image of the target grazing area, the characteristic information of each livestock to be grazed, and each grazing path planning target, where the characteristic information includes the livestock growth period information and health information.

[0048] It should be noted that the execution subject of the embodiments of the present invention is a grazing path planning device.

[0049] Specifically, the livestock to be grazed is the grazing object that executes the planned grazing path in the grazing path planning method provided by the present invention, and the target grazing area is the area where the planned grazing path is located in the grazing path planning method provided by the present invention.

[0050] The types of livestock to be grazed in the embodiments of the present invention can be cattle, sheep, horses, camels, etc.

[0051] It can be understood that the livestock to be grazed and the target grazing area in the embodiments of the present invention can be determined based on actual needs. There are no specific limitations on the livestock to be grazed and the target grazing area in the embodiments of the present invention.

[0052] Optionally, the target grazing area in the embodiments of the present invention can be the area covered by a family ranch. Among them, a family ranch is a ranch operated and managed in units of families.

[0053] In the embodiments of the present invention, the remote sensing image of the target grazing area can be obtained in various ways. For example: in the embodiments of the present invention, satellite remote sensing technology can be used to obtain the remote sensing image of the target grazing area; or, in the embodiments of the present invention, unmanned aerial vehicle technology can also be used to obtain the remote sensing image of the target grazing area.

[0054] In the embodiments of the present invention, the characteristic information of each livestock to be grazed can be obtained based on the user's input or by querying the existing data in the database.

[0055] In the embodiments of the present invention, multiple grazing path planning targets can be obtained based on the user's input or received from other electronic devices.

[0056] It should be noted that each grazing path planning target in the embodiments of the present invention can be determined by the manager of the livestock to be grazed according to the actual situation. There are no limitations on each grazing path planning target in the embodiments of the present invention.

[0057] As an optionally implemented example, each grazing path planning objective includes: the shortest grazing path length, the minimum total turning angle on the grazing path, avoiding prohibited grazing areas, passing through watering points, the single grazing duration of the livestock to be grazed not exceeding the duration threshold, and avoiding multiple ones of the blocks in the overstocked state among the blocks of each forage grade in the target grazing area.

[0058] It should be noted that the carrying capacity refers to the number of livestock that can be accommodated per unit area or in each grazing area during the grazing process. The carrying capacity includes three factors: the number of livestock heads, the grassland area, and the grazing time, and its size is affected by various factors such as livestock species, grazing systems, climate, and soil. In the embodiments of the present invention, the state of the block can be based on the carrying capacity of the block. After dividing the target grazing area into blocks of different forage grades, for any block, the grassland carrying capacity of the above block can be calculated by the following formula:

[0059] Grassland carrying capacity = (forage yield per unit area × grassland utilization rate) / (livestock daily food intake × grazing days)

[0060] After calculating the grassland carrying capacity of the above block, the state of the above block being in an overstocked state, a balanced state, or an understocked state can be determined based on the grassland carrying capacity of the above block.

[0061] Further, based on the state of the above block, a label indicating overstocked carrying capacity, a label indicating balanced carrying capacity, or a label indicating understocked carrying capacity can be added to the above block.

[0062] Step 102: Based on the remote sensing image, divide the target grazing area into blocks of different forage grades, and group each livestock to be grazed based on the characteristic information of each livestock to be grazed, so as to obtain the livestock group to be grazed corresponding to each forage grade. The number of blocks of any forage grade in the target grazing area is multiple.

[0063] Specifically, after obtaining the remote sensing image of the target grazing area, based on the remote sensing image of the target grazing area, the target grazing area can be divided into blocks with excellent forage grade, blocks with good forage grade, and blocks with poor forage grade by means of numerical calculation, deep learning technology, etc.

[0064] It should be noted that in the embodiments of the present invention, the forage yield in the blocks with excellent forage grade is the highest, the nutrient index value is the highest, and the forage quality in the blocks with excellent forage grade is the best; the forage yield in the blocks with good forage grade is the second highest, the nutrient index value is the second highest, and the forage quality in the blocks with good forage grade is the second best; the forage yield in the blocks with poor forage grade is the lowest, the nutrient index value is the lowest, and the forage quality in the blocks with poor forage grade is the worst.

[0065] As an optional embodiment, based on remote sensing images, the target grazing area is divided into blocks with different forage grades, including: based on remote sensing images, obtaining the target vegetation index value of the target grazing area, and the target vegetation index includes the normalized difference vegetation index, enhanced vegetation index, normalized difference red edge index, normalized difference water index, and blue-band normalized difference vegetation index.

[0066] Specifically, based on the spectral information carried in the remote sensing images of the target grazing area, the normalized difference vegetation index value, enhanced vegetation index value, normalized difference red edge index value, normalized difference water index value, and blue-band normalized difference vegetation index value of the target grazing area can be calculated through numerical calculation. The specific calculation formulas are as follows:

[0067]

[0068]

[0069]

[0070]

[0071]

[0072] Wherein, represents the reflectance of the blue band; represents the reflectance of the red band; represents the reflectance of the near-infrared band; represents the reflectance of the short-wave infrared band; represents the reflectance of the red edge band.

[0073] It should be noted that the normalized difference vegetation index (Normalized difference vegetation index, NDVI), enhanced vegetation index (Enhanced vegetation index, EVI), normalized difference red edge index (Normalized difference red edge index, NDRE), normalized difference water index (Normalized differential water index, NDWI), and blue-band normalized difference vegetation index (Blue-normalized difference vegetation index, BNDVI) cover the spectral information of multiple bands, such as blue light, red light, red edge, near-infrared, and short-wave infrared, etc., and can comprehensively reflect the growth status, water content, health status, and biomass and other characteristics of vegetation.

[0074] The reasons for selecting the above-mentioned target vegetation indices in the embodiments of the present invention are as follows: NDVI is a classic and widely used vegetation index. It utilizes data in the red and near-infrared light bands, can effectively distinguish vegetation and non-vegetation areas, and reflect the growth status and density of vegetation. The calculation method of NDVI is simple and highly sensitive to vegetation, thus becoming the preferred index for monitoring vegetation cover and growth changes. EVI is an improvement based on NDVI, introducing corrections for atmospheric and soil backgrounds, especially correcting the influence of atmospheric aerosols through the blue light band. EVI is more sensitive to high-density vegetation areas and can more accurately reflect the health status and growth trends of high-biomass areas, which makes EVI perform better in vegetation monitoring in complex environments. NDRE is calculated using the red-edge band (between red and near-infrared) and can sensitively detect the chlorophyll content and health status of vegetation. The red-edge band is very effective for monitoring the early growth and health changes of vegetation. Especially in refined vegetation health assessment, NDRE is an important tool. NDWI is specifically used to reflect the vegetation water content by combining the near-infrared and short-wave infrared bands. Water is an important factor in vegetation growth and health. NDWI can effectively monitor the plant water status and drought degree, providing key water information support for grassland management and grazing planning. BNDVI utilizes the blue and near-infrared bands and can provide stable vegetation monitoring results under high-reflection backgrounds (such as water bodies or soil). The blue light band is sensitive to short-term vegetation changes and the initial growth stage and can capture the early signals of vegetation growth, which is an important supplement for dynamic monitoring.

[0075] Based on the target vegetation index value of the target grazing area, obtain the grass yield and nutrient index values of the target grazing area.

[0076] Specifically, after obtaining the target vegetation index value of the target grazing area, the grass yield and nutrient index values of the target grazing area can be obtained through methods such as numerical calculation, mathematical statistics, conditional judgment, and deep learning techniques.

[0077] It should be noted that the nutrient index of the grazing area is an important parameter for evaluating the soil and vegetation nutrient status of the grazing area. The nutrient index of the grazing area can include, but is not limited to, soil pH value, soil organic matter content, vegetation nutrient content, trace element content, total nitrogen content, available phosphorus content, and available potassium content, etc.

[0078] As an optional embodiment, based on the target vegetation index value of the target grazing area, obtaining the grass yield and nutrient index values of the target grazing area includes: inputting the target vegetation index value of the target grazing area into the grass yield estimation model, and obtaining the grass yield and nutrient index values of the target grazing area output by the grass yield estimation model.

[0079] Among them, the grass yield estimation model is constructed based on the random forest machine learning model and trained based on the target vegetation index values of the sample grazing areas and the grass yield and nutrient index values of the sample grazing areas.

[0080] It should be noted that the random forest (RF) machine learning model is an ensemble learning model mainly used for classification and regression tasks. The random forest machine learning model is a powerful machine learning model that can provide excellent performance in various classification and regression tasks. By integrating multiple decision trees and introducing randomness, the random forest improves the model accuracy while reducing the risk of overfitting.

[0081] In the embodiment of the present invention, after constructing the initial model based on the random forest machine learning model, the initial model can be trained with the target vegetation index values of the sample grazing areas as training samples and the grass yield and nutrient index values of the sample grazing areas as sample labels to obtain the grass yield estimation model.

[0082] After obtaining the grass product estimation model, the target vegetation index value of the target grazing area can be input into the above grass yield estimation model.

[0083] The above grass yield estimation model can estimate the grass yield and nutrient index values of the target grazing area based on the target vegetation index value of the target grazing area, and then the grass yield and nutrient index values of the target grazing area output by the above grass yield estimation model can be obtained.

[0084] It should be noted that in the embodiment of the present invention, Python 3.9 can be used as the development environment, and the model can be called using the third-party data science library scikit-learn.

[0085] Based on the grass yield and nutrient index values of the target grazing area, the target grazing area is divided into blocks of different forage grades.

[0086] Specifically, after obtaining the grass yield and nutrient index values of the target grazing area, the spatial distribution of the grass yield and the spatial distribution of the nutrient index values of the target grazing area can be analyzed based on the grass yield and nutrient index values of the target grazing area. Then, based on the spatial distribution of the grass yield and the spatial distribution of the nutrient index values of the target grazing area, the spatial distribution of the grass yield and the spatial distribution of the nutrient index values can be subjected to normal distribution standardization processing through the ArcGIS PRO data analysis tool, and the regional grouping method can be used to determine the blocks with excellent forage grade and the blocks with poor forage grade within the target grazing area.

[0087] After determining the areas with excellent forage quality and the areas with poor forage quality in the target grazing area, the remaining areas in the target grazing area can be determined as the areas with good forage quality. Based on the average areas of the areas with excellent forage quality and the areas with poor forage quality in the target grazing area, the areas with good forage quality in the target grazing area can be divided, and the areas with good forage quality in the target grazing area can be divided into multiple blocks with the above average area to obtain the blocks with good forage quality in the target grazing area.

[0088] Figure 2 It is a schematic diagram of the areas of each forage quality level in the target grazing area in the grazing path planning method provided by the present invention. The distribution of the blocks of each forage quality level in the target grazing area is as Figure 2 shown.

[0089] It should be noted that in the embodiments of the present invention, the blocks of different levels are divided proportionally within the confidence interval, Figure 2 and the blocks of each forage quality level are shown as a whole. In the actual application of the invention, the blocks of different forage quality levels often show a more complex and overlapping distribution, thus reflecting the importance of path planning.

[0090] In the embodiments of the present invention, based on the characteristic information of each livestock to be grazed, the livestock to be grazed can be grouped by means of conditional judgment.

[0091] According to the livestock growth period information of the livestock to be grazed, a growth period label can be added to the livestock to be grazed, and the above growth period label includes: young, lactating, pregnant, growth period and non-growth period.

[0092] According to the health information of the livestock to be grazed, a health condition label can be added to the livestock to be grazed, and the above health condition label includes: healthy, sick.

[0093] After adding the growth period label and the health condition label to each livestock to be grazed, based on the growth period label and the health condition label added to each livestock to be grazed, the forage quality level corresponding to each livestock to be grazed can be determined.

[0094] Specifically, when the growth period label of any livestock to be grazed is non-growth period and the health condition label is healthy, it can be determined that the forage quality level corresponding to the above livestock to be grazed is poor.

[0095] When the growth period label of any livestock to be grazed is growth period and the health condition label is healthy, or when the growth period label of any livestock to be grazed is non-growth period and the health condition label is sick, it can be determined that the forage quality level corresponding to the above livestock to be grazed is good.

[0096] When the breeding period label of any livestock to be grazed is the breeding period and the health condition label is sick, or when the breeding period label of any livestock to be grazed is young, lactating or pregnant, the forage grade corresponding to the above-mentioned livestock to be grazed can be determined as excellent.

[0097] Step 103: Construct an objective function corresponding to each grazing path planning objective. For each forage grade, based on the multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning objective, obtain the cruising order of the livestock group to be grazed for each forage grade in each block.

[0098] Specifically, in the embodiments of the present invention, an objective function corresponding to each grazing path planning objective can be constructed according to each grazing path planning objective, and then, based on the multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning objective, the cruising order of the livestock group to be grazed for each forage grade in each block can be obtained by means of numerical calculation.

[0099] As an optional embodiment, obtaining the cruising order of the livestock group to be grazed for each forage grade in each block based on the multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning objective includes: determining the geometric center point of each block of each forage grade in the target grazing area as each node corresponding to each forage grade.

[0100] Determine the predefined grazing starting point as the starting node, and determine the predefined grazing ending point as the ending node. Based on the multi-objective dynamic programming algorithm, the position information of the starting node, the ending node, each node corresponding to each forage grade, and the objective function corresponding to each grazing path planning objective, solve the cruising order of the livestock group to be grazed for each forage grade in each block to obtain candidate solutions for the cruising order of each block corresponding to each forage grade.

[0101] Determine the candidate solution of the cruising order of the block with the smallest objective function value among the candidate solutions of the cruising order of each block corresponding to each forage grade as the optimal solution of the cruising order of the block corresponding to each forage grade.

[0102] Based on the optimal solution of the cruising order of the block corresponding to each forage grade, determine the cruising order of the livestock group to be grazed for each forage grade in each block.

[0103] Specifically, after dividing the target grazing area into blocks with different forage levels, the boundary position information of each block can be obtained in various ways. For example, the boundary position information of each block can be obtained based on the mapping relationship between the remote sensing image of the target grazing area and the target grazing area; or, when the boundary position information of each block has been stored in a CSV file, the boundary position information of each block can be obtained by reading the above CSV file.

[0104] It should be noted that the boundary position information of the block in the embodiment of the present invention can be represented by longitude and latitude.

[0105] It should be noted that in order to facilitate the understanding of the method for obtaining the cruising order of the livestock group to be grazed corresponding to each forage level for each block of each forage level based on the multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning objective provided by the present invention, the following takes each grazing path planning objective including: the shortest grazing path length, the minimum total turning angle on the grazing path, and avoiding the prohibited grazing area as an example to specifically describe the embodiment of the present invention.

[0106] For the forage level k , define the set S including each node corresponding to the forage level k in the target grazing area. According to the boundary position information of each block with the forage level of k in the target grazing area, calculate the distance between any two blocks with the forage level of k , and construct a distance matrix D .

[0107] Define the state indicating the shortest path length and the total turning angle from the starting node, passing through the nodes in the set S , and reaching the node . The state is represented by a binary tuple:

[0108]

[0109] The initial state is represented as: .

[0110] Based on the boundary position information of the prohibited grazing area, generate a taboo table including all grazing points and the prohibited grazing area, assign the value of 0 to the unvisited grazing points, and assign the value of 1 to the visited ones. Before the traversal starts, initialize the value of the prohibited grazing area to 1, thereby achieving the effect of avoiding this area when the path visits the nodes. The state is represented as

[0111] For each state , by selecting nodes from the set in Transfer to the node , the path length and the total turning angle can be updated:

[0112]

[0113] wherein, the turning angle is obtained by calculating the included angle of vectors:

[0114]

[0115] The final state is the optimal solution starting from the starting node, passing through some or all of the nodes in the set and reaching the node :

[0116]

[0117] As an alternative embodiment, based on the multi-objective dynamic programming algorithm, the starting node, the ending node, the position information of each node corresponding to each forage level, and the objective function corresponding to each grazing path planning objective, solve the cruising order of the livestock group to be grazed corresponding to each forage level for each block of each forage level, and obtain the candidate solutions for the cruising order of each block corresponding to each forage level, including: after solving the cruising order of the livestock group to be grazed corresponding to each forage level for each block of each forage level based on the multi-objective dynamic programming algorithm, the starting node, the ending node, the position information of each node, and the objective function corresponding to each grazing path planning objective, and obtaining the candidate solutions for the cruising order of each block corresponding to each forage level in this iteration, determine whether the objective function value corresponding to the candidate solution for the cruising order of each block corresponding to each forage level in this iteration is less than the objective function value corresponding to the candidate solution for the cruising order target of each block corresponding to each forage level. The candidate solution for the cruising order target of each block corresponding to each forage level is the candidate solution for the cruising order corresponding to each forage level obtained before this iteration with the smallest objective function value.

[0118] In the case where the objective function value corresponding to the candidate solution for the cruising order of each block corresponding to each forage level in this iteration is less than the objective function value corresponding to the candidate solution for the cruising order target of each block corresponding to each forage level, update the candidate solution for the cruising order target of each block corresponding to each forage level to the candidate solution for the cruising order of each block corresponding to each forage level in this iteration. In the case where the objective function value corresponding to the candidate solution for the cruising order of each block corresponding to each forage level in this iteration is not less than the objective function value corresponding to the candidate solution for the cruising order target of each block corresponding to each forage level, delete the candidate solution for the cruising order of each block corresponding to each forage level in this iteration.

[0119] It should be noted that in the embodiments of the present invention, in order to save computing resources and improve computing efficiency, during the process of solving the cruising order of the livestock groups to be grazed for each pasture grade for each pasture grade block based on the multi-objective dynamic programming algorithm, the starting node, the ending node, the position information of each node corresponding to each pasture grade, and the objective function corresponding to each grazing path planning objective, pruning rules can be used to prematurely terminate unnecessary calculations and reduce the search space. The specific implementation can be represented by the following formula:

[0120]

[0121] Step 104: Based on the cruising order of the blocks of each pasture grade, determine the grazing paths of the livestock groups to be grazed corresponding to each pasture grade.

[0122] As an optional embodiment, determining the grazing paths of the livestock groups to be grazed corresponding to each pasture grade based on the cruising order of the blocks of each pasture grade includes: sequentially connecting the geometric centers of the blocks of each pasture grade according to the cruising order of the livestock groups to be grazed corresponding to each pasture grade from the grazing starting point, and then taking the connected path as the grazing path of the livestock groups to be grazed corresponding to each pasture grade.

[0123] In the embodiments of the present invention, based on the remote sensing image of the target grazing area, the target grazing area is divided into blocks of different pasture grades. Based on the characteristic information of each livestock to be grazed, the livestock to be grazed are grouped to obtain the livestock groups to be grazed corresponding to each pasture grade. Then, the objective function corresponding to each grazing path planning objective is constructed. For each pasture grade, based on the multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning objective, the cruising order of the livestock groups to be grazed corresponding to each pasture grade for each pasture grade block is obtained. Furthermore, based on the cruising order of the blocks of each pasture grade, the grazing paths of the livestock groups to be grazed corresponding to each pasture grade are determined. When planning the grazing path, factors such as the carrying capacity of the grassland and the differences in the herd structure are considered, optimizing the grazing path planning process under multiple grazing path planning objectives, enabling more efficient acquisition of the optimal grazing path planning in complex scenarios, better solving the problems of overgrazing and untargeted grazing, realizing the automatic selection and autonomous planning of the optimal grazing route, formulating a multi-path intelligent grazing control strategy for the carrying capacity of the grassland and the differences in the herd structure, and providing technical support for environmental protection and the sustainable utilization of grassland resources.

[0124] The grazing path planning method provided by the present invention defines the objective function corresponding to each grazing path planning objective, constructs the state transition equation, and through multi-objective grouping, distributes the livestock group to forage areas of different grades according to different health conditions, gradually solves the optimal grazing path, and realizes the comprehensive optimization of multi-objective allocation, the shortest path length, and the smallest turning angle.

[0125] In the grazing path planning method provided by the present invention, a pruning operation is introduced. By setting pruning rules, unnecessary calculations are terminated in advance, the search space is reduced, and the calculation efficiency of the algorithm is improved.

[0126] The grazing path planning method provided by the present invention automatically selects and independently plans the optimal grazing route according to the grassland carrying capacity and herd structure, can avoid overgrazing and aimless grazing behaviors, and promotes the sustainable development of the grassland ecosystem.

[0127] Figure 3 It is a schematic structural diagram of the grazing path planning device provided by the present invention. The following combines Figure 3 Describe the grazing path planning device provided by the present invention. The grazing path planning device described below can be correspondingly referred to the grazing path planning method described above. As Figure 3 shown, the device includes: a data acquisition module 301, a block division module 302, a path planning module 303, and a path generation module 304.

[0128] The data acquisition module 301 is used to acquire the remote sensing image of the target grazing area, the characteristic information of each livestock to be grazed, and each grazing path planning objective. The characteristic information includes livestock growth period information and health information.

[0129] The block division module 302 is used to divide the target grazing area into blocks of different forage grades based on the remote sensing image, group each livestock to be grazed based on the characteristic information of each livestock to be grazed, and obtain the livestock group to be grazed corresponding to each forage grade. The number of blocks of any forage grade in the target grazing area is multiple.

[0130] The path planning module 303 is used to construct the objective function corresponding to each grazing path planning objective. For each forage grade, based on the multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning objective, obtain the cruising order of the livestock group to be grazed corresponding to each forage grade for the blocks of each forage grade.

[0131] The path generation module 304 is used to obtain the grazing path of the livestock group to be grazed corresponding to each forage grade based on the cruising order of the blocks of each forage grade.

[0132] Specifically, the data acquisition module 301, the block division module 302, the path planning module 303, and the path generation module 304 are electrically connected.

[0133] In the grazing path planning device according to the embodiment of the present invention, based on the remote sensing image of the target grazing area, the target grazing area is divided into blocks with different forage levels. After grouping the to-be-grazed livestock based on the characteristic information of each to-be-grazed livestock and obtaining the to-be-grazed livestock groups corresponding to each forage level, an objective function corresponding to each grazing path planning target is constructed. For each forage level, based on the multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning target, the cruising order of the to-be-grazed livestock groups corresponding to each forage level for the blocks of each forage level is obtained. Then, based on the cruising order of the blocks of each forage level, the grazing paths of the to-be-grazed livestock groups corresponding to each forage level are determined. When planning the grazing path, factors such as the carrying capacity of the grassland and the differences in the herd structure are considered, optimizing the grazing path planning process under multiple grazing path planning targets, being able to more efficiently obtain the optimal grazing path planning in complex scenarios, better solving the problems of overgrazing and untargeted grazing, realizing the automatic selection and autonomous planning of the optimal grazing route, formulating a multi-path intelligent grazing control strategy for the carrying capacity of the grassland and the differences in the herd structure, and being able to provide technical support for environmental protection and the sustainable utilization of grassland resources.

[0134] Figure 4 An example of the physical structure diagram of an electronic device is shown in Figure 4As shown, the electronic device may include: a processor 410, a communications interface 420, a memory 430, and a communication bus 440. Among them, the processor 410, the communications interface 420, and the memory 430 complete communication with each other through the communication bus 440. The processor 410 may call the logical instructions in the memory 430 to execute the grazing path planning method, which includes: obtaining remote sensing images of the target grazing area, the characteristic information of each livestock to be grazed, and each grazing path planning objective, where the characteristic information includes livestock growth period information and health information; based on the remote sensing images, dividing the target grazing area into blocks of different forage levels, and grouping each livestock to be grazed based on the characteristic information of each livestock to be grazed to obtain the livestock groups to be grazed corresponding to each forage level. The number of blocks of any forage level in the target grazing area is multiple; constructing an objective function corresponding to each grazing path planning objective, and for each forage level, based on the multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning objective, obtaining the cruising order of the livestock groups to be grazed corresponding to each forage level for the blocks of each forage level; based on the cruising order of the blocks of each forage level, obtaining the grazing paths of the livestock groups to be grazed corresponding to each forage level.

[0135] In addition, when the logical instructions in the above-mentioned memory 430 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs, etc., which can store program codes.

[0136] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the grazing path planning method provided by the above-mentioned various methods. The method includes: obtaining a remote sensing image of a target grazing area, characteristic information of each livestock to be grazed, and each grazing path planning target, where the characteristic information includes livestock growth period information and health information; based on the remote sensing image, dividing the target grazing area into blocks of different forage levels, and based on the characteristic information of each livestock to be grazed, grouping the livestock to be grazed to obtain a group of livestock to be grazed corresponding to each forage level, and the number of blocks of any forage level in the target grazing area is multiple; constructing an objective function corresponding to each grazing path planning target, and for each forage level, based on the multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning target, obtaining the cruising order of the group of livestock to be grazed corresponding to each forage level for the blocks of each forage level; based on the cruising order of the blocks of each forage level, obtaining the grazing path of the group of livestock to be grazed corresponding to each forage level.

[0137] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is implemented to execute the grazing path planning method provided by the above-mentioned various methods. The method includes: obtaining a remote sensing image of a target grazing area, characteristic information of each livestock to be grazed, and each grazing path planning target, where the characteristic information includes livestock growth period information and health information; based on the remote sensing image, dividing the target grazing area into blocks of different forage levels, and based on the characteristic information of each livestock to be grazed, grouping the livestock to be grazed to obtain a group of livestock to be grazed corresponding to each forage level, and the number of blocks of any forage level in the target grazing area is multiple; constructing an objective function corresponding to each grazing path planning target, and for each forage level, based on the multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning target, obtaining the cruising order of the group of livestock to be grazed corresponding to each forage level for the blocks of each forage level; based on the cruising order of the blocks of each forage level, obtaining the grazing path of the group of livestock to be grazed corresponding to each forage level.

[0138] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0139] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solutions, in essence, or the parts that contribute to the prior art can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0140] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present invention.

Claims

1. A grazing path planning method, characterized in that, Including: Obtaining a remote sensing image of a target grazing area, characteristic information of each livestock to be grazed, and each grazing path planning objective, where the characteristic information includes livestock growth period information and health information; Based on the remote sensing image, dividing the target grazing area into blocks of different forage levels, and grouping each livestock to be grazed based on the characteristic information of each livestock to be grazed, to obtain a group of livestock to be grazed corresponding to each forage level. The number of blocks of any forage level in the target grazing area is multiple; Constructing an objective function corresponding to each grazing path planning objective. For each forage level, based on the multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning objective, obtaining the cruising order of the group of livestock to be grazed corresponding to each forage level for the blocks of each forage level; Based on the cruising order of the blocks of each forage level, determining the grazing path of the group of livestock to be grazed corresponding to each forage level; The obtaining the cruising order of the group of livestock to be grazed corresponding to each forage level for the blocks of each forage level based on the multi-objective dynamic programming algorithm and the objective function corresponding to each grazing path planning objective includes: Determining the geometric center point of each block of each forage level in the target grazing area as a node; Determining a predefined grazing starting point as a starting node and a predefined grazing ending point as an ending node. Based on the multi-objective dynamic programming algorithm, the position information of the starting node, the ending node, and each node, and the objective function corresponding to each grazing path planning objective, solving for the cruising order of the group of livestock to be grazed corresponding to each forage level for the blocks of each forage level, to obtain candidate solutions for the cruising order of each block corresponding to each forage level; Determining the candidate solution for the cruising order of the block with the smallest corresponding objective function value among the candidate solutions for the cruising order of each block corresponding to each forage level as the optimal solution for the cruising order of the block corresponding to each forage level; Based on the optimal solution for the cruising order of the block corresponding to each forage level, determining the cruising order of the group of livestock to be grazed corresponding to each forage level for the blocks of each forage level.

2. The grazing path planning method according to claim 1, wherein The dividing the target grazing area into blocks of different forage levels based on the remote sensing image includes: Based on the remote sensing image, obtaining the target vegetation index value of the target grazing area, where the target vegetation index includes normalized difference vegetation index, enhanced vegetation index, normalized difference red edge index, normalized difference water index, and blue light normalized difference vegetation index; Based on the target vegetation index value of the target grazing area, obtaining the grass yield and nutrient index value of the target grazing area; Based on the grass yield and nutrient index value of the target grazing area, dividing the target grazing area into blocks of different forage levels.

3. The grazing path planning method according to claim 1, wherein Solving the cruising order of the livestock groups to be grazed for each forage grade for each block of each forage grade based on the multi-objective dynamic programming algorithm, the starting node, the ending node, the position information of each node, and the objective function corresponding to each grazing path planning objective, to obtain candidate solutions for the cruising order of each block corresponding to each forage grade, including: After solving the cruising order of the livestock groups to be grazed for each forage grade for each block of each forage grade based on the multi-objective dynamic programming algorithm, the starting node, the ending node, the position information of each node, and the objective function corresponding to each grazing path planning objective, and obtaining candidate solutions for the cruising order of each block corresponding to each forage grade in this iteration, determine whether the objective function value corresponding to the candidate solution for the cruising order of each block corresponding to each forage grade in this iteration is less than the objective function value corresponding to the candidate solution for the target cruising order of each block corresponding to each forage grade. The candidate solution for the target cruising order of each block corresponding to each forage grade is the candidate solution for the cruising order of each block corresponding to each forage grade obtained before this iteration with the smallest objective function value; In the case where the objective function value corresponding to the candidate solution for the cruising order of each block corresponding to each forage grade in this iteration is less than the objective function value corresponding to the candidate solution for the target cruising order of each block corresponding to each forage grade, update the candidate solution for the target cruising order of each block corresponding to each forage grade to the candidate solution for the cruising order of each block corresponding to each forage grade in this iteration. In the case where the objective function value corresponding to the candidate solution for the cruising order of each block corresponding to each forage grade in this iteration is not less than the objective function value corresponding to the candidate solution for the target cruising order of each block corresponding to each forage grade, delete the candidate solution for the cruising order of each block corresponding to each forage grade in this iteration.

4. The grazing path planning method according to claim 2, wherein, Obtaining the grass yield and nutrient index values of the target grazing area based on the target vegetation index value of the target grazing area, including: Inputting the target vegetation index value of the target grazing area into the grass yield estimation model to obtain the grass yield and nutrient index values of the target grazing area output by the grass yield estimation model; Among them, the grass yield estimation model is constructed based on a random forest machine learning model and trained based on the target vegetation index value of the sample grazing area and the grass yield and nutrient index values of the sample grazing area.

5. The grazing path planning method according to claim 1, characterized in that, Determining the grazing path of the livestock groups to be grazed for each forage grade based on the cruising order of each block of each forage grade, including: According to the cruising order of the livestock groups to be grazed for each forage grade for each block of each forage grade, successively connect the geometric centers of each block of each forage grade starting from the grazing starting point, and then use the connected path as the grazing path of the livestock groups to be grazed for each forage grade.

6. The grazing path planning method according to any one of claims 1 to 5, characterized in that, Each of the grazing path planning objectives includes: the shortest grazing path length, the minimum total turning angle on the grazing path, avoiding restricted grazing areas, passing by water points, the single grazing duration of the livestock to be grazed not exceeding a duration threshold, and avoiding multiple blocks in the blocks of each forage grade in the target grazing area that are in an overstocked state.

7. A grazing path planning device, characterized in that, Including: A data acquisition module, configured to acquire a remote sensing image of a target grazing area, characteristic information of each livestock to be grazed, and each grazing path planning objective, where the characteristic information includes livestock growth period information and health information; A block division module, configured to divide the target grazing area into blocks of different forage grades based on the remote sensing image, and group each of the livestock to be grazed based on the characteristic information of each of the livestock to be grazed, to obtain a group of livestock to be grazed corresponding to each forage grade, and the number of blocks of any forage grade in the target grazing area is multiple; A path planning module, configured to construct an objective function corresponding to each of the grazing path planning objectives, and for each forage grade, based on a multi-objective dynamic programming algorithm and the objective function corresponding to each of the grazing path planning objectives, obtain the cruising order of the group of livestock to be grazed corresponding to each forage grade for the blocks of each forage grade; A path generation module, configured to obtain the grazing path of the group of livestock to be grazed corresponding to each forage grade based on the cruising order of the blocks of each forage grade; The path planning module obtains the cruising order of the group of livestock to be grazed corresponding to each forage grade for the blocks of each forage grade based on a multi-objective dynamic programming algorithm and the objective function corresponding to each of the grazing path planning objectives, including: Determining the geometric center point of each block of each forage grade in the target grazing area as a node; Determining a predefined grazing starting point as a starting node, determining a predefined grazing ending point as an ending node, and based on a multi-objective dynamic programming algorithm and the position information of the starting node, the ending node, and each of the nodes, as well as the objective function corresponding to each of the grazing path planning objectives, solving the cruising order of the group of livestock to be grazed corresponding to each forage grade for the blocks of each forage grade, to obtain candidate solutions for the cruising order of each block corresponding to each forage grade; Determining the candidate solution for the cruising order of the block with the minimum objective function value among the candidate solutions for the cruising order of each block corresponding to each forage grade as the optimal solution for the cruising order of the block corresponding to each forage grade; Based on the optimal solution for the cruising order of the block corresponding to each forage grade, determining the cruising order of the group of livestock to be grazed corresponding to each forage grade for the blocks of each forage grade.

8. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, When the processor executes the program, it implements the grazing path planning method according to any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the grazing path planning method according to any one of claims 1 to 6.

Images