Method for optimizing collaborative layout of sensor and mobile equipment in smart agriculture

Abstract

The embodiment of the invention provides an optimization method for collaborative layout of sensors and mobile equipment in smart agriculture, and the method comprises the steps: determining key monitoring points for sensor deployment according to farmland region division and crop distribution conditions; based on a path planning algorithm of the mobile equipment, calculating an optimal data acquisition route, and ensuring that the mobile equipment can efficiently access all sensor nodes; historical data and environmental changes are analyzed by using a machine learning algorithm, and working parameters of the sensor and the inspection frequency of the mobile device are dynamically adjusted, so that the accuracy and timeliness of data acquisition are improved. Through the scheme of the embodiment of the invention, the effective collaborative layout of the sensor and the mobile equipment can be realized in the smart agriculture so as to improve the data acquisition efficiency and precision.

Description

Technical Field

[0001] The present application relates to the field of smart agriculture technology, and in particular to a method for optimizing the coordinated layout of sensors and mobile devices in smart agriculture. Background Art

[0002] The optimization method for the coordinated layout of sensors and mobile devices in smart agriculture aims to improve the efficiency and accuracy of data collection by rationally planning the location and path of sensor networks and mobile devices (such as drones, robots, etc.). Specifically, this method usually involves using algorithms to determine the optimal sensor deployment location and patrol routes of mobile devices so that key areas of the farmland can be covered while reducing unnecessary duplicate coverage or omissions.

[0003] In view of the problems of this method, how to achieve effective collaborative layout of sensors and mobile devices in smart agriculture to improve data collection efficiency and accuracy is a core challenge. This is mainly because the farmland environment is complex and changeable, and the influence of factors such as topography, crop species distribution, and seasonal changes need to be considered. In addition, problems such as limited sensor energy and insufficient battery life of mobile devices need to be solved to ensure the stable operation and long-term maintenance of the entire system. Therefore, designing a collaborative layout solution that can adapt to complex environments and collect data efficiently is one of the current research focuses. Summary of the invention

[0004] In view of this, an embodiment of the present disclosure provides an optimization method for the coordinated layout of sensors and mobile devices in smart agriculture, which at least partially solves the problems existing in the prior art.

[0005] An optimization method for the coordinated layout of sensors and mobile devices in smart agriculture, comprising: Determine the key monitoring points for sensor deployment based on farmland area division and crop distribution; Based on the path planning algorithm of the mobile device, the optimal data collection route is calculated to ensure that the mobile device can efficiently access all sensor nodes; Machine learning algorithms are used to analyze historical data and environmental changes, and dynamically adjust the operating parameters of sensors and the inspection frequency of mobile devices to improve the accuracy and timeliness of data collection.

[0006] In a specific embodiment, Identify different crop planting areas based on farmland topographic maps and crop type distribution maps; Prioritize each planting area based on crop growth cycle and irrigation needs; Using the Kmeans clustering algorithm, the number and approximate locations of monitoring points are determined based on priority and area; The specific locations of the monitoring points are adjusted by minimizing the sum of the Euclidean distances between the monitoring points.

[0007] In a specific embodiment, Initialize the cluster centers and select random locations equal to the number of monitoring points as the initial cluster centers; Assign points within each planting area to the nearest cluster center; Update the location of the cluster center to the average location of all points in the cluster; The iteration stops when the change of cluster center is less than the preset threshold.

[0008] In a specific embodiment, Calculate the distance difference between each cluster center after the current iteration and the cluster center of the previous iteration; If the distance difference of all cluster centers is less than the threshold, the iteration is stopped; The threshold is set as a percentage of the average distance between monitoring points, for example 5%.

[0009] In a specific embodiment, Calculate the sum of distances between all monitoring points; Calculate the average distance between monitoring points; Set the threshold to the average distance multiplied by the percentage, for example 5; If the distance difference of all cluster centers is less than the threshold, it is considered converged.

[0010] In a specific embodiment, For each cluster center, calculate the distance difference between it and the previous iteration position; If the distance difference of all cluster centers is less than the threshold, the cluster center is considered stable; The threshold calculation formula is: Threshold = Average distance * Percentage.

[0011] In a specific embodiment, Determine the starting location of the mobile device and the location of the target sensor node; Establish a grid model of the farmland area and map each sensor node to the corresponding grid unit; Use Dijkstra algorithm or A* search algorithm to calculate the shortest path from the starting location to each sensor node; Based on the calculated shortest path set, the optimal data collection route for the mobile device is generated.

[0012] In a specific embodiment, Initialize the distance matrix, setting the distance from the starting position to all other grid cells to infinity, except for the starting position itself to 0; Add the starting position to the unvisited node set and set it as the current node; For each neighbor node of the current node, if the distance from the current node to the neighbor node is less than the known shortest distance, update the distance of the neighbor node; If the distance of a neighbor node satisfies "newly calculated distance < known shortest distance", the shortest distance of the neighbor node is updated.

[0013] In a specific embodiment, Calculate the actual distance from the current node to the neighboring node; Calculate the total distance from the starting position to the neighboring node, that is, "the actual distance from the current node to the neighboring node + the distance from the starting position to the current node"; If the calculated total distance is less than the shortest distance recorded by the neighbor node, the shortest distance of the neighbor node is updated to the total distance; If the calculated total distance satisfies "total distance < shortest distance recorded by neighboring nodes", the shortest distance of the neighboring nodes is updated.

[0014] In a specific embodiment, Update the shortest distance of neighbor nodes to the calculated total distance; Mark neighbor nodes as updated; If the neighbor node is not in the unvisited node set, add it to the unvisited node set; If the neighbor node is not in the unvisited node set, add it to the unvisited node set.

[0015] The disclosed embodiment provides an optimization method for the coordinated layout of sensors and mobile devices in smart agriculture, including: determining the key monitoring points for sensor deployment according to the farmland area division and crop distribution; calculating the optimal data collection route based on the path planning algorithm of the mobile device to ensure that the mobile device can efficiently access all sensor nodes; using machine learning algorithms to analyze historical data and environmental changes, dynamically adjusting the working parameters of the sensor and the inspection frequency of the mobile device to improve the accuracy and timeliness of data collection. Through the solution of the disclosed embodiment, it is possible to solve how to achieve effective coordinated layout of sensors and mobile devices in smart agriculture to improve data collection efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the exemplary implementation methods of the embodiments of the present disclosure, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the embodiments of the present disclosure and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 It is a flow chart of an optimization method for the coordinated layout of sensors and mobile devices in smart agriculture; Figure 2 It is a flowchart to determine the key monitoring points for sensor deployment based on the farmland area division and crop distribution; Figure 3 It is a flow chart for determining the number and approximate location of monitoring points using the Kmeans clustering algorithm; Figure 4 It is a flowchart that stops iteration when the change of cluster center is less than the preset threshold; Figure 5 is a flow chart with the threshold set as a percentage of the average distance between monitoring points; Figure 6 It is a flowchart that if the distance difference of all cluster centers is less than the threshold, it is considered converged; Figure 7 It is a flow chart based on the path planning algorithm of the mobile device to calculate the optimal data collection route and ensure that the mobile device can efficiently access all sensor nodes; Figure 8 It is a flowchart that uses Dijkstra algorithm or A* search algorithm to calculate the shortest path from the starting location to each sensor node; Fig. 9 If the distance of a neighbor node satisfies "the newly calculated distance < the known shortest distance", the flowchart of updating the shortest distance of the neighbor node is shown; Fig.10 It is a flowchart for updating the shortest distance of neighbor nodes if the calculated total distance satisfies "total distance < shortest distance recorded by neighbor nodes". DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the embodiments of the present disclosure more clear, the embodiments of the present disclosure are further described in detail in combination with the embodiments and drawings. The schematic implementation methods of the embodiments of the present disclosure and their descriptions are only used to explain the embodiments of the present disclosure and are not intended to limit the embodiments of the present disclosure.

[0019] Next, refer to Figure 1 Describe this optimization method for the coordinated layout of sensors and mobile devices in smart agriculture: S101 determines the key monitoring points for sensor deployment based on the division of farmland areas and crop distribution. This process first requires a detailed geographic information analysis of the farmland, including factors such as soil type, topography, irrigation system, and information such as the specific planting location and growth cycle of crops. By comprehensively considering this information, it can be determined which areas are critical to crop growth and require more intensive or more frequent data collection. For example, in a large farmland, if a certain area is planted with crops that are particularly sensitive to moisture, then more humidity sensors and temperature sensors need to be deployed in that area to monitor the changes of these key factors in real time. At the same time, considering the different growth cycles of different crops, more intensive data collection may be required in certain specific periods, so the number of sensors needs to be increased at the monitoring points during these periods. In addition, environmental factors around the farmland, such as wind direction, sunshine time, etc., need to be considered, which may affect the growth of crops, so corresponding monitoring points also need to be set up in these aspects. In this way, it can be ensured that the sensor network covers all important monitoring areas, thus providing a solid foundation for subsequent data collection.

[0020] S102 is to calculate the optimal data collection route based on the path planning algorithm of the mobile device to ensure that the mobile device can efficiently access all sensor nodes. This step involves path planning for mobile devices such as drones and robots, etc., in order to reduce unnecessary moving distances, reduce energy consumption, and ensure that data from all sensor nodes can be collected in a timely manner. Path planning algorithms usually consider factors such as the maximum moving speed of the mobile device, battery life, and the distance between sensor nodes. For example, in a network composed of multiple sensor nodes, the best path from one node to another can be determined by the shortest path algorithm in graph theory such as the Dijkstra algorithm or the A* search algorithm. In addition, heuristic algorithms such as genetic algorithms or ant colony algorithms can be combined to find the global optimal solution to ensure that the entire path planning is both efficient and energy-saving. In practical applications, assuming that there are 50 sensor nodes in a farmland, an optimal path can be calculated through a path planning algorithm, so that the mobile device can collect data from all nodes by moving along this path without having to repeat certain areas or go back, which not only saves time but also reduces energy consumption.

[0021] Through the above steps, the problem of how to achieve effective collaborative layout of sensors and mobile devices in smart agriculture to improve data collection efficiency and accuracy can be solved. Specifically, by accurately determining key monitoring points and rationally planning the path of mobile devices, it can be ensured that data collection is both comprehensive and efficient. At the same time, by dynamically adjusting the working parameters of sensors and the inspection frequency of mobile devices, it is possible to respond flexibly according to actual conditions, further improving the accuracy and timeliness of data collection. For example, during the critical period of crop growth, the sampling frequency of sensors and the number of inspections of mobile devices can be appropriately increased to capture more information about the growth status of crops, thereby helping farmers make more accurate decisions and improve the yield and quality of crops.

[0022] Next, refer to Figure 2 Describe the specific steps of determining the key monitoring points of sensor deployment in an optimization method for the coordinated layout of sensors and mobile devices in smart agriculture: S201 identifies different crop planting areas based on farmland topographic maps and crop type distribution maps. This process involves analyzing farmland topographic data and crop distribution information to determine which areas are planted with specific types of crops. For example, geographic information system (GIS) software can be used to integrate these map data and automatically identify the planting ranges of different crops through image processing technology.

[0023] S202 is to prioritize each planting area according to the crop growth cycle and irrigation needs. This means that information about the growth cycle of each crop and their water requirements need to be collected to determine which areas need to be monitored more closely. For example, crop areas with greater water requirements may be given higher priorities so that they receive more attention in the monitoring network.

[0024] S203 uses the Kmeans clustering algorithm to determine the number and approximate location of monitoring points based on priority and area. This step involves dividing the farmland into several cluster centers, each center representing a potential monitoring point. The Kmeans algorithm determines the number of monitoring points in each area based on the priority and area size of the crop area, and gives the approximate location of these monitoring points. For example, in a larger area with a higher priority, more monitoring points may be set.

[0025] S204 is to adjust the specific positions of the monitoring points by minimizing the sum of the Euclidean distances between the monitoring points. Once the approximate positions of the monitoring points are determined, it is necessary to further optimize these positions to ensure that the distances between the monitoring points are as evenly distributed as possible to achieve the best coverage effect. This can be achieved by iteratively calculating the distances between the monitoring points and adjusting the positions until the preset optimization goal is met. For example, if the distance between two monitoring points is too close, the position of one of the monitoring points can be appropriately adjusted to reduce redundancy and improve the efficiency of the entire monitoring network.

[0026] Next, refer to Figure 3 Describe the specific steps of using the Kmeans clustering algorithm to determine the number and approximate locations of monitoring points in the optimization method for the coordinated layout of sensors and mobile devices in smart agriculture according to claim 2: S301 is to initialize the cluster centers: when implementing the method of the present invention, it is necessary to first determine the approximate number of monitoring points and select random positions equal to the number as the initial cluster centers. These random positions should cover the entire planting area to ensure that the subsequent clustering process can fully reflect the different environmental conditions in the planting area.

[0027] S302 is to assign the points in each planting area to the nearest cluster center: After determining the initial cluster center, the next step is to assign all points in the planting area, such as soil moisture, temperature and other sensor data collection points, to the cluster center closest to them. This process can be completed by calculating the distance from each point to each cluster center, and assigning each point to the cluster center closest to it.

[0028] S303 is to update the location of the cluster center to the average location of all points in the cluster: Once all points are assigned to the corresponding cluster centers, the next step is to recalculate the location of each cluster center. The new cluster center location should be the average value of the coordinates of all points in the cluster, so that the cluster center can more accurately represent the characteristics of the points it contains.

[0029] S304 is to stop iteration when the change of the cluster center is less than the preset threshold: the above steps will be repeated until the position change of the cluster center is less than the preset threshold. This threshold can be set according to actual needs, usually to ensure the stability of the clustering results. When this condition is met, the algorithm will stop iterating, and the cluster center obtained at this time is the final monitoring point position.

[0030] Through the above steps, the Kmeans clustering algorithm can be effectively used to determine the number and approximate locations of monitoring points in the optimization scheme for the coordinated layout of sensors and mobile devices in smart agriculture.

[0031] Next, refer to Figure 4 Describe the specific steps of stopping iteration when the change of cluster center is less than a preset threshold value in the optimization method of the coordinated layout of sensors and mobile devices in smart agriculture according to claim 3: S401 is to calculate the distance difference between each cluster center after the current iteration and the cluster center of the previous iteration after each iteration. This process can be implemented in many ways, such as using the Euclidean distance formula to measure the spatial distance between two cluster centers to obtain the distance difference.

[0032] S402 is to check whether the distance difference of all cluster centers is less than a preset threshold value. If the distance difference of all cluster centers is less than the threshold value, it is considered that the cluster centers have stabilized and further iterations can be stopped. This judgment process can be implemented by writing a conditional statement, such as using an if statement to compare the distance difference and the threshold value.

[0033] S403 is to set the threshold value as a percentage of the average distance between monitoring points, for example, 5%. This means that if the change in the cluster center is less than 5% of the average distance between monitoring points, the cluster center is considered to have converged. The setting of the threshold value can be completed by statistically analyzing the average distance between monitoring points and calculating the corresponding percentage.

[0034] Next, refer to Figure 5 Describe the specific steps of setting the threshold as a percentage of the average distance between monitoring points in an optimization method for the coordinated layout of sensors and mobile devices in smart agriculture: S501 is to calculate the sum of the distances between all monitoring points. This process involves traversing all monitoring points and calculating the spatial distance between each pair of monitoring points, and finally accumulating these distances to obtain the sum. For example, in a scene containing N monitoring points, it is necessary to calculate the distances between N*N-1 / 2 pairs of monitoring points.

[0035] S502 is to calculate the average distance between monitoring points. This step is completed by dividing the sum of the distances between all monitoring points obtained in S501 by the number of monitoring point pairs. For example, if there are N monitoring points in total, the number of monitoring point pairs is N*N-1 / 2, and the average distance is the total distance divided by this number.

[0036] S503 is to set the threshold value to the average distance multiplied by a preset percentage. For example, the threshold value can be set to 50% of the average distance. This means that if the distance between two monitoring points is less than half of the average distance, they are considered close enough to be considered in the same area.

[0037] S504 is to determine whether the distance difference of all cluster centers is less than the threshold. Once the threshold is set, the algorithm will check whether the distance difference between all cluster centers is less than the threshold. If the distance difference of all cluster centers is less than the threshold, it is considered that the algorithm has converged, that is, the layout of the monitoring points has reached a stable state.

[0038] Through the above steps, it can be effectively determined whether the layout of monitoring points has reached the optimal or stable configuration, thereby ensuring that the sensors in the smart agriculture system can work together efficiently.

[0039] Next, refer to Figure 6 Describe the specific steps of cluster center convergence judgment in an optimization method for the coordinated layout of sensors and mobile devices in smart agriculture: S601 is to calculate the distance difference between each cluster center and the position of the previous iteration. This process involves comparing the position coordinates of the current cluster center with the position coordinates of the previous iteration, and calculating the Euclidean distance or Manhattan distance between them to quantify the degree of position change.

[0040] S602 is to check whether the distance difference of all cluster centers is less than a preset threshold value. This step is completed by traversing the distance difference values ​​of all cluster centers and comparing these differences with the threshold value. If all the distance differences are less than the threshold value, it can be considered that the cluster centers have stabilized, i.e., have reached a convergence state.

[0041] S603 is to calculate the threshold value based on the average distance and a percentage. Specifically, the average distance between all cluster centers needs to be calculated first, and then multiplied by a pre-set percentage, such as 0.01 representing 1%, as the threshold value for determining whether the cluster center is stable.

[0042] The technical implementation of the above features is illustrated as follows: Suppose we have three cluster centers C1, C2 and C3, and their positions after the nth iteration are 10, 20, 30, 40 and 50, 60 respectively, while their positions at the n-1th iteration are 9, 19, 31, 41 and 51, 61 respectively. We first calculate the position change of each cluster center between two iterations. For example, for C1, the position change is sqrt10-9^2 + 20-19^2 = sqrt2 ≈ 1.41.

[0043] Assuming that the threshold we set is 1% of the average distance, we need to first calculate the average distance between all cluster centers. Assuming that the average distance is 100 unit lengths, the threshold is 100 * 0.01 = 1 unit length. Next, we check whether the position change of each cluster center is less than 1 unit length. If so, the cluster centers are considered to have converged.

[0044] In this example, if the position change of all cluster centers is less than 1 unit length, the cluster centers are considered stable, and the algorithm can stop iterating and output the final result. Otherwise, the next round of iteration will continue until the convergence condition is met.

[0045] Next, refer to Figure 7 Describe the specific steps of an optimization method for the coordinated layout of sensors and mobile devices in smart agriculture: S701 is to determine the starting position of the mobile device and the position of the target sensor node. At this stage, it is necessary to clarify the initial point where the mobile device starts to perform the task and the specific location information of each sensor node. This location information can be obtained through GPS coordinates or other positioning technologies, and is usually stored in the system database in advance for subsequent processing.

[0046] S702 is to establish a grid model of the farmland area and map each sensor node to the corresponding grid unit. In order to simplify the problem and improve the calculation efficiency, the entire farmland is divided into multiple grid units, and each sensor node is assigned to the nearest grid unit. In this way, the actual spatial problem can be converted into a discretized grid problem, which is convenient for the application of the subsequent path planning algorithm.

[0047] S703 is to use Dijkstra algorithm or A* search algorithm to calculate the shortest path from the starting position to each sensor node. In this step, it is very important to choose a suitable path planning algorithm. For example, if the sensor nodes are evenly distributed and there are few obstacles, Dijkstra algorithm may be a better choice; when there are many obstacles in the environment, A* search algorithm can often find the optimal path more efficiently because it considers heuristic information such as the estimated distance from the target node.

[0048] S704 is to generate the optimal data collection route of the mobile device based on the calculated shortest path set. After obtaining the shortest path from the starting position to each sensor node, it is necessary to further combine these paths through a certain strategy to ensure that the mobile device can visit all sensor nodes with the minimum total travel distance. This step may involve the solution method of classic problems such as the traveling salesman problem TSP.

[0049] For example, in a specific embodiment, suppose we have a rectangular farmland with 10 sensor nodes deployed to monitor soil moisture, temperature and other information. First, we determined that the starting position of the mobile device is located in a corner of the farmland and recorded the precise coordinates of each sensor node. Next, we divided the entire farmland into a 50×50 grid model and mapped each sensor node to the nearest grid unit. Subsequently, the A* search algorithm was used to calculate the shortest path from the starting position to each sensor node. Finally, these shortest paths were combined and optimized through methods such as greedy algorithms or genetic algorithms to obtain the optimal data collection route for the mobile device to access all sensor nodes.

[0050] Next, refer to Figure 8 Describe the specific steps of using Dijkstra algorithm or A* search algorithm to calculate the shortest path from the starting position to each sensor node in an optimization method for the coordinated layout of sensors and mobile devices in smart agriculture: S801 is to initialize the distance matrix: In this step, the system first creates a matrix representing the distance between each node in the grid. For the distance from the starting position to all other grid cells, the initial value is set to infinity ∞ to indicate that no valid path to these nodes has been found; and the distance from the starting position to itself is set to 0, because it can reach itself without passing through any intermediate nodes.

[0051] S802 is to add the starting position to the unvisited node set and set it as the current node: At this stage, the starting position is marked as the node currently being processed and is added to a node set containing all nodes whose shortest paths have not yet been determined. This step ensures that the algorithm can gradually explore the optimal path starting from the starting position.

[0052] S803 is to check each neighbor node of the current node: the system will traverse all neighbor nodes of the current node and calculate the distance from the current node to these neighbor nodes. If the distance from the current node to a neighbor node is shorter than the shortest distance previously recorded, the shortest distance of the neighbor node is updated.

[0053] S804 is to update the shortest distance of the neighbor node: if the newly calculated distance to a neighbor node through the current node is less than the known shortest distance, the shortest distance of the neighbor node is updated, which means that a better path to the neighbor node is found.

[0054] S805 is to repeat the above process until all nodes have been visited: the algorithm will continue to select the next unvisited node with the smallest known distance as the new current node, and repeat S803 and S804 until all nodes have been visited and their shortest paths to the starting position have been determined.

[0055] Through the above steps, the shortest path from the starting position to each sensor node can be effectively calculated, thereby optimizing the collaborative layout of sensors and mobile devices in smart agriculture.

[0056] Next, refer to Fig. 9 Describe the specific steps of updating the shortest path of neighbor nodes in an optimization method for the coordinated layout of sensors and mobile devices in smart agriculture: S901 starts from the current node and calculates the actual distance to each neighboring node. This process is usually completed by measurement or based on a preset network topology. For example, in a smart agriculture scenario, the actual distance can be estimated by wireless signal strength, GPS coordinates, or preset distances between nodes.

[0057] S902 is to calculate the total distance from the starting position to the neighboring node, which involves adding the actual distance from the current node to the neighboring node and the distance from the starting position to the current node. In practical applications, the distance from the starting position to the current node can be a value that has been calculated and recorded before.

[0058] S903 is to compare the total distance calculated with the shortest distance recorded by the neighboring nodes. If the total distance calculated is less than the shortest distance recorded by the neighboring nodes, the shortest distance of the neighboring nodes is updated to the total distance. This comparison and update operation ensures that the path selection algorithm can find the shortest path from the starting position to each node.

[0059] S904 is to check whether the calculated total distance satisfies "total distance < shortest distance recorded by neighboring nodes". If the condition is met, the update operation in S903 is performed; if not, the shortest distance of neighboring nodes is kept unchanged. This mechanism ensures the efficiency of the algorithm and avoids unnecessary calculations.

[0060] Through the above steps, the present invention provides an effective method to optimize the collaborative layout between sensors and mobile devices in smart agriculture, thereby improving the efficiency of data collection and transmission.

[0061] Next, refer to Fig.10 Specific steps to describe the features of an optimization method for the coordinated layout of sensors and mobile devices in smart agriculture: S1001 is a method, when executing the path optimization algorithm, for each neighbor node of the current node, the total distance from the current node to the neighbor node is calculated. The total distance here refers to the sum of the distances from the starting node through a series of intermediate nodes to the current neighbor node.

[0062] S1002 is to compare the calculated total distance with the shortest distance recorded by the neighboring node. If the calculated total distance is less than the shortest distance recorded by the neighboring node, then perform subsequent operations to update the information of the neighboring node.

[0063] S1003 is to update the shortest distance of the neighbor node to the smaller total distance calculated. This step ensures that the neighbor node records the shortest path to the node.

[0064] S1004 is marking the neighbor node as updated, indicating that the shortest distance information of the node has been optimized. This mark helps to avoid repeated calculations.

[0065] S1005 is to check whether the neighbor node is already in the unvisited node set. If the neighbor node has not been added to the unvisited node set, it is added to the set. The unvisited node set is used to store the nodes to be processed so that the algorithm can continue to find the shortest path to other nodes.

[0066] S1006 is to add the neighbor node to the unvisited node set again if it is found in S1005 that the neighbor node is not in the unvisited node set. This step is to ensure that all possible paths are considered and to ensure the integrity of the algorithm.

[0067] To illustrate how this feature is technically implemented, let's assume that we are using the Dijkstra algorithm to optimize the layout of sensor networks in smart agriculture. In a specific example, assume that we have a starting node S, which is connected to three neighboring nodes A, B, and C. Initially, the shortest distances of all neighboring nodes are set to infinity except for the starting node S, which has a shortest distance of 0. When the algorithm starts running, it first calculates the direct distances from S to A, B, and C, assuming that these distances are 10, 20, and 15, respectively. The algorithm then checks whether these distances are less than the shortest distance currently recorded by the neighboring nodes. Since these distances are all infinite at the beginning, the new distances are accepted and updated. For example, the shortest distance of node A will be updated to 10, and A will be marked as updated and added to the set of unvisited nodes. The algorithm then continues to process the nodes in the set of unvisited nodes until the shortest path to all nodes is found.

[0068] In actual operation, when this device is used, it is first necessary to divide the farmland area in detail, and determine which locations are the key monitoring points for sensor deployment in combination with the specific distribution of crops. The selection of these key monitoring points is crucial because they directly affect the quality and efficiency of subsequent data collection. Once the location of the monitoring point is determined, the corresponding sensor nodes will be deployed next. These sensors will be responsible for collecting information about environmental parameters such as soil moisture, temperature, and light intensity, and transmitting these data to mobile devices or other receiving terminals through wireless communication technology. In order to ensure that the mobile device can efficiently access all sensor nodes and collect data, the present invention also proposes a method based on a path planning algorithm to calculate the optimal data collection route. The path planning algorithm takes into account factors such as terrain, obstacle distribution, and the distance between sensor nodes, thereby generating a path that can cover all key monitoring points and has the shortest total travel distance. This not only reduces the energy consumption of the mobile device, but also increases the speed of data collection. In addition, the present invention also introduces a machine learning algorithm to further improve the intelligence level of the system. Through the analysis of historical data and the monitoring of current environmental changes, the system can dynamically adjust the working parameters of the sensor such as the sampling frequency and the inspection frequency of the mobile device. For example, during the critical period of crop growth or when encountering abnormal weather conditions, the system may automatically increase the frequency of data collection in order to obtain relevant information in a more timely manner. This dynamic adjustment mechanism not only helps to improve the accuracy and timeliness of data collection, but also effectively reduces unnecessary energy consumption, thereby achieving rational allocation and utilization of resources. In summary, the present invention provides an efficient optimization method for the coordinated layout of sensors and mobile devices for smart agriculture by comprehensively using sensor technology, path planning algorithms, and machine learning technology, which greatly improves the level of intelligence in agricultural production management.

[0069] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of the embodiments of the present disclosure. It should be understood that the above description is only the specific implementation method of the embodiments of the present disclosure and is not intended to limit the protection scope of the embodiments of the present disclosure. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present disclosure should be included in the protection scope of the embodiments of the present disclosure.

Claims

1. A method for optimizing the coordinated layout of sensors and mobile devices in smart agriculture, characterized in that: include: Determine the key monitoring points for sensor deployment based on farmland area division and crop distribution; Based on the path planning algorithm of the mobile device, the optimal data collection route is calculated to ensure that the mobile device can efficiently access all sensor nodes; Machine learning algorithms are used to analyze historical data and environmental changes, and dynamically adjust the operating parameters of sensors and the inspection frequency of mobile devices to improve the accuracy and timeliness of data collection.

2. The optimization method for the coordinated layout of sensors and mobile devices in smart agriculture according to claim 1 is characterized in that: According to the farmland area division and crop distribution, the key monitoring points for sensor deployment are determined as follows: Identify different crop planting areas based on farmland topographic maps and crop type distribution maps; Prioritize each planting area based on crop growth cycle and irrigation needs; Using the Kmeans clustering algorithm, the number and approximate locations of monitoring points are determined based on priority and area; The specific locations of the monitoring points are adjusted by minimizing the sum of the Euclidean distances between the monitoring points.

3. The optimization method for the coordinated layout of sensors and mobile devices in smart agriculture according to claim 2 is characterized in that: The Kmeans clustering algorithm is used to determine the number and approximate locations of monitoring points, including: Initialize the cluster centers and select random locations equal to the number of monitoring points as the initial cluster centers; Assign points within each planting area to the nearest cluster center; Update the location of the cluster center to the average location of all points in the cluster; The iteration stops when the change of cluster center is less than the preset threshold.

4. The optimization method for the coordinated layout of sensors and mobile devices in smart agriculture according to claim 3 is characterized in that: When the change of cluster center is less than the preset threshold, the iteration is stopped, including: Calculate the distance difference between each cluster center after the current iteration and the cluster center of the previous iteration; If the distance difference of all cluster centers is less than the threshold, the iteration is stopped; The threshold is set as a percentage of the average distance between monitoring points, for example 5%.

5. The optimization method for the coordinated layout of sensors and mobile devices in smart agriculture according to claim 4 is characterized in that: Thresholds set as percentages of the average distance between monitoring points include: Calculate the sum of distances between all monitoring points; Calculate the average distance between monitoring points; Set the threshold to the average distance multiplied by the percentage, for example 0.5; If the distance difference of all cluster centers is less than the threshold, it is considered converged.

6. The optimization method for the coordinated layout of sensors and mobile devices in smart agriculture according to claim 5 is characterized in that: If the distance difference of all cluster centers is less than the threshold, it is considered convergent: For each cluster center, calculate the distance difference between it and the previous iteration position; If the distance difference of all cluster centers is less than the threshold, the cluster center is considered stable; The threshold calculation formula is: Threshold = Average distance * Percentage.

7. The optimization method for the coordinated layout of sensors and mobile devices in smart agriculture according to claim 1 is characterized in that: The path planning algorithm based on the mobile device calculates the optimal data collection route to ensure that the mobile device can efficiently access all sensor nodes, including: Determine the starting location of the mobile device and the location of the target sensor node; Establish a grid model of the farmland area and map each sensor node to the corresponding grid unit; Use Dijkstra algorithm or A* search algorithm to calculate the shortest path from the starting location to each sensor node; Based on the calculated shortest path set, the optimal data collection route for the mobile device is generated.

8. The optimization method for the coordinated layout of sensors and mobile devices in smart agriculture according to claim 1 is characterized in that: The method of calculating the shortest path from the starting position to each sensor node using the Dijkstra algorithm or the A* search algorithm includes: Initialize the distance matrix, setting the distance from the starting position to all other grid cells to infinity, except for the starting position itself to 0; Add the starting position to the unvisited node set and set it as the current node; For each neighbor node of the current node, if the distance from the current node to the neighbor node is less than the known shortest distance, update the distance of the neighbor node; If the distance of a neighbor node satisfies "newly calculated distance < known shortest distance", the shortest distance of the neighbor node is updated.

9. The optimization method for the coordinated layout of sensors and mobile devices in smart agriculture according to claim 1, characterized in that: If the distance of the neighbor node satisfies "the newly calculated distance < the known shortest distance", updating the shortest distance of the neighbor node includes: Calculate the actual distance from the current node to the neighboring node; Calculate the total distance from the starting position to the neighboring node, that is, "the actual distance from the current node to the neighboring node + the distance from the starting position to the current node"; If the calculated total distance is less than the shortest distance recorded by the neighbor node, the shortest distance of the neighbor node is updated to the total distance; If the calculated total distance satisfies "total distance < shortest distance recorded by neighboring nodes", the shortest distance of neighboring nodes is updated.

10. The optimization method for the coordinated layout of sensors and mobile devices in smart agriculture according to claim 1, characterized in that: If the calculated total distance satisfies "total distance < shortest distance recorded by neighboring nodes", updating the shortest distance of neighboring nodes includes: Update the shortest distance of neighbor nodes to the calculated total distance; Mark neighbor nodes as updated; If the neighbor node is not in the unvisited node set, add it to the unvisited node set; If the neighbor node is not in the unvisited node set, add it to the unvisited node set.

Images