Intelligent agent collaboration method, device and system and agricultural task execution method and device

Abstract

The invention provides an agent collaboration method, device and system and an agricultural task execution method and device, and relates to the technical fields of artificial intelligence, natural language processing, computer vision and the like. According to the specific implementation scheme, the method comprises the steps of collecting multi-source agricultural data of a crop in a historical period and a current growth stage; constructing a spatial-temporal characteristic spectrum based on the multi-source agricultural data; calculating a crop stress index and a crop growth index based on the spatial-temporal characteristic spectrum; determining a target agent from an agent set based on the crop stress index and the crop growth index; and sending the crop growth index and the crop stress index to the target agent, so that the target agent executes a corresponding agricultural task on the crop based on the crop growth index and the crop stress index, and links related agents in the agent set based on environmental data, and the globality and collaboration of agricultural decision making are improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to natural language processing, artificial intelligence, computer vision, and other technical fields. Specifically, the present disclosure provides an intelligent agent collaboration method and device, a system, an agricultural task execution method and device, an electronic device, and a computer-readable storage medium. Background Art

[0002] With the advancement of agricultural modernization, crop models and agricultural information systems are playing an increasingly important role in agricultural production. Traditional crop growth models are primarily based on mechanistic modeling. These models establish mathematical equations based on crop physiological and ecological principles to simulate the entire crop growth process, from sowing to harvest. However, these models rely on extensive parameter calibration and often require expert input, making them difficult to adapt to complex environmental changes and large-scale application.

[0003] In recent years, with the development of artificial intelligence, particularly deep learning, a number of agricultural AI models have been proposed to solve specific problems. Examples include using convolutional neural networks for pest and disease identification, using time series models for yield forecasting, and monitoring soil moisture using remote sensing data. However, these "small models" are often optimized for specific tasks and lack cross-task collaboration, creating so-called "task silos."

[0004] Furthermore, traditional models lack the ability to adapt to dynamic environmental changes and are unable to automatically adjust model parameters to address differences in different regions or climate conditions. Furthermore, due to the diverse sources and complex structures of agricultural data, effectively integrating multimodal and heterogeneous data such as remote sensing, IoT sensors, weather stations, and field management logs has become a challenge in current research. Summary of the Invention

[0005] The present disclosure provides an intelligent agent collaboration method and device, a system, an agricultural task execution method and device, an electronic device, and a computer-readable storage medium.

[0006] According to the first aspect, an intelligent agent collaboration method is provided, which includes: collecting multi-source agricultural data of crops in historical periods and current growth stages; constructing a spatiotemporal feature map based on the multi-source agricultural data; calculating a crop stress index and a crop growth index based on the spatiotemporal feature map; determining a target intelligent agent from an intelligent agent set based on the crop stress index and the crop growth index; sending the crop growth index and the crop stress index to the target intelligent agent, so that the target intelligent agent performs corresponding agricultural tasks on the crops based on the crop growth index and the crop stress index, and links relevant intelligent agents in the intelligent agent set based on environmental data.

[0007] According to the second aspect, a method for executing an agricultural task is provided, which includes: receiving a crop growth index and a crop stress index of a crop; executing an agricultural task on the crop based on the crop growth index and the crop stress index; obtaining environmental data when executing the agricultural task; and linking relevant intelligent agents in a set of intelligent agents based on the environmental data.

[0008] According to the third aspect, an intelligent agent collaboration device is provided, which includes: a collection unit configured to collect multi-source agricultural data of crops in historical periods and current growth stages; a map construction unit configured to construct a spatiotemporal feature map based on the multi-source agricultural data; an index construction unit configured to calculate a crop stress index and a crop growth index based on the spatiotemporal feature map; a determination unit configured to determine a target intelligent agent from an intelligent agent set based on the crop stress index and the crop growth index; a sending unit configured to send the crop growth index and the crop stress index to the target intelligent agent, so that the target intelligent agent performs corresponding agricultural tasks on the crops based on the crop growth index and the crop stress index, and links relevant intelligent agents in the intelligent agent set based on environmental data.

[0009] According to the fourth aspect, an agricultural task execution device is provided, which includes: a receiving unit configured to receive a crop growth index and a crop stress index of a crop; an execution unit configured to execute an agricultural task on the crop based on the crop growth index and the crop stress index; an acquisition unit configured to acquire environmental data when executing the agricultural task; and a linkage unit configured to link relevant intelligent agents in a set of intelligent agents based on the environmental data.

[0010] According to the fifth aspect, an intelligent agent collaborative system is provided, which includes: a central intelligent agent and an intelligent agent set; the central intelligent agent is used to collect multi-source agricultural data of crops in historical periods and current growth stages; based on the multi-source agricultural data, a spatiotemporal feature map is constructed; based on the spatiotemporal feature map, a crop stress index and a crop growth index are calculated; based on the crop stress index and the crop growth index, a target intelligent agent is determined from the intelligent agent set; the crop growth index and the crop stress index are sent to the target intelligent agent; the target intelligent agent performs corresponding agricultural tasks on crops based on the crop growth index and the crop stress index, and links relevant intelligent agents in the intelligent agent set based on environmental data.

[0011] According to the sixth aspect, an electronic device is provided, which includes: at least one processor; and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method described in any implementation of the first aspect or the second aspect.

[0012] According to a seventh aspect, a non-transitory computer-readable storage medium storing computer instructions is provided, where the computer instructions are used to cause a computer to execute the method described in any implementation of the first aspect or the second aspect.

[0013] The disclosed embodiments provide an agent collaboration method and device. First, multi-source agricultural data from historical and current crop growth stages is collected. Second, a spatiotemporal feature map is constructed based on the multi-source agricultural data. Third, a crop stress index and a crop growth index are calculated based on the spatiotemporal feature map. Then, a target agent is identified from a set of agents based on the crop stress index and crop growth index. Finally, the crop growth index and crop stress index are sent to the target agent, enabling the target agent to perform agricultural tasks on the crops based on the crop growth index and crop stress index, and to coordinate with related agents in the agent set based on environmental data. This method constructs a spatiotemporal feature map from multi-source agricultural data and further generates a crop stress index and crop growth index, which accurately reflect the growth status and stress conditions faced by the crops. Based on these indices, a target agent is identified from an initial agent and relevant data is sent to the target agent, enabling it to perform agricultural tasks based on the accurate data and coordinate with related agents in the agent set, thus achieving intelligent, precise, and collaborative agricultural production. This not only improves agricultural production efficiency and reduces labor costs, but also enables timely response to various problems in the crop growth process, improves crop yield and quality, enhances the stability and sustainability of agricultural production, and provides strong support for the development of smart agriculture.

[0014] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings are used to better understand the present invention and do not constitute a limitation of the present invention. Figure 1 is a flow chart of an embodiment of the intelligent agent collaboration method according to the present disclosure; Figure 2 It is a schematic diagram of a structure of collaborative work between a set of intelligent agents in the present disclosure; Figure 3 is a flow chart of an embodiment of a method for performing an agricultural task according to the present disclosure; Figure 4 It is a structural diagram of an embodiment of the intelligent agent collaboration device disclosed in the present invention; Figure 5 It is a structural schematic diagram of an embodiment of the agricultural task execution device disclosed herein; Figure 6It is a structural diagram of the intelligent agent collaborative system disclosed in the present invention; Figure 7 It is a block diagram of an electronic device used to implement the intelligent agent collaboration method of the embodiment of the present disclosure. DETAILED DESCRIPTION

[0016] Unless expressly stated otherwise, throughout the specification and claims, the term "comprise" or variations such as "include" or "comprising", etc., will be understood to include the stated elements or components but not to exclude other elements or other components.

[0017] The technical solutions of the present disclosure are described below through specific examples. It should be understood that one or more steps mentioned in the present disclosure do not exclude the existence of other methods and steps before and after the combination step, or other methods and steps may be inserted between these explicitly mentioned steps. It should also be understood that these examples are only used to illustrate the present disclosure and are not used to limit the scope of the present disclosure. Unless otherwise specified, the numbering of each method step is only for the purpose of identifying each method step, and does not limit the order of arrangement of each method or limit the scope of implementation of the present disclosure. Changes or adjustments in their relative relationships can also be regarded as the scope of implementation of the present disclosure without substantial changes in the technical content.

[0018] The sources of the raw materials and instruments used in the examples are not particularly limited and can be purchased from the market or prepared according to conventional methods known to those skilled in the art.

[0019] Problems faced by the agricultural field in existing technologies: Weak data fusion capabilities: Agricultural data comes from various sources, including remote sensing images, sensors, meteorological data, and agricultural operation records. Existing models find it difficult to efficiently integrate these multimodal data, which affects the model's prediction accuracy and decision-making quality.

[0020] Severe task isolation: Existing agricultural models operate independently. For example, the plant protection model is only used for pest and disease control, the irrigation model is only used for water management, and the nutrition model only simulates nitrogen balance. There is a lack of information exchange and collaborative decision-making mechanisms between them, making it difficult to form a globally optimal planting management strategy.

[0021] High reliance on manual intervention: Model construction and parameter calibration rely heavily on expert experience and lack the ability to self-feedback and correct errors based on real-time data. This results in poor model adaptability and makes it difficult to cope with climate change and regional differences. Traditional crop models are based on limited experimental data and struggle to accurately simulate complex crop-environment interactions, especially under atypical climate conditions.

[0022] To address the shortcomings of traditional technologies, this paper proposes a collaborative approach for intelligent agents. This collaborative processing method fully utilizes data fusion capabilities, information exchange, and collaborative decision-making mechanisms. Based on the self-feedback and correction capabilities of real-time data, it improves agricultural production efficiency, reduces labor costs, and can promptly respond to various issues during crop growth, thereby increasing crop yield and quality, enhancing the stability and sustainability of agricultural production, and providing strong support for the development of smart agriculture. Figure 1 A process 100 according to an embodiment of the agent collaboration method of the present disclosure is shown. The agent collaboration method includes the following steps: Step 101: Collect multi-source agricultural data of crops in historical periods and current growth stages.

[0023] In this embodiment, multi-source agricultural data refers to a collection of data from different sources and types related to crops from historical periods and current crop growth stages. This data may include meteorological data (such as temperature, humidity, and rainfall), soil data (such as soil fertility, pH, and texture), remote sensing imagery (for monitoring crop growth and land use), agricultural production activity data (such as sowing times, fertilizer application rates, and irrigation records), and market data (such as agricultural product prices and market demand). The integration of multi-source agricultural data can provide a more comprehensive perspective and richer information for agricultural production, facilitating precision agriculture management and decision support.

[0024] In this embodiment, the intelligent agent collaboration method may be run on a central intelligent agent that controls each intelligent agent in the intelligent agent set. By coordinating the intelligent agents in the intelligent agent set and the collaborative work between related intelligent agents, intelligent control of crops can be achieved.

[0025] In this embodiment, collecting multi-source agricultural data from historical and current crop growth stages can be achieved by integrating various technologies and methods. First, multi-source agricultural data from historical crop periods is obtained by screening historical data. Satellite remote sensing technology is used to obtain macro-data such as soil moisture, vegetation index, and land use type during the current crop growth stage. Simultaneously, a network of IoT sensors deployed in farmland collects real-time soil temperature, moisture, nutrient content, and meteorological data such as temperature, precipitation, and light intensity during the current crop growth stage. Furthermore, low-altitude remote sensing using drones is combined to capture high-resolution farmland images for accurate identification of detailed information such as crop growth status, pest and disease distribution, and other details during the current crop growth stage. Agricultural experts then conduct field research to collect humanistic management data such as crop varieties, planting patterns, and irrigation and fertilization records. Finally, this data from various channels is aggregated into an agricultural big data platform. After cleaning, integration, and analysis, it provides comprehensive and accurate data support for agricultural production decision-making, contributing to the development of smart agriculture.

[0026] Step 102: construct a spatiotemporal feature map based on multi-source agricultural data.

[0027] In this embodiment, a spatiotemporal feature map is a tool or model that visualizes and analyzes data characteristics across time and space. It can demonstrate the patterns and patterns of data changes over time in different geographic locations (spatial dimensions). For example, in agriculture, a spatiotemporal feature map can show the seasonal distribution of crop growth in a given region, or the changing trends in soil fertility across different plots of land over different years. By constructing a spatiotemporal feature map, key factors influencing agricultural production and their dynamic changes can be more intuitively identified, providing a scientific basis for optimizing the allocation of agricultural resources, early warning of pests and diseases, and yield forecasting.

[0028] In this embodiment, a spatiotemporal feature map is constructed based on multi-source agricultural data. First, the data from different sources must be standardized to ensure consistency in data format, unit, and temporal resolution. Then, a Geographic Information System (GIS) is used to associate the data with geographic location information to form a spatialized first dataset. Next, time series analysis methods are used to extract the temporal characteristics of the first dataset, such as seasonal fluctuations and long-term trends. Finally, spatial and temporal characteristics are combined, and visualization tools (such as maps and charts) are used to generate a spatiotemporal feature map for the first dataset. This intuitively displays the dynamic changes in agricultural data across different regions and time periods, providing strong support for agricultural production decision-making.

[0029] Optionally, step 102 further includes: before constructing the spatiotemporal feature map, associating the multi-source agricultural data with the agricultural knowledge map to obtain associated data, standardizing the associated data, and associating the associated data with geographic location information using a geographic information system to form a spatialized second dataset; extracting temporal variation characteristics of the second dataset using a time series analysis method. Finally, combining the spatial and temporal features and using visualization tools to generate a spatiotemporal feature map from the second dataset.

[0030] Step 103: Calculate the crop stress index and the crop growth index based on the spatiotemporal characteristic map.

[0031] In this embodiment, the Crop Stress Index (CSI) is an indicator used to quantify the degree of stress on crops. Stress refers to the impact of adverse environmental factors (such as drought, high temperature, salinity, pests and diseases) on crops during their growth process. The higher the stress index, the more severe the stress on the crop, and the greater the negative impact on its growth and development.

[0032] In this embodiment, the Crop Growth Index (CGI) is an indicator used to measure crop growth and health. It typically combines the comprehensive effects of crop physiological characteristics (such as chlorophyll content, leaf area, and biomass) and environmental factors (such as light, temperature, and moisture) on crop growth. A higher CGI indicates better crop growth and health.

[0033] In this embodiment, CSI and CGI are time-series change indicators ranging from 0 to 1. In CGI, 0 represents the beginning of seed germination after sowing, and 1 represents crop maturity. In CSI, 0 indicates severe stress and abnormal growth, and 1 indicates normal growth of unstressed crops. The plant protection, irrigation, nutrition, and yield estimation intelligent agents dynamically adjust their strategies based on CSI and CGI. During the crop growth process, CGI changes from 0 to 1, and the corresponding plant protection plan, irrigation plan, nutrition plan, and dynamic yield estimation are triggered as the value changes.

[0034] In this embodiment, crop stress index and crop growth index are calculated based on a spatiotemporal feature map. First, key spatiotemporal feature parameters, such as soil moisture, temperature, light intensity, and leaf color, need to be extracted from the map. Then, a corresponding mathematical model is established based on the known relationships between these parameters and crop growth and stress. For example, the contribution of drought stress to the index is determined by analyzing the degree of inhibition of crop growth when soil moisture falls below a certain threshold. Simultaneously, changes in the spectral reflectance of leaf color are combined to assess crop health and assign a value to the growth index. Finally, these weighted parameters are combined to form an index value that reflects the overall stress and growth status of the crop.

[0035] Step 104 : determining a target intelligent agent from the intelligent agent set based on the crop stress index and the crop growth index.

[0036] In this embodiment, the execution entity dynamically adjusts resources based on real-time changes in the Crop Stress Index (CSI) and the Crop Growth Index (CGI). There are two common approaches to resource scheduling: one that determines the crop growth stage based on the CGI and, based on the CGI, invokes corresponding agent tasks at different stages of crop growth to implement plant protection, irrigation, nutrition, and dynamic yield estimation measures. The other approach triggers the corresponding target agent based on the growth stage corresponding to the CGI when the CSI is less than 1.

[0037] In this embodiment, an agent is a system or model that can perceive, make decisions, and execute specific tasks. An agent can be a sensor network monitoring crop health, an algorithmic model analyzing satellite remote sensing data, or a decision-making system with pre-set rules and algorithms.

[0038] In this embodiment, the crop stress index and the crop growth index are two key indicators for evaluating intelligent agents. The crop stress index may reflect the degree of influence of adverse environmental factors (such as drought, pests and diseases, etc.) faced by crops during the growth process on their growth. A higher value usually indicates a greater degree of stress; while the crop growth index measures the growth status and health level of crops. A higher value indicates better growth conditions. In a specific implementation, the system will comprehensively consider these two indexes according to preset rules or algorithms, evaluate and screen the initial intelligent agents, and obtain the target intelligent agent. For example, an intelligent agent with a lower stress index and a higher growth index may be preferentially selected as the target intelligent agent, because this means that the intelligent agent can still maintain a good growth state when facing an adverse environment, and is therefore considered to be the best choice. Among them, the intelligent agent set includes multiple intelligent agents, such as Figure 2 As shown, the intelligent agents include: plant protection intelligent agent, irrigation intelligent agent, nutrition intelligent agent, and yield intelligent agent, among which the plant protection intelligent agent is used for disease identification and disease prevention; the irrigation intelligent agent is used to balance water and provide irrigation plans for crops; the nutrition intelligent agent is used to balance nutrients and provide fertilization recommendations; the yield intelligent agent is used to predict crop yields and generate yield maps of crops in various regions; the target intelligent agent is at least one intelligent agent with specific functions selected by the central intelligent agent from multiple intelligent agents, that is, the target intelligent agent includes: one or more intelligent agents, each of which has corresponding functions.

[0039] The plant protection agent uses spore counters, insect monitoring instruments, and image recognition algorithms to monitor the occurrence and development of pests and diseases. It also combines meteorological data to predict the risk of disease spread and generate preventive measures in advance. The CSI and CGI indices trigger the agent to perform corresponding agricultural tasks at different stages: 0: After sowing and before seedling emergence—weeding; 0.1: Seedling emergence—efficiency enhancement and weeding; 0.3: Jointing stage—aphids, two-spotted leaf beetles, corn borers, foliar diseases, and aerial spray adjuvants; 0.5: Tasseling and silking stage—stress resistance and quality improvement, aphids, two-spotted leaf beetles, corn borers, and foliar diseases. Furthermore, pest and disease monitoring and early warning programs are run based on the CSI, monitoring equipment, and meteorological factors, allowing for targeted pest and disease control, making these agricultural tasks more professional. The agricultural tasks of the plant protection intelligent body also include: matching disease types with the compatibility of control agents, and building a corresponding knowledge base for the corresponding pests and diseases in routine agricultural treatments and the disease types obtained through early warnings, which includes information such as introductions to pest and disease categories, occurrence patterns, and control plans.

[0040] The agricultural tasks of the irrigation agent include: real-time analysis of soil moisture, meteorological data, and crop water requirement curves; analyzing the relationship between crop water requirement and soil moisture content by combining crop stress index and growth index, and dynamically adjusting irrigation plans; and outputting irrigation recommendations to the control system to achieve automated irrigation scheduling. The irrigation agent is triggered based on different CSI and CGI values, calculates crop transpiration based on meteorological data, and calculates the required irrigation amount based on crop growth and soil moisture content, as shown in Equation (1).

[0041] Irrigation volume = Kc·ETo·A·η(1) In formula (1), Kc is the crop coefficient (reflecting the crop demand characteristics), and different CSIs correspond to different Kcs; A is the irrigation area (m 2 ); η is the irrigation efficiency (drip irrigation efficiency is 0.9); ETo is calculated according to the standard Penman-Monteith method for calculating reference crop evapotranspiration.

[0042] The Nutrition Agent's agricultural tasks include analyzing soil nutrient content, crop growth stage, and target yield; analyzing the relationship between crop nutrients and soil nutrient supply using stress and growth indices, and dynamically recommending fertilizer types, dosages, and application times; and integrating remote sensing NDVI data to determine foliar nutrient status and assist in diagnosing nutrient deficiency symptoms. The Nutrition Agent is triggered based on CGI and CSI data. The nitrogen content of crop leaves changes dynamically at different growth stages. Based on historical crop leaf nitrogen levels, the normal nitrogen content standard for each CGI stage is calculated. This is then compared with current crop leaf nitrogen levels to provide fertilization recommendations, adjusting the amount of topdressing based on normal application rates.

[0043] The yield estimator agent trains a prediction model based on crop growth models and historical yield data. Its agricultural tasks include dynamically forecasting yields based on factors such as current year climate conditions, management practices, and stress indices. It also integrates remote sensing NDVI data to support multi-scale predictions (field and regional). Based on changes in CGI and CSI values, the dynamic yield estimator agent is activated during the grain development period, utilizing NDVI, meteorological, CSI, and management data to assess yield changes.

[0044] Optionally, the above step 104 includes: based on the crop stress index, detecting whether the crop is in a stress state; in response to detecting that the crop is in a stress state, selecting an agent that can release the stress state from the agent set; at the same time, based on the crop growth index, determining the growth state of the crop, selecting an agent that is conducive to the maturity of the crop from the agent set, and using the above-mentioned agent that can release the stress state and the above-mentioned agent that is conducive to the maturity of the crop as target agents.

[0045] Step 105 , the crop growth index and the crop stress index are sent to the target agent, so that the target agent performs corresponding agricultural tasks on the crops based on the crop growth index and the crop stress index, and links related agents in the agent set based on the environmental data.

[0046] In this embodiment, the target agent refers to an intelligent device or software system with autonomous decision-making and execution capabilities in an agricultural system, such as an intelligent irrigation system, an automatic fertilization robot, or an agricultural drone. It can receive data and perform corresponding agricultural tasks on crops according to preset rules or algorithms.

[0047] In this embodiment, environmental data refers to various data related to the crop growth environment, such as soil moisture, temperature, light intensity, carbon dioxide concentration, etc. These data are collected by sensors and transmitted to the system to evaluate whether the environmental conditions are suitable for crop growth.

[0048] In this embodiment, related agents refer to other intelligent devices or systems in the agent set that work in conjunction with the target agent, such as a weather monitoring agent or a greenhouse control agent. Related agents and the target agent can collaborate through data sharing and linkage mechanisms. It should be noted that each time the target agent is determined, the related agents will also be different.

[0049] In this embodiment, in an automated agricultural management system, sensors collect crop growth and environmental data, calculating a crop growth index and a crop stress index. These indices are then transmitted to a target agent, such as an intelligent irrigation system. The target agent determines whether the crop is in its peak growth phase based on the crop growth index and whether the crop is experiencing water shortage or other stress based on the crop stress index. If the crop is found to be water-deficient, the target agent automatically initiates an irrigation task and simultaneously transmits environmental data (such as soil moisture) to a related agent, such as a greenhouse control agent. The greenhouse control agent then adjusts ventilation and shading systems based on the soil moisture data to optimize environmental conditions within the greenhouse, thereby enabling linkage between multiple agents and ensuring healthy crop growth.

[0050] In this embodiment, the intelligent agent set includes four agents: plant protection agent, irrigation agent, nutrition agent, and yield agent, which work together to influence crop growth and development. The plant protection agent relies on irrigation for moisture control, while the nutrition agent handles stress resistance, forecasting, and early warning. The irrigation agent must respond to plant protection drug efficacy, nutrient dissolution requirements, and predict critical water demand periods. The nutrition agent must adapt to irrigation migration, interactions between plant protection drugs and fertilizers, and predict fertilizer requirements. The yield agent's predictions must incorporate plant protection losses, irrigation stress, and nutrient surplus and deficit data.

[0051] The embodiment of the present disclosure provides an intelligent agent collaboration method. First, multi-source agricultural data from the historical period and current growth stage of crops are collected. Second, a spatiotemporal feature map is constructed based on the multi-source agricultural data. Third, the crop stress index and crop growth index are calculated based on the spatiotemporal feature map. Then, based on the crop stress index and crop growth index, a target intelligent agent is determined from the intelligent agent set. Finally, the crop growth index and crop stress index are sent to the target intelligent agent, so that the target intelligent agent performs corresponding agricultural tasks on the crops based on the crop growth index and crop stress index, and links related intelligent agents in the intelligent agent set based on environmental data. This collaborative processing method achieves high efficiency, accuracy, and flexibility in information processing in agricultural scenarios, can effectively improve the decision-making efficiency and management level of agricultural production, and provides strong support for the development of intelligent agriculture.

[0052] In some optional implementations of the present disclosure, the above-mentioned multi-source agricultural data include: remote sensing data, Internet of Things data, field management data, meteorological data, and pest and disease monitoring data in historical periods and current growth stages. Based on the multi-source agricultural data, constructing a spatiotemporal feature map includes: performing data preprocessing on the remote sensing data, Internet of Things data, field management data, meteorological data, and pest and disease monitoring data; using a spatiotemporal attention mechanism to align the frequency and spatial resolution of the preprocessed remote sensing data, Internet of Things data, field management data, meteorological data, and pest and disease monitoring data to obtain aligned data; and constructing a spatiotemporal feature map based on the aligned data.

[0053] In this optional implementation, if Figure 2 As shown in the figure, remote sensing data, IoT data, field management data, meteorological data, and pest and disease monitoring data are multi-source heterogeneous data. Data preprocessing includes data cleaning and cross-source alignment. Cross-source alignment involves unifying spatiotemporal benchmarks, matching resolutions, semantic alignment, and quality assessment. Through cross-source alignment, remote sensing data, IoT data, field management data, meteorological data, and pest and disease monitoring data can be transformed into high-quality fused datasets that can be directly used for agricultural monitoring, yield forecasting, or pest and disease early warning. For example, remote sensing data, field management data, meteorological data, and pest and disease monitoring data can be aligned with IoT data from multi-source agricultural data. Key data can then be extracted from these aligned data to construct spatiotemporal feature maps.

[0054] In this optional implementation, a spatiotemporal feature map is constructed by integrating agricultural data from multiple sources. Specifically, this data includes remote sensing data from historical time periods (e.g., crop growth as reflected in satellite imagery), data collected by IoT devices (e.g., soil moisture and temperature sensor data), field management records (e.g., the time and location of fertilization and irrigation operations), meteorological data (information on weather changes), and pest and disease monitoring data (information on the occurrence of pests and diseases). To integrate these different types of data into a unified spatiotemporal feature map, a spatiotemporal attention mechanism is employed. This mechanism adjusts and aligns the frequency (the time interval between data collection) and spatial resolution (the spatial accuracy of the data) of different data sources based on their temporal and spatial importance. For example, remote sensing data may have a low temporal frequency but a high spatial resolution, while IoT data may have a high temporal frequency but a limited spatial extent. The spatiotemporal attention mechanism aligns the frequency and spatial resolution of these data sources to a consistent level, generating a comprehensive spatiotemporal feature map for a more comprehensive analysis and understanding of the various factors and their interrelationships in agricultural production.

[0055] In some optional implementations of the present disclosure, the above-mentioned calculation of the crop stress index and the crop growth index based on the spatiotemporal characteristic map includes: obtaining the remote sensing normalized vegetation index and the maximum normalized vegetation index based on the remote sensing unit in the spatiotemporal characteristic map; calculating the relative value of the vegetation index based on the remote sensing normalized vegetation index and the maximum normalized vegetation index; calculating the water stress index based on the available water and crop water requirement in the spatiotemporal characteristic map; calculating the nutrient stress index based on the utilization rate of crop growth nitrogen and the theoretical nitrogen requirement of the crop in the spatiotemporal characteristic map; calculating the temperature stress index based on the actual effective accumulated temperature during the growth period and the most suitable accumulated temperature for the crop in the spatiotemporal characteristic map. index; the crop stress index is calculated based on the relative value of the vegetation index, the water stress index, the nutrient stress index and the temperature stress index; the average temperature is calculated based on the temperature in the spatiotemporal characteristic map; the soil photosynthetic capacity value is calculated based on the soil water content in the spatiotemporal characteristic map; the fertility photosynthetic capacity value is calculated based on the total nitrogen content of the soil and fertilizer in the spatiotemporal characteristic map; the precipitation photosynthetic capacity value is calculated based on the precipitation in the spatiotemporal characteristic map; the temperature photosynthetic capacity value is calculated based on the average temperature; the crop growth index is calculated based on the relative value of the vegetation index, the soil photosynthetic capacity value, the fertility photosynthetic capacity value, the precipitation photosynthetic capacity value and the temperature photosynthetic capacity value.

[0056] In this optional implementation, the stress state of crops is considered from the intrinsic aspects such as water, nutrients, pests and diseases, and the relative index value NDVI is added to reflect the crop growth state from the surface. The multimodal data is mapped into a vector representation of unified dimension to obtain the crop stress index shown in formula (2).

[0057] CSI = f(Sndvi,Sw,Sn,St)(2) In formula (2), Sndvi is the relative value of the vegetation index, Sw is the water stress index, Sn is the nutrient stress index, and St is the temperature stress index. Specifically, the calculation process of the crop stress index can be obtained using the method shown in formula (3).

[0058] CSI = a×Sndvi×b×Sw×c×Sn×d×St (3) In formula (3), a, b, c, and d are dynamic weights. Each stress factor ranges from [0–1], with 1 indicating no stress and 0 indicating severe stress. The core of the calculation can use a geometric mean rather than an arithmetic mean to reflect the nonlinear additive effects of stress factors (such as the interaction between water and nutrients). The weights a, b, c, and d are dynamically adjusted through AI learning. The stress factor is input, and the output is the yield reduction rate that intersects the historical optimal yield.

[0059] In this optional implementation, the Normalized Difference Vegetation Index (NDVI) is a remotely sensed vegetation index used to measure vegetation growth and coverage. Its value typically ranges from -1 to 1, with higher values ​​indicating better vegetation growth. The maximum NDVI, the highest value of the remotely sensed NDVI over a historical period (e.g., a crop growing season), serves as a reference.

[0060] In this optional implementation, the relative vegetation index value can be the ratio of the remote sensing normalized vegetation index to the maximum normalized vegetation index, which is used to measure the level of current vegetation growth relative to the optimal state. Optionally, the relative vegetation index value can also be a value calculated using formula (4).

[0061] Sndvi = (NDVIt-NDVImin) / (NDVImax-NDVImin) (4) In formula (3), Sndvi is the relative value of vegetation index, reflecting the strength of photosynthetic capacity, NDVIt is the current remote sensing NDVI value, NDVImax is the maximum NDVI of the plot in the same period in the past historical period, and NDVImin is the bare soil NDVI value generally 0.1.

[0062] In this optional implementation, the water stress index is an indicator calculated based on the available water and the crop water requirement, and is used to reflect the degree of stress suffered by crops due to insufficient or excessive water, as shown in formula (5), where Sw represents the water stress index.

[0063] Sw = min(1,TAWact / TAWreq)(5) In Equation (5), TAWact is the actual available water in the root zone (field management data), obtained from soil sensors (or publicly available soil moisture values ​​inverted from remote sensing). TAWreq is the crop water requirement (field management data), calculated from potential evapotranspiration (ET0) × crop coefficient Kc.

[0064] In this optional implementation, the nutrient stress index is calculated based on the nitrogen utilization rate of crop growth and the theoretical nitrogen requirement of the crop. It is used to reflect the degree of stress to the crop due to nutrient deficiency or excess. As shown in formula (6), the nutrient stress index is Sn.

[0065] Sn = min(1,Nup / Ndm)(6) In formula (6), Nup is the actual nitrogen content of crop growth (field management data), Nup = (Soil_Nmin+Fertilizer_N×Efficiency)×(1-Leaching_loss), Efficiency is the nitrogen utilization efficiency. Different crops have different nitrogen utilization efficiencies. For example, the utilization efficiency of corn is 50%. Ndm is the theoretical nitrogen requirement of the crop, Ndm = k×(1- )×Nmax, c, k are crop parameters.

[0066] In this optional implementation, the temperature stress index is calculated based on the difference between the actual effective accumulated temperature during the growth period and the optimal accumulated temperature for the crop. It is used to reflect the degree of stress experienced by the crop due to temperature discomfort. As shown in Equation (7), St represents temperature stress.

[0067] St = exp(-0.5×|GDDt-GDDopt| / σ)(7) In formula (7), GDDt is the actual effective accumulated temperature during the growth period (meteorological data), GDDopt is the most suitable accumulated temperature for crops (the accumulated temperature when planting high yield in normal years in history), and σ is the tolerance range (usually 10% of the length of the crop growth period).

[0068] In this optional implementation, the crop stress index is an indicator calculated by combining the relative value of the vegetation index, the water stress index, the nutrient stress index and the temperature stress index, and is used to comprehensively reflect the comprehensive impact of various stress factors on crops during their growth process.

[0069] The Crop Growth Index (CGI) reflects the growth and development stages of crops from seeding to germination, emergence, flowering, and maturity. The CGI is specifically represented by the vector shown in formula (8).

[0070] CGI = f(NDVI,SWC,Ntotal,Tavg,P)(8) In formula (8), SWC is soil water content, Ntotal is the total nitrogen content of soil and fertilizer, Tavg is average temperature, and P is rainfall. In actual calculations, the crop growth index CGI can be calculated using the function shown in formula (9).

[0071] CGI = F(a)×(k1×f(NDVI)+k2×f(SWC)+k3×f(Ntotal)+k4×f(Tavg)+k5×f(P)) (9) In formula (9), a is the phenological period adjustment coefficient [0-1], k1-k5 are dynamic weight factors, with default values ​​set to (0.15, 0.4, 0.1, 0.25, 0.1), and the default values ​​are recalculated using past data for calibration; F(a) represents the effective accumulated temperature function; SWC is soil water content, obtained using soil moisture instruments (or obtained based on soil moisture data inverted by remote sensing); Ntoal is obtained from fertilizer application and soil nutrients; P is rainfall; soil photosynthetic capacity value f(SWC) is an indicator calculated based on soil water content, which is used to reflect the soil's ability to support photosynthesis; fertility photosynthetic capacity value f(Ntotal) is an indicator calculated based on the total nitrogen content of the soil and fertilizer, which is used to reflect the soil fertility's ability to support photosynthesis; precipitation photosynthetic capacity value f(P) is calculated based on precipitation The index is used to reflect the support capacity of precipitation for photosynthesis; the temperature photosynthetic capacity value f(Tavg) is an index calculated based on the average temperature, which is used to reflect the support capacity of temperature for photosynthesis; the crop growth index is an index calculated from the relative value of the comprehensive vegetation index, the soil photosynthetic capacity value, the fertility photosynthetic capacity value, the precipitation photosynthetic capacity value and the temperature photosynthetic capacity value, which is used to comprehensively reflect the growth status and potential of crops. The following uses the effective accumulated temperature function shown in formula (10) and the photosynthetic capacity function shown in formula (11) to illustrate the representation of f() F(a) = 1 / (1+exp(-b(GDD-GDD50)))(10) In formula (10), GDD is the effective accumulated temperature at the base temperature Tbase (e.g., 8°C), GDD = ∑max(0, (Tmax+Tmin) / 2 - Tbase), b is the growth rate (0.005-0.01), and GDD50 is the accumulated temperature at 50% growth potential.

[0072] F(NDVI) = 1 / (1+exp(-b(NDVI-NDVIc)))(11) In formula (11), NDVIc is the dynamic threshold of crop phenological period (e.g., 0.15 at budding stage and 0.85 at flowering stage), and b is the slope parameter (5-10).

[0073] In this optional implementation, the remote sensing unit data in the spatiotemporal characteristic map is first used to calculate the remote sensing Normalized Difference Vegetation Index (NDVI) and the Maximum Normalized Difference Vegetation Index. By dividing the remote sensing Normalized Difference Vegetation Index by the Maximum Normalized Difference Vegetation Index, a relative vegetation index is obtained, which measures the current vegetation growth status relative to the optimal state. Next, the water stress index is calculated based on the available water and crop water requirement in the spatiotemporal characteristic map. The nutrient stress index is calculated based on the nitrogen utilization rate for crop growth and the theoretical nitrogen requirement for the crop. The temperature stress index is calculated based on the difference between the actual effective accumulated temperature during the growth period and the optimal accumulated temperature for the crop. Finally, the relative vegetation index, water stress index, nutrient stress index, and temperature stress index are combined to obtain the crop stress index, which is used to comprehensively assess the combined impact of various stress factors on the crop.

[0074] At the same time, the average temperature is calculated based on the temperature data in the spatiotemporal characteristic map; the soil photosynthetic capacity is calculated based on the soil moisture content; the fertility photosynthetic capacity is calculated based on the total nitrogen content of the soil and fertilizer; the precipitation photosynthetic capacity is calculated based on the precipitation; and the temperature photosynthetic capacity is calculated based on the average temperature. The relative value of the vegetation index, the soil photosynthetic capacity, the fertility photosynthetic capacity, the precipitation photosynthetic capacity, and the temperature photosynthetic capacity are comprehensively calculated to obtain the growth crop index, which comprehensively reflects the growth status and potential of crops.

[0075] In some optional implementations of the present disclosure, the above-mentioned determination of the target intelligent agent from the intelligent agent set based on the crop stress index and the crop growth index includes: detecting whether the crop stress index is within the non-stress range; in response to detecting that the crop stress index is within the non-stress range, determining the crop growth stage based on the crop growth index; and determining the target intelligent agent from the intelligent agent set based on the crop growth stage.

[0076] In this optional implementation, the system first checks to see if the crop stress index is within the non-stress range. If the test indicates the crop is not under stress, the system further determines the current crop growth stage based on the crop growth index. Based on the determined growth stage, the system automatically selects and triggers the corresponding target agent to implement precise agricultural management operations, such as activating irrigation or fertilization equipment at specific crop growth stages, thereby improving agricultural production efficiency and crop yields.

[0077] Figure 3 A process 300 according to an embodiment of the method for performing an agricultural task of the present disclosure is shown. The method for performing an agricultural task includes the following steps: Step 301: receiving a crop growth index and a crop stress index of a crop.

[0078] In this embodiment, the execution subject on which the agricultural task execution method runs can be any agent in the agent set, and the execution subject can receive the crop growth index and the crop stress index from the central agent.

[0079] In this embodiment, the central intelligent agent can monitor the multi-source agricultural data of crops in real time through the sensor network installed in the farmland, and calculate the crop growth index and stress index by analyzing and processing the multi-source agricultural data. The specific calculation process can be referred to Figure 1 The illustrated embodiment describes an agent collaboration method.

[0080] Step 302: Perform agricultural tasks on crops based on the crop growth index and the crop stress index.

[0081] In this embodiment, thresholds are set based on the numerical ranges of these indices. When the growth index falls below the normal range, fertilization or irrigation may be necessary; when the stress index rises above the warning value, pest control or drought relief measures may be necessary. Based on the real-time data of these indices, the execution entity automatically triggers corresponding agricultural tasks, such as activating irrigation equipment, releasing pesticides, or notifying farmers to take manual intervention measures, thereby achieving precision agricultural management and improving crop yield and quality.

[0082] Step 303, obtaining environmental data while performing the agricultural task; In this embodiment, when performing agricultural tasks (such as irrigation), environmental data is collected through a sensor network installed in the farmland. These sensors can monitor key indicators such as soil moisture, air temperature, and humidity in real time. For example, a soil moisture sensor can accurately measure the moisture content in the soil. When the soil moisture falls below a preset threshold, the execution entity automatically triggers irrigation to ensure that crops receive the appropriate amount of water. Simultaneously, combined with weather data provided by a meteorological station (such as rainfall forecasts for the next few days), the execution entity can further optimize irrigation plans to avoid over-irrigation. This approach not only improves water resource utilization efficiency but also reduces the frequency of manual intervention, achieving intelligent and precise management of agricultural production.

[0083] Step 304: Based on the environmental data, related agents in the agent set are linked.

[0084] In this embodiment, an agent set is controlled by an execution subject on which the agent collaboration method is running. The agent set includes multiple agents, each of which has a corresponding function. When the execution subject on which the agricultural task execution method is running is executing a corresponding agricultural task, if the environmental data is found to be related to a related agent in the agent set, the related agent is linked. The related agent is one or more agents in the agent set, and the related agent is related to the environmental data.

[0085] In this embodiment, sensors and other devices collect farmland environmental data (such as soil moisture, temperature, light intensity, and carbon dioxide concentration). This data is then transmitted to an execution entity via IoT technology. The execution entity analyzes the data based on pre-set algorithms and models to determine the growth status and needs of the crops. Subsequently, the execution entity automatically sends control signals to relevant agents in the agent collection (such as irrigation agents, fertilization agents, and greenhouse agents) to precisely operate these agents. For example, when soil moisture falls below a set threshold, the irrigation agent is automatically activated; when light levels are insufficient, the greenhouse lighting agent is adjusted. This intelligent linkage model can improve agricultural production efficiency, reduce labor costs, optimize the crop growth environment, and increase yield and quality.

[0086] In this embodiment, the intelligent agent set may include: a plant protection intelligent agent, an irrigation intelligent agent, a nutrition intelligent agent, and a yield intelligent agent. When the execution subject on which the agricultural task execution method runs is a plant protection intelligent agent, the irrigation intelligent agent may be linked. Irrigation will affect the humidity in the field and increase the risk of disease. Irrigation should be suspended within 24 hours before applying the pesticide to avoid dilution of the liquid medicine and ensure that the leaves are dry. Micro-irrigation can be started after applying the pesticide to promote the penetration of the liquid medicine into the root system and enhance the effect of systemic pesticides. The nutrition intelligent agent is linked to provide nutrient balance. Good nutrition can thicken the cell walls of crops, thereby reducing the probability of pest infection. The plant protection intelligent agent dynamically adjusts the drug ratio accordingly. The yield intelligent agent is linked to predict the yield level in advance based on growth, agricultural data, and meteorological data. The plant protection intelligent agent conducts inspections and prevention in advance based on the predicted yield level to avoid risks.

[0087] When the agricultural task execution method runs on an irrigation agent, it can be linked to the plant protection agent. For example, when spraying herbicides, delaying irrigation for at least 12 hours is crucial to prevent the drug from losing its effectiveness. The nutrition agent can also be linked to the water requirement for nitrogen fertilizer dissolution, which is directly related to irrigation depth. For example, urea requires a soil moisture content greater than 18% to dissolve in the 30cm soil layer; otherwise, ammonia volatilization can occur. The irrigation agent should adjust drip irrigation duration in sync with the nutrition agent's fertilization plan. The yield agent can also be linked to identify water-sensitive periods (such as the tasseling and silking period of corn) based on crop growth models. Based on this, the irrigation module increases water allocation priority by 50% to ensure grain formation.

[0088] When the agricultural task execution method is executed by a nutrition agent, the plant protection agent is linked to the nutrition agent. After applying certain fungicides, the nutrition agent needs to suspend magnesium fertilizer application for three days to prevent the agent from becoming ineffective due to photolysis. If the plant protection agent detects an aphid outbreak, the nutrition agent can increase the proportion of silicon fertilizer to enhance the plant's mechanical resistance. The irrigation agent is linked to the irrigation agent. When the drip irrigation tape is buried at a depth of 5 cm, the nitrogen migration distance at the wetting front reaches 40 cm, while at a depth of 10 cm, it is only 25 cm. The nutrition module should optimize fertilizer application based on this information to avoid deep seepage. The yield agent is linked to the nutrition agent and, based on the yield forecast, instructs the nutrition agent to increase fertilizer application to compensate for the yield.

[0089] When the yield agent executes the agricultural task execution method, it collaborates with the plant protection agent, providing real-time data (e.g., a 15% decrease in leaf area index due to rust) that is directly fed into the yield model to correct prediction errors. Seven days after the plant protection agent applies pesticides, if plant protection stress decreases or disappears, the yield model will return to normal prediction levels. In conjunction with the irrigation agent, soil water potential data (e.g., <-60 kPa for five consecutive days) triggers the yield model to activate the drought reduction coefficient, affecting yield. If actual irrigation volume falls below the planned value, the model automatically adjusts yield projections downward. In conjunction with the nutrition agent, low nitrogen content in crops is marked as a "potential yield reduction zone" by the nutrition module, and the yield model accordingly increases its prediction error. After topdressing, the yield model's response is delayed by seven days.

[0090] The agricultural task execution method provided by the embodiments of the present disclosure first receives the crop growth index and crop stress index of the crop; secondly, based on the crop growth index and crop stress index, the agricultural task is executed on the crop; then, environmental data is obtained when executing the agricultural task; finally, based on the environmental data, the relevant intelligent agents in the intelligent agent set are linked to execute the agricultural task on the crop through the crop growth index and the crop stress index, and based on the environmental data, the relevant intelligent agents are linked to improve the reliability of collaborative operations between multiple intelligent agents.

[0091] In some optional implementations of the present disclosure, the above-mentioned agricultural task is a fertilization task. Based on the crop growth index and the crop stress index, the agricultural task performed on the crops includes: determining the fertilization index based on the crop growth index and the crop stress index; determining the amount of topdressing based on the fertilization index; calculating the standard value of nitrogen content based on historical crop leaf nitrogen content; and determining fertilization recommendation information in the fertilization task based on the standard value of nitrogen content and the amount of topdressing.

[0092] In this optional implementation, the fertilization index is an indicator used to guide fertilization, which is calculated based on the crop growth index and the crop stress index, and reflects the urgency of the crop's current demand for fertilizer; the amount of topdressing is the amount of additional fertilizer that needs to be applied based on the fertilization index, which is used to supplement the nutrients required during crop growth; the standard nitrogen content is a reference value of the nitrogen content of crop leaves calculated based on historical data, which is used to assess whether the current nutritional status of the crop has reached the ideal level; the fertilization recommendation information is specific suggestions on fertilization generated after comprehensively considering the fertilization index, topdressing amount and nitrogen content standard value, including fertilization time, fertilization amount, fertilization method, etc.

[0093] In this optional implementation, a fertilization index is calculated based on the crop growth index and crop stress index. This index reflects the crop's current fertilizer needs. Next, a standard value for the nitrogen content of crop leaves is calculated based on historical data, which serves as a reference for assessing the crop's nutritional status. Finally, the amount of topdressing fertilizer to be applied is determined based on the fertilization index, and specific fertilization recommendations are generated based on the standard nitrogen content. This provides agricultural producers with precise fertilization guidance, thereby optimizing fertilization results and improving crop yield and quality.

[0094] Further references Figure 4 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of an intelligent collaborative device, which is similar to Figure 1 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0095] like Figure 4 As shown, the agent collaboration device 400 provided in this embodiment includes: a collection unit 401, a map construction unit 402, an index construction unit 403, a determination unit 404, and a sending unit 405. The collection unit 401 can be configured to collect multi-source agricultural data on crops from historical periods and current growth stages. The map construction unit 402 can be configured to construct a spatiotemporal feature map based on the multi-source agricultural data. The index construction unit 403 can be configured to calculate a crop stress index and a crop growth index based on the spatiotemporal feature map. The determination unit 404 can be configured to determine a target agent from a set of agents based on the crop stress index and the crop growth index. The sending unit 405 can be configured to send the crop growth index and the crop stress index to the target agent, so that the target agent performs corresponding agricultural tasks on the crops based on the crop growth index and the crop stress index, and links related agents in the set of agents based on environmental data.

[0096] In this embodiment, the specific processing of the acquisition unit 401, the graph construction unit 402, the index construction unit 403, the determination unit 404, and the sending unit 405 and the technical effects thereof can be referred to respectively. Figure 1 The relevant descriptions of step 101, step 102, step 103, step 104 and step 105 in the corresponding embodiment are not repeated here.

[0097] In some embodiments of the present disclosure, the multi-source agricultural data include: remote sensing data, Internet of Things data, field management data, meteorological data, and pest and disease monitoring data of historical periods and current growth stages, and the map construction unit 402 is configured to: perform data preprocessing on remote sensing data, Internet of Things data, field management data, meteorological data, and pest and disease monitoring data; use a spatiotemporal attention mechanism to align the frequency and spatial resolution of the preprocessed remote sensing data, Internet of Things data, field management data, meteorological data, and pest and disease monitoring data to obtain aligned data; and construct a spatiotemporal feature map based on the aligned data.

[0098] In some embodiments of the present disclosure, the index construction unit 403 is further configured to: obtain the remote sensing normalized vegetation index and the maximum normalized vegetation index based on the remote sensing unit in the spatiotemporal characteristic map; calculate the relative value of the vegetation index based on the remote sensing normalized vegetation index and the maximum normalized vegetation index; calculate the water stress index based on the available water and crop water requirement in the spatiotemporal characteristic map; calculate the nutrient stress index based on the utilization rate of crop growth nitrogen and the theoretical nitrogen requirement of crops in the spatiotemporal characteristic map; calculate the temperature stress index based on the actual effective accumulated temperature during the growth period and the most suitable accumulated temperature for crops in the spatiotemporal characteristic map; calculate the relative value of the vegetation index based on the available water and crop water requirement ... The crop stress index is calculated based on the relative value of vegetation index, water stress index, nutrient stress index and temperature stress index; the average temperature is calculated based on the temperature in the spatiotemporal characteristic map; the soil photosynthetic capacity value is calculated based on the soil moisture content in the spatiotemporal characteristic map; the fertility photosynthetic capacity value is calculated based on the total nitrogen content of soil and fertilizer in the spatiotemporal characteristic map; the precipitation photosynthetic capacity value is calculated based on the precipitation in the spatiotemporal characteristic map; the temperature photosynthetic capacity value is calculated based on the average temperature; the crop growth index is calculated based on the relative value of vegetation index, soil photosynthetic capacity value, fertility photosynthetic capacity value, precipitation photosynthetic capacity value and temperature photosynthetic capacity value.

[0099] In some embodiments of the present disclosure, the above-mentioned determination unit 404 is also configured to: detect whether the crop stress index is within the non-stress range; in response to detecting that the crop stress index is within the non-stress range, determine the crop growth stage based on the crop growth index; based on the crop growth stage, determine the target intelligent agent from the intelligent agent set.

[0100] The embodiment of the present disclosure provides an intelligent agent collaboration device. First, the collection unit 401 collects multi-source agricultural data of crops in historical periods and current growth stages; the map construction unit 402 constructs a spatiotemporal feature map based on the multi-source agricultural data; the index construction unit 403 calculates the crop stress index and the crop growth index based on the spatiotemporal feature map; the determination unit 404 determines the target intelligent agent from the intelligent agent set based on the crop stress index and the crop growth index; the sending unit 405 sends the crop growth index and the crop stress index to the target intelligent agent, so that the target intelligent agent performs corresponding agricultural tasks on the crops based on the crop growth index and the crop stress index, and links the relevant intelligent agents in the intelligent agent set based on the environmental data.

[0101] Further references Figure 5 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of an agricultural task execution device, which is similar to Figure 3 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0102] like Figure 5 As shown, the agricultural task execution device 500 provided in this embodiment includes: a receiving unit 501, an executing unit 502, an acquiring unit 503, and a linkage unit 504. The receiving unit 501 can be configured to receive a crop growth index and a crop stress index of a crop. The executing unit 502 can be configured to execute an agricultural task on the crop based on the crop growth index and the crop stress index. The acquiring unit 503 can be configured to acquire environmental data when executing the agricultural task. The linkage unit 504 can be configured to link related agents in the agent set based on the environmental data.

[0103] In this embodiment, the specific processing of the receiving unit 501, the executing unit 502, the acquiring unit 503, and the linkage unit 504 and the technical effects thereof can be referred to in the respective embodiments. Figure 3 The relevant descriptions of step 301, step 302, step 303, and step 304 in the corresponding embodiment are not repeated here.

[0104] In some embodiments of the present disclosure, the above-mentioned execution unit 502 is also configured to: determine the fertilization index based on the crop growth index and the crop stress index; determine the amount of topdressing based on the fertilization index; calculate the standard value of nitrogen content based on historical crop leaf nitrogen content; and determine the fertilization recommendation information in the fertilization task based on the standard value of nitrogen content and the amount of topdressing.

[0105] The agricultural task execution device provided by the embodiments of the present disclosure first comprises a receiving unit 501 receiving a crop growth index and a crop stress index; an executing unit 502 executing an agricultural task on the crop based on the crop growth index and the crop stress index; an acquiring unit 503 acquiring environmental data while executing the agricultural task; and a linkage unit 504 linking related agents in the agent set based on the environmental data. This collaborative working model ensures a more efficient, precise, and adaptable agricultural production process, improving crop yield and quality while reducing resource waste and environmental impact, ultimately achieving intelligent and sustainable agricultural production.

[0106] Further references Figure 6 The present disclosure also provides an intelligent agent collaborative system, which includes: a central intelligent agent 601 and an intelligent agent set 602.

[0107] The central intelligent agent 601 is used to collect multi-source agricultural data of crop a in the historical period and the current growth stage; construct a spatiotemporal feature map based on the multi-source agricultural data; calculate the crop stress index and the crop growth index based on the spatiotemporal feature map; determine the target intelligent agent 6021 from the intelligent agent set 602 based on the crop stress index and the crop growth index; and send the crop growth index and the crop stress index to the target intelligent agent 6021.

[0108] The target agent 6021 performs corresponding agricultural tasks on crop a based on the crop growth index and the crop stress index, and links the relevant agents 6022 in the agent set 602 based on the environmental data.

[0109] The intelligent agent collaborative system provided by the present disclosure improves the globality and coordination of agricultural decision-making through the cooperation between the central intelligent agent and the intelligent agents in the intelligent agent set: through the task scheduling mechanism of the multi-agent system, it realizes the linkage optimization of plant protection, irrigation, nutrition, yield prediction and other links, avoiding the local optimal trap caused by a single model.

[0110] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0111] Figure 7A schematic block diagram of an example electronic device 700 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their modes are provided for example only and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0112] like Figure 7 As shown, electronic device 700 includes a computing unit 701, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 702 or a computer program loaded from a storage unit 708 into a random access memory (RAM) 703. Various programs and data required for the operation of electronic device 700 may also be stored in RAM 703. Computing unit 701, ROM 702, and RAM 703 are interconnected via a bus 704. An input / output (I / O) interface 705 is also connected to bus 704.

[0113] Multiple components in the electronic device 700 are connected to the I / O interface 705, including an input unit 706, such as a keyboard, a mouse, etc.; an output unit 707, such as various types of displays, speakers, etc.; a storage unit 708, such as a magnetic disk, an optical disk, etc.; and a communication unit 709, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 709 allows the electronic device 700 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0114] The computing unit 701 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 701 performs the various methods and processes described above, such as the agent collaboration method. For example, in some embodiments, the agent collaboration method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 708. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 700 via the ROM 702 and / or the communication unit 709. When the computer program is loaded into the RAM 703 and executed by the computing unit 701, one or more steps of the agent collaboration method described above can be performed. Alternatively, in other embodiments, the computing unit 701 can be configured to perform the agent collaboration method in any other suitable manner (e.g., via firmware).

[0115] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0116] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable intelligent collaborative device, so that when the program code is executed by the processor or controller, the modes / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0117] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fibers, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0118] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0119] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0120] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not limited herein.

[0121] The foregoing descriptions of specific exemplary embodiments of the present disclosure are for purposes of illustration and description. These descriptions are not intended to limit the present disclosure to the precise forms disclosed, and it is apparent that many variations and modifications are possible in light of the foregoing teachings. The exemplary embodiments have been selected and described for the purpose of explaining the specific principles of the present disclosure and their practical application, thereby enabling those skilled in the art to realize and utilize a variety of exemplary embodiments of the present disclosure and various options and modifications. The scope of the present disclosure is intended to be defined by the claims and their equivalents.

Claims

1. An agent collaboration method, characterized in that: The method comprises: Collect multi-source agricultural data on crops during historical periods and current growth stages; Based on the multi-source agricultural data, construct a spatiotemporal feature map; Calculating a crop stress index and a crop growth index based on the spatiotemporal characteristic map; Determining a target intelligent agent from an intelligent agent set based on the crop stress index and the crop growth index; The crop growth index and the crop stress index are sent to the target intelligent agent, so that the target intelligent agent performs corresponding agricultural tasks on the crops based on the crop growth index and the crop stress index, and links related intelligent agents in the intelligent agent set based on environmental data.

2. The method according to claim 1, characterized in that The multi-source agricultural data includes: remote sensing data, Internet of Things data, field management data, meteorological data, and pest and disease monitoring data of the historical period and the current growth stage. The construction of a spatiotemporal feature map based on the multi-source agricultural data includes: Performing data preprocessing on the remote sensing data, the Internet of Things data, the field management data, the meteorological data, and the pest and disease monitoring data; Using the spatiotemporal attention mechanism, the frequency and spatial resolution of pre-processed remote sensing data, IoT data, field management data, meteorological data, and pest and disease monitoring data are aligned to obtain aligned data. Based on the aligned data, a spatiotemporal feature map is constructed.

3. The method according to claim 1, characterized in that The calculating of the crop stress index and the crop growth index based on the spatiotemporal characteristic map includes: Based on the remote sensing units in the spatiotemporal characteristic map, a remote sensing normalized difference vegetation index and a maximum normalized difference vegetation index are obtained; Calculating a relative value of a vegetation index based on the remote sensing normalized difference vegetation index and the maximum normalized difference vegetation index; Calculating a water stress index based on available water and crop water requirement in the spatiotemporal characteristic map; Calculating a nutrient stress index based on the crop growth nitrogen utilization rate and the crop theoretical nitrogen requirement in the spatiotemporal characteristic map; Calculating a temperature stress index based on the actual effective accumulated temperature and the most suitable accumulated temperature for the crop during the growth period in the spatiotemporal characteristic map; Calculating a crop stress index based on the relative value of the vegetation index, the water stress index, the nutrient stress index, and the temperature stress index; Calculating an average temperature based on the temperatures in the spatiotemporal characteristic map; Calculating the soil photosynthetic capacity value based on the soil moisture content in the spatiotemporal characteristic map; Calculating the fertility photosynthetic capacity value based on the soil and total nitrogen content of fertilizer in the spatiotemporal characteristic map; Calculating the precipitation photosynthetic capacity value based on the precipitation in the spatiotemporal characteristic map; Calculating a temperature photosynthetic capacity value based on the average temperature; A crop growth index is calculated based on the relative value of the vegetation index, the soil photosynthetic capacity value, the fertility photosynthetic capacity value, the precipitation photosynthetic capacity value, and the temperature photosynthetic capacity value.

4. The method according to claim 1, characterized in that The step of determining a target intelligent agent from an intelligent agent set based on the crop stress index and the crop growth index includes: detecting whether the crop stress index is within a non-stress range; In response to detecting that the crop stress index is in a non-stress range, determining a crop growth stage based on the crop growth index; Based on the crop growth stage, a target agent is determined from the agent set.

5. A method for executing agricultural tasks, applied to a target intelligent agent, wherein the target intelligent agent is determined by the intelligent agent collaboration method according to any one of claims 1 to 4, characterized in that: The method comprises: Crop growth index and crop stress index of recipient crops; performing an agricultural task on the crop based on the crop growth index and the crop stress index; While performing the agricultural task, acquiring environmental data; Based on the environmental data, related agents in the agent set are linked.

6. The method according to claim 5, wherein the agricultural task is a fertilization task, and performing the agricultural task on the crop based on the crop growth index and the crop stress index comprises: determining a fertilization index based on the crop growth index and the crop stress index; Determining the amount of topdressing based on the fertilization index; Calculate the standard value of nitrogen content based on historical crop leaf nitrogen content; Fertilization recommendation information in the fertilization task is determined based on the nitrogen content standard value and the topdressing amount.

7. An intelligent agent collaboration device, comprising: a collection unit configured to collect multi-source agricultural data of crops in historical periods and current growth stages; A map construction unit is configured to construct a spatiotemporal feature map based on the multi-source agricultural data; an index construction unit configured to calculate a crop stress index and a crop growth index based on the spatiotemporal characteristic map; a determining unit configured to determine a target intelligent agent from an intelligent agent set based on the crop stress index and the crop growth index; The sending unit is configured to send the crop growth index and the crop stress index to the target intelligent agent, so that the target intelligent agent performs corresponding agricultural tasks on the crops based on the crop growth index and the crop stress index, and links related intelligent agents in the intelligent agent set based on environmental data.

8. An agricultural task execution device, applied to a target intelligent agent, wherein the target intelligent agent is determined by the intelligent agent collaboration device according to claim 7, the device comprising: a receiving unit configured to receive a crop growth index and a crop stress index of a crop; an execution unit configured to execute an agricultural task on the crop based on the crop growth index and the crop stress index; an acquisition unit configured to acquire environmental data when performing the agricultural task; The linkage unit is configured to link related agents in the agent set based on the environmental data.

9. An intelligent agent collaboration system, comprising: Central agent, agent collection; A central agent that collects multi-source agricultural data on crops during historical periods and current growth stages; Based on the multi-source agricultural data, a spatiotemporal feature map is constructed; based on the spatiotemporal feature map, a crop stress index and a crop growth index are calculated; based on the crop stress index and the crop growth index, a target agent is determined from a set of agents; and the crop growth index and the crop stress index are sent to the target agent; The target intelligent agent performs corresponding agricultural tasks on the crops based on the crop growth index and the crop stress index, and links related intelligent agents in the intelligent agent set based on environmental data.

10. An electronic device, characterized in that: include: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 6.

11. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to enable the computer to execute the method according to any one of claims 1 to 6.

Images