Artificial intelligence-based simulated human brain robot control method and device

Abstract

The invention relates to the technical field of intelligent control, in particular to a simulated human brain robot control method and device based on artificial intelligence, and the method comprises the steps: obtaining cultivated land crop growth information and soil monitoring information of a plurality of recorders in cultivated land, forming a crop growth dynamic log based on the health state of crops, and inputting the crop growth dynamic log into a pre-trained growth analysis model to decide whether disease and pest growth exists, and triggering a corresponding maintenance prompt mode according to the risk level of disease and pest growth in a disease and pest record set, and performing time dimension statistical analysis on the farmland environment real-time information and the crop health state to output farmland planting historical information, and dynamically updating a maintenance plan in a preset environment maintenance plan library according to maintenance plan adjustment content. Fusion analysis of multi-source data is fully utilized, comprehensive environment perception and intelligent decision making are difficult to achieve, and the method has great significance in achieving efficient planting history and safe maintenance of intelligent cultivated land.

Description

Technical Field

[0001] The present application relates to the field of intelligent control technology, and in particular to a control method and device for a human brain-simulating robot based on artificial intelligence. Background Art

[0002] Traditional farmland maintenance relies primarily on manual oversight, which not only results in high labor costs but also low efficiency. With the advancement of technology, existing automated maintenance systems often employ fixed maintenance plans and lack adaptability to diverse environments. This is particularly true when dealing with fluctuations in survival rates and unexpected record processing, making it difficult for existing systems to respond promptly and effectively.

[0003] At the same time, existing farmland maintenance systems have significant shortcomings in identifying and addressing pest and disease growth. Traditional growth monitoring systems primarily focus on passive recording, lacking active identification and notification methods, making them ineffective in preventing and promptly addressing pest and disease records. These systems also fail to fully utilize the fusion analysis of multi-source data, making it difficult to achieve comprehensive environmental perception and intelligent decision-making.

[0004] Furthermore, the existing system is weak in analyzing historical planting information and adjusting plans. The system lacks an adaptive plan adjustment method, making it impossible to dynamically adjust and adapt maintenance plans based on historical planting information, hindering the continued improvement of maintenance effectiveness. Addressing these issues is crucial for achieving efficient and safe maintenance of smart farmland. Summary of the Invention

[0005] To achieve the above objectives, this application provides the following technical solutions: According to a first aspect of the present invention, the present invention seeks protection for a method for controlling a human brain-simulating robot based on artificial intelligence, the method comprising: Acquire crop growth information and soil quality monitoring information from multiple recorders in the cultivated land, perform target analysis on the crop growth information to collect crop health status and crop organ development status, perform feature collection on the soil quality monitoring information to output cultivated land soil indicators, input the crop health status, the crop organ development status and the cultivated land soil indicators into a pre-trained environment detection network, output real-time cultivated land environment information, perform semantic analysis on the real-time cultivated land environment information to output main feature indicators of the real-time environment, form an environmental feature sequence based on the main feature indicators, calculate the matching degree between the environmental feature sequence and the environmental description sequence in a preset environmental maintenance plan library to obtain a maintenance plan with the highest matching degree, match the interval data, survival rate information and soil quality data in the real-time environmental information with a preset interval condition library, survival rate threshold condition library and soil indicator condition library in turn to output associated interval maintenance sub-plan, survival rate maintenance sub-plan and soil maintenance sub-plan, and integrate the sub-plans to form an environmental maintenance plan; A crop growth dynamic log is generated based on the crop health status, and the crop growth dynamic log is input into a pre-trained growth analysis model to determine whether there is growth of pests and diseases. When pests and diseases are found, the growth type, occurrence time, and location data of the pests and diseases are stored in a pest and disease record set. A corresponding maintenance reminder method is triggered according to the risk level of pest and disease growth in the pest and disease record set. At the same time, a related accident treatment plan is adopted from a preset accident treatment plan library based on the real-time information of the cultivated land environment and the pest and disease growth type; A statistical analysis of the real-time information of the cultivated land environment and the crop health status in the time dimension is performed to output cultivated land planting history information, the cultivated land planting history information is input into an adaptive adjustment model to obtain cultivated land maintenance plan adjustment content, the maintenance plan in the preset environmental maintenance plan library is dynamically updated according to the maintenance plan adjustment content, and at the same time, the cultivated land planting history information, the pest and disease record set and the maintenance plan adjustment content are transmitted to a cloud maintenance platform for storage, and the cloud maintenance platform obtains a cultivated land planting history analysis report based on the acquired data.

[0006] Furthermore, the crop health status, the crop organ development status, and the farmland soil index are input into a pre-trained environment detection network to output real-time farmland environment information, and semantic analysis is performed on the real-time farmland environment information to output the main characteristic indicators of the real-time environment, including: The crop health status, the crop organ development status and the cultivated land soil index are subjected to feature fusion according to a preset template to form a feature matrix, the feature matrix is ​​preprocessed, the predicted feature data is input into an environmental monitoring network, and real-time information of the cultivated land environment is output through the environmental monitoring network; The text record information in the real-time information of the cultivated land environment is semantically split and word-segmented, the cultivated land rotation characteristics and soil quality characteristics in the environmental tags are collected, and the main characteristic indicators of the real-time environment are output by weighted calculation of each feature based on the preset feature item weight.

[0007] Furthermore, forming an environmental feature sequence according to the main feature indicators, and calculating the matching degree between the environmental feature sequence and the environmental description sequence in the preset environmental maintenance plan library to obtain the maintenance plan with the highest matching degree includes: The Jaccard matching calculation method is used to calculate the matching scores between the pre-environmental feature sequence and each environmental description sequence in the preset environmental maintenance plan library. The matching scores are screened based on a preset matching threshold, and the maintenance plan associated with the environmental description sequence with the highest matching score that is greater than the matching threshold is obtained as the matching plan for the real-time environment.

[0008] Furthermore, the interval data, survival rate information and soil quality data in the real-time environmental information are matched with the preset interval condition library, survival rate threshold condition library and soil index condition library in sequence to output the associated interval maintenance sub-plan, survival rate maintenance sub-plan and soil maintenance sub-plan, and the sub-plans are integrated to form an environmental maintenance plan, including: Parsing the real-time environmental information into interval data, survival rate information, and soil quality data, dividing the interval data according to a preset time granularity and comparing it with the rule items in the interval condition library to output an interval maintenance sub-plan, performing density calculation on the survival rate information and comparing it with the rule items in the survival rate threshold condition library to output a survival rate maintenance sub-plan, and performing indicator classification on the soil quality data and comparing it with the rule items in the soil indicator condition library to output a soil maintenance sub-plan; Based on the preset plan priorities, the interval maintenance sub-plan, the survival rate maintenance sub-plan and the soil maintenance sub-plan are subjected to risk analysis and mitigation processing, and the risk-free sub-plans are integrated and adjusted through a plan fusion algorithm to obtain an environmental maintenance plan. According to the environmental maintenance plan, fertilization control requests, growth monitoring requests and soil adjustment requests are obtained.

[0009] Furthermore, the step of generating a crop growth dynamic log based on the crop health status and inputting the crop growth dynamic log into a pre-trained growth analysis model to determine whether pests and diseases are growing includes: sorting the crop health status according to timestamps and segmenting it into a health status log, performing time window division and feature completion processing on the health status log to output a fixed-length growth dynamic log, and sequentially preprocessing static features, fertilization history features, and pest and disease elimination features in the growth dynamic log; The predicted growth dynamic log is input into the pest and disease growth detection network formed based on the recurrent neural network. The dynamic mapping features and individual features in the growth log are collected through the pest and disease growth detection network. The collected features are analyzed and classified for pests and diseases based on the preset growth mode feature library, and the pest and disease growth decision results are output.

[0010] Furthermore, when the growth of pests and diseases is detected, the growth type, occurrence time, and location data of the pests and diseases are stored in a pest and disease record set, and a corresponding maintenance prompt mode is triggered according to the risk level of the pest and disease growth in the pest and disease record set. At the same time, according to the real-time information of the cultivated land environment and the growth type of the pests and diseases, an associated accident handling plan is adopted from a preset accident handling plan library, including: Determine the type of pest growth obtained through the analysis through a preset growth action object, obtain spatial coordinate information of the occurrence of the pest growth, and write the pest record consisting of the pest growth type, timestamp, and spatial coordinate information into a pest record set. Score the pest growth in the pest record set according to a preset risk level assessment rule, and determine the prompt level based on the risk level score. According to the prompt level, the preset maintenance prompt condition library is queried to obtain a prompt request, the real-time information of the cultivated land environment and the growth type information of the pests and diseases are input into an accident treatment plan selection model formed based on a decision tree, and the associated treatment plan is obtained from the preset accident treatment plan library through the accident treatment plan selection model to obtain an accident treatment request log.

[0011] Furthermore, the statistical analysis of the real-time farmland environment information and the crop health status in the time dimension is performed to output farmland planting history information, and the farmland planting history information is input into the adaptive adjustment model to obtain the farmland maintenance plan adjustment content, including: The real-time information of the farmland environment and the health status of the crops are aggregated in segments according to a preset time window, and multi-dimensional feature collection is performed on the aggregated data to output the survival rate change trend, crop maturity cycle distribution, crop organ disease and pest infection frequency, and soil index change curve. Based on a preset data analysis model, correlation analysis and trend prediction are performed on each indicator to obtain the historical information of farmland planting; The cultivated land planting history information is input into a plan adjustment model formed based on a reinforcement learning algorithm. The execution effect of the existing maintenance plan is evaluated and scored through the plan adjustment model. The reward value of the plan execution is calculated based on the preset plan evaluation index. The plan gradient method is used to adjust the indicators of the maintenance plan to obtain the plan adjustment content.

[0012] According to a second aspect of the present invention, the present invention seeks protection for an artificial intelligence-based human brain-simulating robot control device, comprising: an environment determination module for acquiring cultivated land crop growth information and soil quality monitoring information from multiple recorders in cultivated land, performing target analysis on the cultivated land crop growth information to collect crop health status and crop organ development status, performing feature collection on the soil quality monitoring information to output cultivated land soil indicators, inputting the crop health status, the crop organ development status, and the cultivated land soil indicators into a pre-trained environment detection network to output real-time cultivated land environment information, performing semantic analysis on the real-time cultivated land environment information to output main characteristic indicators of the real-time environment, forming an environmental feature sequence based on the main characteristic indicators, performing matching calculation on the environmental feature sequence with an environmental description sequence in a preset environmental maintenance plan library to obtain a maintenance plan with the highest matching degree, matching the interval data, survival rate information, and soil quality data in the real-time environmental information with a preset interval condition library, a survival rate threshold condition library, and a soil indicator condition library in sequence to output associated interval maintenance sub-plans, survival rate maintenance sub-plans, and soil maintenance sub-plans, and integrating the sub-plans to form an environmental maintenance plan; An accident handling module is configured to generate a crop growth dynamic log based on the crop health status, input the crop growth dynamic log into a pre-trained growth analysis model to determine whether there is growth of pests and diseases, and when pests and diseases are detected, store the growth type, occurrence time, and location data of the pests and diseases in a pest and disease record set; trigger a corresponding maintenance prompt method based on the risk level of pest and disease growth in the pest and disease record set; and simultaneously adopt a related accident handling plan from a preset accident handling plan library based on the real-time information of the cultivated land environment and the pest and disease growth type; a plan update module for performing a statistical analysis of the real-time farmland environment information and the crop health status in a time dimension to output farmland planting history information, inputting the farmland planting history information into an adaptive adjustment model to obtain farmland maintenance plan adjustment content, dynamically updating the maintenance plan in the preset environmental maintenance plan library based on the maintenance plan adjustment content, and simultaneously transmitting the farmland planting history information, the pest and disease record set, and the maintenance plan adjustment content to a cloud maintenance platform for storage, wherein the cloud maintenance platform obtains a farmland planting history analysis report based on the acquired data; The artificial intelligence-based human brain simulation robot control device is used to execute the artificial intelligence-based human brain simulation robot control method.

[0013] The present application relates to the field of intelligent control technology, and more particularly to an artificial intelligence-based human brain-simulating robot control method and device. The method obtains crop growth information and soil quality monitoring information from multiple recorders in cultivated land, generates a dynamic crop growth log based on the crop health status, inputs the dynamic crop growth log into a pre-trained growth analysis model to determine whether pests and diseases are growing, triggers corresponding maintenance prompts based on the risk level of pest growth in a pest and disease record set, performs a time-dimensional statistical analysis of real-time cultivated land environmental information and crop health status, outputs cultivated land planting history information, and dynamically updates the maintenance plan in a preset environmental maintenance plan library based on the maintenance plan adjustment content. The present invention fully utilizes the fusion analysis of multi-source data, which is difficult to achieve comprehensive environmental perception and intelligent decision-making, and is of great significance for realizing efficient planting history and safe maintenance of smart cultivated land. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A flowchart of a control method for a human brain-simulating robot based on artificial intelligence claimed in an embodiment of the present application; Figure 2 A second workflow diagram of a method for controlling a human brain-simulating robot based on artificial intelligence as claimed in an embodiment of the present application; Figure 3 A third workflow diagram of a method for controlling a human brain-simulating robot based on artificial intelligence as claimed in an embodiment of the present application; Figure 4 This is a fourth workflow diagram of a method for controlling a human brain-simulating robot based on artificial intelligence claimed in an embodiment of the present application; Figure 5 This is a fifth workflow diagram of a method for controlling a human brain-simulating robot based on artificial intelligence as claimed in an embodiment of the present application; Figure 6 This is a sixth workflow diagram of a method for controlling a human brain-simulating robot based on artificial intelligence claimed in an embodiment of the present application; Figure 7 This is a structural module diagram of a human brain simulation robot control device based on artificial intelligence that is protected by an embodiment of the present application. DETAILED DESCRIPTION

[0015] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments of soil fertility obtained by ordinary technology in this field without creative work are within the scope of protection of this application.

[0016] The terms "first," "second," and "third" in this application are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features identified. Therefore, features identified as "first," "second," or "third" may explicitly or implicitly include at least one of such features. In the description of this application, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined. All directional designations in the embodiments of this application (such as up, down, left, right, front, back, etc.) are intended only to explain the relative positional relationships and movement of components in a specific static state (as shown in the accompanying drawings). If the specific static state changes, the directional designations will also change accordingly. Furthermore, the terms "including," "having," and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or farming robot comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or farming robot.

[0017] Reference herein to an "embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the present application. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, that the embodiments described herein may be combined with other embodiments.

[0018] Taking into account the problems existing in the prior art, the present application provides a method and device for controlling a human brain-simulating robot based on artificial intelligence, which obtains farmland data through multiple recorders and sensors, and collects crop health status, crop organ development status and soil indicators. An innovative environmental detection network and growth analysis model are formed to realize intelligent identification of environmental status and real-time monitoring of the growth of pests and diseases. Maintenance plans are dynamically matched based on environmental feature sequences, and detailed maintenance sub-plans are formulated in combination with intervals, survival rates and soil conditions. At the same time, planting history information is analyzed through an adaptive adjustment model to realize dynamic updating and adjustment of maintenance plans. This method breaks through the limitations of traditional fixed maintenance methods and provides a comprehensive technical solution for the unmanned planting history of smart farmland.

[0019] In order to effectively improve the efficiency and reliability of unmanned planting in smart farmland, this application provides an embodiment of a control method of a human brain-simulating robot based on artificial intelligence, see Figure 1 The control method of the artificial intelligence-based human brain simulation robot specifically includes the following contents: Step S101: Acquire crop growth information and soil quality monitoring information from multiple recorders in cultivated land, perform target analysis on the crop growth information to collect crop health status and crop organ development status, perform feature collection on the soil quality monitoring information to output cultivated land soil indicators, input the crop health status, the crop organ development status, and the cultivated land soil indicators into a pre-trained environment detection network, output real-time cultivated land environment information, perform semantic analysis on the real-time cultivated land environment information to output main feature indicators of the real-time environment, form an environmental feature sequence based on the main feature indicators, calculate the matching degree between the environmental feature sequence and the environmental description sequence in a preset environmental maintenance plan library to obtain a maintenance plan with the highest matching degree, match the interval data, survival rate information, and soil quality data in the real-time environmental information with a preset interval condition library, survival rate threshold condition library, and soil indicator condition library in turn to output associated interval maintenance sub-plan, survival rate maintenance sub-plan, and soil maintenance sub-plan, and integrate the sub-plans to form an environmental maintenance plan; Alternatively, this embodiment combines multi-source data acquisition with deep learning models to achieve adaptive identification of smart farmland environments and generate maintenance plans. First, multiple high-definition recorders are deployed across the farmland, covering key areas. Sensors for temperature, humidity, light, and air quality are also deployed to form a multi-dimensional data acquisition network.

[0020] This embodiment utilizes an improved target parsing algorithm to process crop growth information on cultivated land. Using a pre-trained model, the algorithm collects static crop characteristics, fertilization history, and pest and disease infestation data. For example, fruit key point analysis can identify whether crops are ripening before plowing or whether crop organs are being harvested. Furthermore, crop organ location areas are monitored in real time to identify the location and number of crop organs.

[0021] In terms of soil data processing, this embodiment performs dynamic correlation analysis and pest and disease value processing on various types of soil monitoring information. The changing trends of soil indicators are calculated using a sliding time window, and combined with the spatial layout characteristics of cultivated land, a multidimensional feature sequence reflecting the soil state is obtained.

[0022] Step S102: generating a crop growth dynamic log based on the crop health status, inputting the crop growth dynamic log into a pre-trained growth analysis model to determine whether there is pest growth; when pest growth is detected, storing the pest growth type, occurrence time, and location data in a pest record set; triggering a corresponding maintenance prompt method based on the pest growth risk level in the pest record set; and simultaneously adopting a related accident handling plan from a preset accident handling plan library based on the real-time farmland environment information and the pest growth type; Optionally, this embodiment addresses the problem of analyzing pest and disease growth in smart farmland by designing a growth identification solution based on dynamic feature analysis. Crop health status, including key fruit points, fertilization history, and pest and disease infestation patterns, collected from crop growth information is compiled into a health status log in timestamp order. To ensure the continuity and integrity of the feature log, a sliding window approach is used to segment the raw feature log, with the window size dynamically adjusted based on the duration of typical pest and disease growth.

[0023] Step S103: Perform statistical analysis on the real-time information of the cultivated land environment and the crop health status in the time dimension to output cultivated land planting history information, input the cultivated land planting history information into the adaptive adjustment model to obtain cultivated land maintenance plan adjustment content, dynamically update the maintenance plan in the preset environmental maintenance plan library according to the maintenance plan adjustment content, and at the same time transmit the cultivated land planting history information, the pest and disease record set and the maintenance plan adjustment content to the cloud maintenance platform for storage, and the cloud maintenance platform obtains a cultivated land planting history analysis report based on the acquired data.

[0024] As can be seen from the above description, the artificial intelligence-based simulated human brain robot control method provided in the embodiment of the present application can obtain cultivated land data through multiple recorders and sensors, and collect crop health status, crop organ development status and soil indicators. Innovatively form an environmental detection network and a growth analysis model to achieve intelligent identification of environmental conditions and real-time monitoring of the growth of pests and diseases. Dynamically match maintenance plans based on environmental feature sequences, and formulate detailed maintenance sub-plans based on intervals, survival rates and soil conditions. At the same time, the planting history information is analyzed through an adaptive adjustment model to achieve dynamic updating and adjustment of maintenance plans. This method breaks through the limitations of traditional fixed maintenance methods and provides a comprehensive technical solution for the unmanned planting history of smart farmland.

[0025] In one embodiment of the control method of the artificial intelligence-based human brain simulating robot of the present application, see Figure 2 , and may also include the following content: Step S201: performing feature fusion on the crop health status, the crop organ development status, and the cultivated land soil index according to a preset template to form a feature matrix, preprocessing the feature matrix, inputting the predicted feature data into an environmental monitoring network, and outputting real-time cultivated land environment information through the environmental monitoring network; Step S202: semantically split and word-segment associate the text record information in the real-time information of the cultivated land environment, collect the cultivated land rotation characteristics and soil quality characteristics in the environmental tags, and perform weighted calculation on each feature based on the preset feature item weights to output the main feature indicators of the real-time environment.

[0026] Optionally, this embodiment addresses the collection of environmental features and state identification for smart farmland by designing a multimodal feature fusion environmental analysis solution. First, crop health status (including plant static state, fertilization history, growth type, etc.), crop organ development status (including crop organ location, growth density, and category distribution), and farmland soil indicators (including lighting, temperature, and air quality) are serialized. During the feature fusion process, attention is used to dynamically adjust the weight templates of different features.

[0027] In one embodiment of the artificial intelligence-based human brain simulation robot control method of the present application, the following contents may also be specifically included: The Jaccard matching calculation method is used to calculate the matching scores between the pre-environmental feature sequence and each environmental description sequence in the preset environmental maintenance plan library. The matching scores are screened based on a preset matching threshold, and the maintenance plan associated with the environmental description sequence with the highest matching score that is greater than the matching threshold is obtained as the matching plan for the real-time environment.

[0028] Optionally, this embodiment designs a feature-sequencing-based environmental matching solution for preprocessing farmland environmental features and matching them with maintenance plans. During feature selection, key dimensions that significantly influence environmental status decisions, such as soil fertility density, activity intensity, and soil indices, are retained.

[0029] In practical applications, this embodiment can effectively handle various environments in the history of cultivated land cultivation.

[0030] This embodiment achieves precise matching between environmental features and maintenance plans through feature serialization and matching degree calculation. This solution can effectively respond to various environmental changes in cultivated land throughout its planting history and provide timely and accurate plans for maintaining soil fertility.

[0031] In one embodiment of the control method of the artificial intelligence-based human brain simulating robot of the present application, see Figure 3 , and may also include the following content: Step S301: parsing the real-time environmental information into interval data, survival rate information, and soil quality data; dividing the interval data according to a preset time granularity and comparing it with the rule items in the interval condition library to output an interval maintenance sub-plan; performing density calculation on the survival rate information and comparing it with the rule items in the survival rate threshold condition library to output a survival rate maintenance sub-plan; performing indicator classification on the soil quality data and comparing it with the rule items in the soil indicator condition library to output a soil maintenance sub-plan; Step S302: Based on the preset plan priority, the interval maintenance sub-plan, the survival rate maintenance sub-plan and the soil maintenance sub-plan are subjected to risk analysis and mitigation processing, and the risk-free sub-plans are integrated and adjusted through the plan fusion algorithm to obtain an environmental maintenance plan. According to the environmental maintenance plan, a fertilization control request, a growth monitoring request and a soil adjustment request are obtained.

[0032] This embodiment demonstrates good adaptability and reliability in practical applications. Through comprehensive analysis of interval characteristics, maturity characteristics, and soil characteristics, the system can accurately grasp environmental change trends and prepare maintenance plans in advance, effectively improving the level of intelligent farmland maintenance and historical planting efficiency.

[0033] In one embodiment of the control method of the artificial intelligence-based human brain simulating robot of the present application, see Figure 4 , and may also include the following content: Step S401: sorting the crop health status by timestamp and segmenting it into a health status log, performing time window division and feature completion on the health status log to output a fixed-length growth dynamic log, and preprocessing the static features, fertilization history features, and pest and disease control features in the growth dynamic log in sequence; Step S402: Input the predicted growth dynamic log into the pest and disease growth detection network formed based on the recurrent neural network, collect the dynamic mapping features and individual features in the growth log through the pest and disease growth detection network, analyze and classify the collected features based on the preset growth mode feature library, and output the pest and disease growth decision result.

[0034] This example uses deep learning technology to achieve real-time identification of pests and diseases in farmland environments. This solution can accurately capture subtle pests and diseases in growth logs and provide timely prompt information.

[0035] In one embodiment of the control method of the artificial intelligence-based human brain simulating robot of the present application, see Figure 5 , and may also include the following content: Step S501: Determine the type of pest growth obtained through the analysis through a preset growth action object, obtain spatial coordinate information of the pest growth, and write the pest record into a pest record set by including the pest growth type, timestamp, and spatial coordinate information. Score the pest growth in the pest record set according to a preset risk level assessment rule, and determine a warning level based on the risk level score. Step S502: According to the prompt level, the preset maintenance prompt condition library is queried to obtain a prompt request, the real-time information of the cultivated land environment and the growth type information of the pests and diseases are input into an accident handling plan selection model formed based on a decision tree, and the associated handling plan is obtained from the preset accident handling plan library through the accident handling plan selection model to obtain an accident handling request log.

[0036] Optionally, this embodiment designs a pest and disease record response and processing solution for smart farmland maintenance. First, the analyzed pest and disease growth is identified based on the growth action object, including multiple pest and disease types. A multi-target tracking algorithm is used to obtain real-time spatial positioning data of the pest and disease growth entity, ensuring the precise positioning of the pest and disease record.

[0037] This embodiment realizes the dynamic integration of plans in the process of obtaining accident handling requests. Based on the real-time environmental status and pest health status, the most suitable basic plan is obtained from the accident handling plan library, and then the plan is adjusted and adjusted according to the actual situation.

[0038] In one embodiment of the control method of the artificial intelligence-based human brain simulating robot of the present application, see Figure 6 , and may also include the following content: Step S601: The real-time farmland environment information and the crop health status are aggregated in segments according to a preset time window. Multi-dimensional feature collection is performed on the aggregated data to output the survival rate change trend, crop maturity cycle distribution, crop organ disease and pest infection frequency, and soil index change curve. Correlation analysis and trend prediction are performed on each indicator based on a preset data analysis model to obtain farmland planting history information. Step S602: Input the cultivated land planting history information into the plan adjustment model formed based on the reinforcement learning algorithm, evaluate and score the execution effect of the existing maintenance plan through the plan adjustment model, calculate the reward value of the plan execution based on the preset plan evaluation index, and use the plan gradient method to adjust the indicators of the maintenance plan to obtain the plan adjustment content.

[0039] In order to effectively improve the efficiency and reliability of unmanned planting in smart farmland, this application provides an embodiment of a human brain simulation robot control device based on artificial intelligence, see Figure 7 The artificial intelligence-based human brain simulation robot control device specifically includes the following contents: An environment determination module 10 is configured to obtain crop growth information and soil quality monitoring information from multiple recorders in the cultivated land, perform target analysis on the crop growth information to collect crop health status and crop organ development status, perform feature collection on the soil quality monitoring information to output cultivated land soil indicators, input the crop health status, crop organ development status, and cultivated land soil indicators into a pre-trained environment detection network, output real-time cultivated land environment information, perform semantic analysis on the real-time cultivated land environment information to output main characteristic indicators of the real-time environment, form an environmental feature sequence based on the main characteristic indicators, calculate a matching degree between the environmental feature sequence and an environmental description sequence in a preset environmental maintenance plan library to obtain a maintenance plan with the highest matching degree, match the interval data, survival rate information, and soil quality data in the real-time environmental information with a preset interval condition library, survival rate threshold condition library, and soil indicator condition library in sequence to output associated interval maintenance sub-plans, survival rate maintenance sub-plans, and soil maintenance sub-plans, and integrate the sub-plans to form an environmental maintenance plan; The accident handling module 20 is configured to generate a crop growth dynamic log based on the crop health status, input the crop growth dynamic log into a pre-trained growth analysis model to determine whether there is pest growth, and when pest growth is detected, store the pest growth type, occurrence time, and location data in a pest record set, trigger a corresponding maintenance prompt mode based on the pest growth risk level in the pest record set, and simultaneously adopt a related accident handling plan from a preset accident handling plan library based on the real-time farmland environment information and the pest growth type; The plan update module 30 is used to perform statistical analysis on the real-time information of the cultivated land environment and the health status of the crops in the time dimension to output the cultivated land planting history information, input the cultivated land planting history information into the adaptive adjustment model to obtain the cultivated land maintenance plan adjustment content, dynamically update the maintenance plan in the preset environment maintenance plan library according to the maintenance plan adjustment content, and at the same time transmit the cultivated land planting history information, the pest and disease record set and the maintenance plan adjustment content to the cloud maintenance platform for storage, and the cloud maintenance platform obtains the cultivated land planting history analysis report based on the acquired data.

[0040] As can be seen from the above description, the artificial intelligence-based simulated human brain robot control device provided in the embodiment of the present application can obtain farmland data through multiple recorders and sensors, and collect crop health status, crop organ development status and soil indicators. Innovatively form an environmental detection network and a growth analysis model to achieve intelligent identification of environmental conditions and real-time monitoring of the growth of pests and diseases. Dynamically match maintenance plans based on environmental feature sequences, and formulate detailed maintenance sub-plans based on intervals, survival rates and soil conditions. At the same time, the planting history information is analyzed through an adaptive adjustment model to achieve dynamic updating and adjustment of maintenance plans. This method breaks through the limitations of traditional fixed maintenance methods and provides a comprehensive technical solution for the unmanned planting history of smart farmland.

[0041] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0042] In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. The above is only an implementation method of the present application and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the content in the description and drawings of this application, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

[0043] The above detailed description of the specific embodiments of the invention is intended only as an example, and the present application is not limited to the specific embodiments described above. For the technical soil fertility in the field, any equivalent modification or substitution of the invention is also within the scope of the present application. Therefore, equalization, modification, and improvement made without departing from the spirit and principle of the present application should be included in the scope of the present application.

Claims

1. A control method for a human brain simulating robot based on artificial intelligence, characterized in that: The method comprises: Acquire crop growth information and soil quality monitoring information from multiple recorders in the cultivated land, perform target analysis on the crop growth information to collect crop health status and crop organ development status, perform feature collection on the soil quality monitoring information to output cultivated land soil indicators, input the crop health status, the crop organ development status and the cultivated land soil indicators into a pre-trained environment detection network, output real-time cultivated land environment information, perform semantic analysis on the real-time cultivated land environment information to output main feature indicators of the real-time environment, form an environmental feature sequence based on the main feature indicators, calculate the matching degree between the environmental feature sequence and the environmental description sequence in a preset environmental maintenance plan library to obtain a maintenance plan with the highest matching degree, match the interval data, survival rate information and soil quality data in the real-time environmental information with a preset interval condition library, survival rate threshold condition library and soil indicator condition library in turn to output associated interval maintenance sub-plan, survival rate maintenance sub-plan and soil maintenance sub-plan, and integrate the sub-plans to form an environmental maintenance plan; A crop growth dynamic log is generated based on the crop health status, and the crop growth dynamic log is input into a pre-trained growth analysis model to determine whether there is growth of pests and diseases. When pests and diseases are found, the growth type, occurrence time, and location data of the pests and diseases are stored in a pest and disease record set. A corresponding maintenance reminder method is triggered according to the risk level of pest and disease growth in the pest and disease record set. At the same time, a related accident treatment plan is adopted from a preset accident treatment plan library based on the real-time information of the cultivated land environment and the pest and disease growth type; A statistical analysis of the real-time information of the cultivated land environment and the crop health status in the time dimension is performed to output cultivated land planting history information, the cultivated land planting history information is input into an adaptive adjustment model to obtain cultivated land maintenance plan adjustment content, the maintenance plan in the preset environmental maintenance plan library is dynamically updated according to the maintenance plan adjustment content, and at the same time, the cultivated land planting history information, the pest and disease record set and the maintenance plan adjustment content are transmitted to a cloud maintenance platform for storage, and the cloud maintenance platform obtains a cultivated land planting history analysis report based on the acquired data.

2. The control method of the artificial intelligence-based human brain simulating robot according to claim 1, characterized in that: The step of inputting the crop health status, the crop organ development status, and the cultivated land soil index into a pre-trained environment detection network, outputting real-time cultivated land environment information, and performing semantic analysis on the real-time cultivated land environment information to output main characteristic indicators of the real-time environment includes: The crop health status, the crop organ development status and the cultivated land soil index are subjected to feature fusion according to a preset template to form a feature matrix, the feature matrix is ​​preprocessed, the predicted feature data is input into an environmental monitoring network, and real-time information of the cultivated land environment is output through the environmental monitoring network; The text record information in the real-time information of the cultivated land environment is semantically split and word-segmented, the cultivated land rotation characteristics and soil quality characteristics in the environmental tags are collected, and the main characteristic indicators of the real-time environment are output by weighted calculation of each feature based on the preset feature item weight.

3. The control method of the artificial intelligence-based human brain simulating robot according to claim 1, characterized in that: The forming of an environmental feature sequence according to the main feature indicators and performing matching calculation between the environmental feature sequence and an environmental description sequence in a preset environmental maintenance plan library to obtain a maintenance plan with the highest matching degree include: The Jaccard matching calculation method is used to calculate the matching scores between the pre-environmental feature sequence and each environmental description sequence in the preset environmental maintenance plan library. The matching scores are screened based on a preset matching threshold, and the maintenance plan associated with the environmental description sequence with the highest matching score that is greater than the matching threshold is obtained as the matching plan for the real-time environment.

4. The control method of the artificial intelligence-based human brain simulating robot according to claim 1, characterized in that: The interval data, survival rate information and soil quality data in the real-time environmental information are matched with the preset interval condition library, survival rate threshold condition library and soil index condition library in sequence to output the associated interval maintenance sub-plan, survival rate maintenance sub-plan and soil maintenance sub-plan, and the sub-plans are integrated to form an environmental maintenance plan, including: Parsing the real-time environmental information into interval data, survival rate information, and soil quality data, dividing the interval data according to a preset time granularity and comparing it with the rule items in the interval condition library to output an interval maintenance sub-plan, performing density calculation on the survival rate information and comparing it with the rule items in the survival rate threshold condition library to output a survival rate maintenance sub-plan, and performing indicator classification on the soil quality data and comparing it with the rule items in the soil indicator condition library to output a soil maintenance sub-plan; Based on the preset plan priorities, the interval maintenance sub-plan, the survival rate maintenance sub-plan and the soil maintenance sub-plan are subjected to risk analysis and mitigation processing, and the risk-free sub-plans are integrated and adjusted through a plan fusion algorithm to obtain an environmental maintenance plan. According to the environmental maintenance plan, fertilization control requests, growth monitoring requests and soil adjustment requests are obtained.

5. The control method of the artificial intelligence-based human brain simulating robot according to claim 1, characterized in that: The step of generating a crop growth dynamic log based on the crop health status and inputting the crop growth dynamic log into a pre-trained growth analysis model to determine whether pests and diseases are growing includes: sorting the crop health status according to timestamps and segmenting it into a health status log, performing time window division and feature completion processing on the health status log to output a fixed-length growth dynamic log, and sequentially preprocessing static features, fertilization history features, and pest and disease elimination features in the growth dynamic log; The predicted growth dynamic log is input into the pest and disease growth detection network formed based on the recurrent neural network. The dynamic mapping features and individual features in the growth log are collected through the pest and disease growth detection network. The collected features are analyzed and classified for pests and diseases based on the preset growth mode feature library, and the pest and disease growth decision results are output.

6. The control method of the artificial intelligence-based human brain simulating robot according to claim 1, characterized in that: When the growth of pests and diseases is detected, the growth type, occurrence time and location data of the pests and diseases are stored in a pest and disease record set, and a corresponding maintenance prompt mode is triggered according to the risk level of the pest and disease growth in the pest and disease record set. At the same time, according to the real-time information of the cultivated land environment and the growth type of the pests and diseases, a related accident handling plan is adopted from a preset accident handling plan library, including: Determine the type of pest growth obtained through the analysis through a preset growth action object, obtain spatial coordinate information of the occurrence of the pest growth, and write the pest record consisting of the pest growth type, timestamp, and spatial coordinate information into a pest record set. Score the pest growth in the pest record set according to a preset risk level assessment rule, and determine the prompt level based on the risk level score. According to the prompt level, the preset maintenance prompt condition library is queried to obtain a prompt request, the real-time information of the cultivated land environment and the growth type information of the pests and diseases are input into an accident treatment plan selection model formed based on a decision tree, and the associated treatment plan is obtained from the preset accident treatment plan library through the accident treatment plan selection model to obtain an accident treatment request log.

7. The control method of the artificial intelligence-based human brain simulating robot according to claim 1, characterized in that: The statistical analysis of the real-time farmland environment information and the crop health status in the time dimension is performed to output farmland planting history information, and the farmland planting history information is input into the adaptive adjustment model to obtain farmland maintenance plan adjustment content, including: The real-time information of the farmland environment and the health status of the crops are aggregated in segments according to a preset time window, and multi-dimensional feature collection is performed on the aggregated data to output the survival rate change trend, crop maturity cycle distribution, crop organ disease and pest infection frequency, and soil index change curve. Based on a preset data analysis model, correlation analysis and trend prediction are performed on each indicator to obtain the historical information of farmland planting; The cultivated land planting history information is input into a plan adjustment model formed based on a reinforcement learning algorithm. The execution effect of the existing maintenance plan is evaluated and scored through the plan adjustment model. The reward value of the plan execution is calculated based on the preset plan evaluation index. The plan gradient method is used to adjust the indicators of the maintenance plan to obtain the plan adjustment content.

8. An artificial intelligence-based human brain-simulating robot control device, characterized in that: include: an environment determination module for acquiring cultivated land crop growth information and soil quality monitoring information from multiple recorders in cultivated land, performing target analysis on the cultivated land crop growth information to collect crop health status and crop organ development status, performing feature collection on the soil quality monitoring information to output cultivated land soil indicators, inputting the crop health status, the crop organ development status, and the cultivated land soil indicators into a pre-trained environment detection network to output real-time cultivated land environment information, performing semantic analysis on the real-time cultivated land environment information to output main characteristic indicators of the real-time environment, forming an environmental feature sequence based on the main characteristic indicators, performing matching calculation on the environmental feature sequence with an environmental description sequence in a preset environmental maintenance plan library to obtain a maintenance plan with the highest matching degree, matching the interval data, survival rate information, and soil quality data in the real-time environmental information with a preset interval condition library, a survival rate threshold condition library, and a soil indicator condition library in sequence to output associated interval maintenance sub-plans, survival rate maintenance sub-plans, and soil maintenance sub-plans, and integrating the sub-plans to form an environmental maintenance plan; An accident handling module is configured to generate a crop growth dynamic log based on the crop health status, input the crop growth dynamic log into a pre-trained growth analysis model to determine whether there is growth of pests and diseases, and when pests and diseases are detected, store the growth type, occurrence time, and location data of the pests and diseases in a pest and disease record set; trigger a corresponding maintenance prompt method based on the risk level of pest and disease growth in the pest and disease record set; and simultaneously adopt a related accident handling plan from a preset accident handling plan library based on the real-time information of the cultivated land environment and the pest and disease growth type; a plan update module for performing a statistical analysis of the real-time farmland environment information and the crop health status in a time dimension to output farmland planting history information, inputting the farmland planting history information into an adaptive adjustment model to obtain farmland maintenance plan adjustment content, dynamically updating the maintenance plan in the preset environmental maintenance plan library based on the maintenance plan adjustment content, and simultaneously transmitting the farmland planting history information, the pest and disease record set, and the maintenance plan adjustment content to a cloud maintenance platform for storage, wherein the cloud maintenance platform obtains a farmland planting history analysis report based on the acquired data; The artificial intelligence-based human brain simulation robot control device is used to execute the artificial intelligence-based human brain simulation robot control method as described in any one of claims 1 to 7.

Images