Robot remote cooperative scheduling method and system based on artificial intelligence

Abstract

The invention relates to the technical field of data processing, provides a robot remote collaborative scheduling method and system based on artificial intelligence, and aims to solve the problems of large data noise and low decision accuracy in a crop planting environment. By preprocessing sensor data, constructing a crop growth simulation model, monitoring a farmland environment in real time and optimizing a data processing algorithm, precise scheduling of robot operation is realized. The core of the invention lies in improving the scheduling efficiency of the agricultural production equipment, reducing the cost, enhancing the decision accuracy and promoting the agricultural development by using the data processing technology. The application of the invention can significantly improve the intelligent level of agricultural production, and provides powerful support for agricultural modernization.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a method and system for remote collaborative scheduling of robots based on artificial intelligence. Background Art

[0002] In the field of crop planting, a remote collaborative scheduling system for robots based on artificial intelligence is gradually showing its application potential. The remote collaborative scheduling of robots based on artificial intelligence realizes remote monitoring and intelligent scheduling of the whole process of crop planting by integrating sensor technology, the Internet of Things, big data analysis, and artificial intelligence algorithms. For example, in the farmland management link, the system can collect data such as soil humidity, temperature, and nutrient content in real time, and analyze and process it through data processing algorithms, so as to accurately guide operations such as irrigation and fertilization. At the same time, the system can also remotely schedule intelligent agricultural machinery and equipment, such as unmanned tractors, self-propelled sprayers, etc., to achieve autonomous navigation and precise operations, greatly improving agricultural production efficiency. In addition, in the prevention and control of pests and diseases, the remote collaborative scheduling system for robots based on artificial intelligence can also play an important role. Through image recognition and deep learning algorithms, it can achieve early warning and precise prevention and control of pests and diseases, effectively ensuring the healthy growth of crops.

[0003] However, in the application process of the remote collaborative scheduling system for robots based on artificial intelligence in the field of crop planting, due to the complex and changeable crop planting environment, the data collected by sensors often have noise and uncertainty, which affect the analysis and decision-making accuracy of data processing algorithms. For example, environmental factors such as soil humidity and temperature are affected by various factors such as weather and terrain, resulting in large fluctuations in measurement data. Any delay or error in the data will lead to mistakes in agricultural production decisions, and it is necessary to continuously optimize the accuracy of data transmission. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention provides a method and system for remote collaborative scheduling of robots based on artificial intelligence, which solves the problem that due to the complex and changeable crop planting environment, the data collected by sensors often have noise and uncertainty, affecting the analysis and decision-making accuracy of data processing algorithms.

[0005] To solve the above technical problems, the specific technical solutions of the present invention are as follows: In the first aspect, the method for remote collaborative scheduling of robots based on artificial intelligence provided by the present invention includes: Step S101: Real-time collect farmland environmental data, where the farmland environmental data includes farmland soil humidity, farmland soil temperature, and farmland soil nutrient content. Preprocess the collected farmland environmental data to obtain preprocessed farmland environmental data. Extract data features from the preprocessed farmland environmental data, and then perform weighted evaluation on the extracted farmland environmental data features to obtain a farmland environmental data evaluation result; Step S102: Receive the standard parameters for using farmland environmental data. Based on the standard parameters for using farmland environmental data, screen out the valid farmland environmental data from the farmland environmental data evaluation result. Extract features from the valid farmland environmental data based on the preset sensor identity information to obtain a sensor data group. Classify the data in the sensor data group according to a preset time period to obtain sensor time-segmented data. Substitute the sensor time-segmented data into a preset warning model to obtain sensor warning data; Step S103: If there is incorrect data in the sensor warning data, retrieve the sensor identity information corresponding to the incorrect data in the sensor warning data, collect the time information of the sensor identity information corresponding to the incorrect data in the sensor warning data, and substitute the time information of the sensor identity information corresponding to the incorrect data in the sensor warning data into a preset crop growth data simulation model to obtain the simulation data corresponding to the incorrect data; Step S104: Replace the real-time collected farmland environmental data with the simulation data corresponding to the incorrect data to obtain modified farmland environmental data. Collect real-time farmland crop image data, and substitute the modified farmland environmental data into a preset crop growth image model to generate an expected farmland crop image; Step S105: Perform hashing on the expected image of the farm crops to obtain the hash value of the expected image of the farm crops. Substitute the expected image of the farm crops and the hash value of the expected image of the farm crops into the preset metaverse image model to obtain the metaverse image of the farm crops. Perform hashing on the real-time collected farmland environment data to obtain the hash value of the real-time collected farmland environment data. Substitute the real-time collected farmland environment data and the hash value of the real-time collected farmland environment data into the preset metaverse image model to obtain the metaverse environment data of the farm crops. Configure the metaverse environment data of the farm crops in the metaverse image of the farm crops. After the configuration is complete, run the preset metaverse image model. If the preset metaverse image model runs normally after the configuration is complete, retrieve the operation data of the preset metaverse image model after the configuration is complete. Match the operation data of the preset metaverse image model after the configuration is complete with the crop production equipment to obtain the operation data of the metaverse crop production equipment. Substitute the operation data of the metaverse crop production equipment into the preset artificial intelligence robot scheduling model to generate an artificial intelligence robot scheduling command. Collect the image data after the robot executes the artificial intelligence robot scheduling command. Then substitute the image data after the robot executes the artificial intelligence robot scheduling command into the preset metaverse image model after the configuration is complete. If the preset metaverse image model runs abnormally after the configuration is complete, generate a warning message for the error in the execution of the scheduling command.

[0006] Further, in the method for remote cooperative scheduling of robots based on artificial intelligence according to the present invention, the step S102 includes: Receive the standard parameters for the use of farmland environment data, and the standard parameters for the use of farmland environment data include soil environment parameters, meteorological environment parameters, and water quality environment parameters; From the evaluation results of the preprocessed farmland environment data, compare each piece of data in the evaluation results of the preprocessed farmland environment data with the standard parameters for the use of farmland environment data, and only retain the data entries that meet all the standard parameters for the use of farmland environment data as the effective farmland environment data; For each piece of effective farmland environment data, according to the sensor identity information associated with each piece of effective farmland environment data, extract the data features related to the sensor identity information. The data features related to the sensor identity information include the average value, maximum value, and minimum value within a period. Combine the data features related to the sensor identity information into a sensor data group, and each sensor data group corresponds to a sensor and the effective data features collected by the sensor.

[0007] Further, in the method for remote cooperative scheduling of robots based on artificial intelligence according to the present invention, the step S103 includes: In the sensor warning data, identify the existing error data and locate the position of the error data in the warning data; According to the location information of the error data, the sensor identity information corresponding to the error data is retrieved, where the sensor identity information includes the identification, location and type of the sensor; Collect the time information when the error data occurs, including the timestamp of year, month, day, hour and minute; Substituting the time information and the sensor identity information into a preset crop growth data simulation model, running the preset crop growth data simulation model, and calculating the crop growth state when the error data occurs according to the input time information and the sensor identity information; The preset crop growth data simulation model outputs simulation data corresponding to the erroneous data, and verifies the generated simulation data.

[0008] Furthermore, the artificial intelligence-based robot remote collaborative scheduling method of the present invention, step S104, includes: Substituting the farmland environment data into a preset crop growth image model, the preset crop growth image model calculates the expected growth state of the crop based on the input farmland environment data; Using image processing technology, the calculation results are converted into visual images of expected crops; The generated expected image of farmland crops is verified. If there are any abnormalities or unreasonableness in the expected image of farmland crops, return to check the data replacement and model substitution steps.

[0009] In a second aspect, the present invention provides a robot remote collaborative scheduling system based on artificial intelligence, which is applied to the robot remote collaborative scheduling method based on artificial intelligence, comprising: The data acquisition module is used to collect farmland environmental data in real time. The farmland environmental data includes farmland soil moisture, farmland soil temperature and farmland soil nutrient content. The collected farmland environmental data is preprocessed to obtain preprocessed farmland environmental data, data features are extracted from the preprocessed farmland environmental data, and then weighted evaluation is performed on the extracted farmland environmental data features to obtain farmland environmental data evaluation results. The first data processing module is used to receive the farmland environment data usage standard parameters, screen out effective farmland environment data from the farmland environment data evaluation results based on the farmland environment data usage standard parameters, extract features of the effective farmland environment data based on preset sensor identity information to obtain a sensor data group, classify the data of the sensor data group according to a preset time period to obtain sensor time segment data, substitute the sensor time segment data into a preset early warning model to obtain sensor early warning data; Second data processing module: If there is incorrect data in the sensor warning data, it is used to retrieve the sensor identity information corresponding to the incorrect data in the sensor warning data, collect the time information of the sensor identity information corresponding to the incorrect data in the sensor warning data, and substitute the time information of the sensor identity information corresponding to the incorrect data in the sensor warning data into the preset crop growth data simulation model to obtain the simulation data corresponding to the incorrect data; Image generation module: It is used to replace the real-time collected farmland environment data with the simulation data corresponding to the incorrect data to obtain the modified farmland environment data, collect the real-time farmland crop image data, and substitute the modified farmland environment data into the preset crop growth image model to generate the expected farmland crop image; Data analysis module: It is used to perform hash processing on the expected farmland crop image to obtain the hash value of the expected farmland crop image, substitute the expected farmland crop image and the hash value of the expected farmland crop image into the preset metaverse image model to obtain the metaverse image of the farmland crop, perform hash processing on the real-time collected farmland environment data to obtain the hash value of the real-time collected farmland environment data, substitute the real-time collected farmland environment data and the hash value of the real-time collected farmland environment data into the preset metaverse image model to obtain the metaverse environment data of the farmland crop, configure the metaverse environment data of the farmland crop in the metaverse image of the farmland crop, and run the preset metaverse image model after the configuration is complete. If the preset metaverse image model runs normally after the configuration is complete, retrieve the operation data of the preset metaverse image model after the configuration is complete, match the operation data of the preset metaverse image model after the configuration is complete with the crop production equipment to obtain the operation data of the metaverse crop production equipment, substitute the operation data of the metaverse crop production equipment into the preset artificial intelligence robot scheduling model to generate an artificial intelligence robot scheduling command, collect the image data after the robot executes the artificial intelligence robot scheduling command, and then substitute the image data after the robot executes the artificial intelligence robot scheduling command into the preset metaverse image model after the configuration is complete. If the preset metaverse image model runs abnormally after the configuration is complete, generate a warning information for the error in the execution of the scheduling command.

[0010] Furthermore, in the robot remote cooperative scheduling system based on artificial intelligence of the present invention, the first data processing module is further used for: Receiving the standard parameters for the use of farmland environment data, where the standard parameters for the use of farmland environment data include soil environment parameters, meteorological environment parameters, and water quality environment parameters; From the evaluation results of the preprocessed farmland environment data, comparing each data in the evaluation results of the preprocessed farmland environment data with the standard parameters for the use of farmland environment data, and only retaining the data entries that meet all the standard parameters for the use of farmland environment data as the valid farmland environment data; For each valid farmland environmental data, data features related to the sensor identity information are extracted according to the sensor identity information associated with each valid farmland environmental data. The data features related to the sensor identity information include the average value, maximum value, and minimum value within a time period. The data features related to the sensor identity information are combined into a sensor data group. Each sensor data group corresponds to a sensor and the valid data features collected by the sensor.

[0011] Furthermore, in the artificial intelligence-based robot remote collaborative scheduling system of the present invention, the second data processing module is also used for: In the sensor warning data, identify the existing erroneous data and locate the position of the erroneous data in the warning data; According to the location information of the error data, the sensor identity information corresponding to the error data is retrieved, where the sensor identity information includes the identification, location and type of the sensor; Collect the time information when the error data occurs, including the timestamp of year, month, day, hour and minute; Substituting the time information and the sensor identity information into a preset crop growth data simulation model, running the preset crop growth data simulation model, and calculating the crop growth state when the error data occurs according to the input time information and the sensor identity information; The preset crop growth data simulation model outputs simulation data corresponding to the erroneous data, and verifies the generated simulation data.

[0012] Furthermore, in the artificial intelligence-based robot remote collaborative scheduling system of the present invention, the image generation module is also used for: Substituting the farmland environment data into a preset crop growth image model, the preset crop growth image model calculates the expected growth state of the crop based on the input farmland environment data; Using image processing technology, the calculation results are converted into visual images of expected crops; The generated expected image of farmland crops is verified. If there are any abnormalities or unreasonableness in the expected image of farmland crops, return to check data replacement and model substitution.

[0013] Beneficial effects of the present invention: The artificial intelligence-based robot remote collaborative scheduling method of the present invention has shown significant beneficial effects in the crop planting environment, which is specifically reflected in the following aspects: By receiving and analyzing farmland environmental data, such as soil moisture, temperature, light, etc., the data processing algorithm can accurately guide the robot to carry out farming activities such as sowing, fertilizing, and irrigating, achieving precise operations, reducing resource waste, and improving production efficiency. The robot can automatically complete farming activities according to the instructions of the data processing algorithm without manual intervention, greatly reducing the labor intensity of farmers and improving the operation efficiency.

[0014] Through precise operations and automated operations, the present invention can significantly reduce the waste of agricultural resources such as chemical fertilizers, pesticides, and water resources, and reduce production costs. The robot can replace manual labor to complete most farming activities, reducing the dependence on human resources and thus reducing labor costs.

[0015] The present invention preprocesses and filters the noise of the data collected by the sensor, eliminates the outliers and noise in the data, improves the accuracy and reliability of the data, and thus enhances the decision-making accuracy of the data processing algorithm. By continuously optimizing the data processing model and algorithm, the present invention can more accurately predict the crop growth status, the occurrence of pests and diseases, etc., and provide more scientific decision-making basis for farmers.

[0016] Precision management based on the data processing algorithm can ensure that crops grow in the best growth environment, thereby increasing the yield and quality of crops. By real-time monitoring and analyzing crop growth environment data, the present invention can timely detect and warn of the occurrence of pests and diseases, guide farmers to take effective control measures, and reduce the damage of pests and diseases to crops. Through precise fertilization and spraying, the present invention can significantly reduce the usage amount of chemical fertilizers and pesticides, reduce environmental pollution, and promote the sustainable development of agriculture. By optimizing the allocation and utilization of agricultural resources, the present invention can improve resource utilization rate, reduce resource waste, and achieve green production of agriculture.

[0017] In summary, the robot remote collaborative scheduling method based on artificial intelligence of the present invention shows significant beneficial effects in the crop planting environment, including improving production efficiency, reducing costs, enhancing decision-making accuracy, increasing crop yield and quality, and promoting the sustainable development of agriculture, etc. Brief Description of the Drawings

[0018] In order to more clearly illustrate the technical solution of the present invention, the drawings required to be used in the embodiments will be briefly introduced below. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to the drawings.

[0019] Figure 1 It is a schematic flowchart of the robot remote collaborative scheduling method based on artificial intelligence provided by the embodiment of the present invention. Detailed Embodiments

[0020] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention. The technical solutions provided by each embodiment of the present invention will be described in detail below with reference to the drawings.

[0021] To better understand the objectives of the present invention, the present invention will be further described in detail below.

[0022] As Figure 1 shown, in a first aspect, the robot remote collaborative scheduling method based on artificial intelligence provided by the present invention includes: Step S101: Collect farmland environmental data in real time. The farmland environmental data includes farmland soil humidity, farmland soil temperature, and farmland soil nutrient content. Preprocess the collected farmland environmental data to obtain preprocessed farmland environmental data. Extract data features from the preprocessed farmland environmental data, and then perform weighted evaluation on the extracted farmland environmental data features to obtain a farmland environmental data evaluation result; Data collection: Use sensors installed in the farmland to collect farmland environmental data in real time, including farmland soil humidity, farmland soil temperature, and farmland soil nutrient content.

[0023] The sensors should be calibrated regularly to ensure the accuracy of the data.

[0024] Preprocessing: Denoise the collected raw data to eliminate random errors and outliers.

[0025] Smooth the data to reduce data fluctuations and improve data quality.

[0026] Feature extraction: Extract key features from the preprocessed data, such as the average value, maximum value, minimum value, etc. of the soil humidity.

[0027] These features should be able to reflect the main change trends and states of the farmland environment.

[0028] Weighted evaluation: Assign weights to each feature according to the influence degree of different features on the farmland environmental state.

[0029] Calculate the weighted average value or weighted sum to obtain a farmland environmental data evaluation result.

[0030] Step S102: Receive the standard parameters for using farmland environmental data. Based on the standard parameters for using farmland environmental data, screen out the valid farmland environmental data from the evaluation results of farmland environmental data. Extract features from the valid farmland environmental data based on the preset sensor identity information to obtain a sensor data group. Classify the data in the sensor data group according to a preset time period to obtain sensor time-segmented data. Substitute the sensor time-segmented data into a preset warning model to obtain sensor warning data; Receiving standard parameters: Obtain the standard parameters for using farmland environmental data from relevant standards or databases, including the reasonable ranges of soil humidity, soil temperature, soil nutrient content, etc.

[0031] Data screening: Compare the preprocessed evaluation results of farmland environmental data with the standard parameters, and screen out the valid farmland environmental data that meet the standards.

[0032] Feature extraction: For the valid farmland environmental data, perform further feature extraction according to the preset sensor identity information (such as sensor type, location, etc.). Extract data features related to the sensor identity information, such as the average value, maximum value, minimum value, etc. within a specific time period.

[0033] Data classification: Classify the sensor data group according to a preset time period (such as hours, days, weeks, etc.) to obtain sensor time-segmented data.

[0034] Model substitution: Substitute the sensor time-segmented data into a preset warning model, and the model calculates and outputs sensor warning data based on the input data.

[0035] Step S103: If there is incorrect data in the sensor warning data, retrieve the sensor identity information corresponding to the incorrect data in the sensor warning data, collect the time information of the sensor identity information corresponding to the incorrect data in the sensor warning data, and substitute the time information of the sensor identity information corresponding to the incorrect data in the sensor warning data into a preset crop growth data simulation model to obtain the simulation data corresponding to the incorrect data; Identifying incorrect data: Identify the incorrect data in the sensor warning data that is significantly inconsistent with the actual situation.

[0036] Retrieving identity information: According to the location information of the incorrect data, retrieve the sensor identity information corresponding to the incorrect data from the database.

[0037] Collecting time information: Record the time information when the incorrect data occurs, including the specific time stamps such as year, month, day, hour, minute, etc.

[0038] Model substitution and simulation data generation: Substitute the time information and sensor identity information corresponding to the incorrect data into a preset crop growth data simulation model.

[0039] The model calculates the growth state of crops when error data occurs based on the input information and outputs simulated data to replace the error data.

[0040] Step S104: Replace the real-time collected farmland environment data with the simulated data corresponding to the error data to obtain the modified farmland environment data. Collect the real-time farmland crop image data, and substitute the modified farmland environment data into the preset crop growth image model to generate the expected image of the farmland crops. Data replacement: Use the simulated data generated in step S103 to replace the error data in the real-time collected farmland environment data to obtain the modified farmland environment data.

[0041] Image data collection: Use devices such as drones or ground cameras to collect real-time farmland crop image data to reflect the current growth state of the crops.

[0042] Model substitution and image generation: Substitute the modified farmland environment data into the preset crop growth image model.

[0043] The model calculates the expected growth state of the crops based on the input data and uses image processing technology to generate the expected image of the farmland crops.

[0044] Step S105: Perform hash processing on the expected image of the farmland crops to obtain the hash value of the expected image of the farmland crops. Substitute the expected image of the farmland crops and the hash value of the expected image of the farmland crops into the preset metaverse image model to obtain the metaverse image of the farmland crops. Perform hash processing on the real-time collected farmland environment data to obtain the hash value of the real-time collected farmland environment data. Substitute the real-time collected farmland environment data and the hash value of the real-time collected farmland environment data into the preset metaverse image model to obtain the metaverse environment data of the farmland crops. Configure the metaverse environment data of the farmland crops in the metaverse image of the farmland crops. After the configuration is complete, run the preset metaverse image model. If the preset metaverse image model runs normally after the configuration is complete, retrieve the running data of the preset metaverse image model after the configuration is complete. Match the running data of the preset metaverse image model after the configuration is complete with the crop production equipment to obtain the running data of the metaverse crop production equipment. Substitute the running data of the metaverse crop production equipment into the preset artificial intelligence robot scheduling model to generate an artificial intelligence robot scheduling command. Collect the image data after the robot executes the artificial intelligence robot scheduling command, and then substitute the image data after the robot executes the artificial intelligence robot scheduling command into the preset metaverse image model after the configuration is complete. If the preset metaverse image model runs abnormally after the configuration is complete, generate a warning message for the error in the execution of the scheduling command.

[0045] Hashing Process: Use a secure hashing algorithm (such as SHA-256) to hash the expected images of farm crops, generating a unique hash value as the identifier for the expected images of farm crops.

[0046] Perform the same hashing process on the real-time collected farmland environmental data to generate corresponding hash values.

[0047] Model Substitution: Take the hashed expected images of farm crops and their hash values as inputs and substitute them into a preset metaverse image model.

[0048] At the same time, also substitute the real-time collected farmland environmental data and their hash values into the metaverse image model as inputs for environmental data.

[0049] Configuration: In the metaverse image model, configure the metaverse scene according to the input expected images of farm crops and environmental data. This includes adjusting environmental factors such as the growth state of crops, lighting, and temperature to simulate a real farmland environment.

[0050] Model Execution: Start the metaverse image model to let it run in the configured metaverse scene. Monitor the running state of the model to ensure that all components work properly without any abnormal error reports.

[0051] Data Matching: If the metaverse image model runs normally, extract the operation data of the metaverse crop production equipment from its output. Match these data with the actual crop production equipment data to verify their accuracy and feasibility.

[0052] Scheduling Command Generation: Substitute the matched operation data of the metaverse crop production equipment into a preset artificial intelligence robot scheduling model. The scheduling model calculates and generates scheduling commands for the artificial intelligence robot based on the input data, including the moving path and operation tasks of the robot.

[0053] Image Data Collection: During the process of the robot executing the scheduling commands, use a camera or other image collection devices to collect real-time operation image data of the robot. These data are used to verify whether the robot executes the operations correctly according to the scheduling commands.

[0054] Early Warning Information Generation: Substitute the collected operation image data of the robot into the fully configured metaverse image model for further analysis and verification. If it is found that the robot operation does not match the expectation or the metaverse image model runs abnormally, immediately generate an early warning message for the execution error of the scheduling command. The early warning message should include key information such as the specific description of the error, the occurrence time, and the impact scope, so that relevant personnel can respond and handle it in a timely manner.

[0055] Specifically, for the method of remote collaborative scheduling of robots based on artificial intelligence described in the present invention, step S102 includes: Receive the standard parameters for farmland environment data. The standard parameters for farmland environment data include soil environment parameters, meteorological environment parameters, and water quality environment parameters; From the evaluation results of the preprocessed farmland environment data, compare each piece of data in the evaluation results of the preprocessed farmland environment data with the standard parameters for farmland environment data, and only retain the data entries that meet all the standard parameters for farmland environment data as valid farmland environment data; For each piece of valid farmland environment data, extract the data features related to the sensor identity information according to the sensor identity information associated with each piece of valid farmland environment data. The data features related to the sensor identity information include the average value, maximum value, and minimum value within a period, and combine the data features related to the sensor identity information into a sensor data group. Each sensor data group corresponds to a sensor and the valid data features collected by the sensor.

[0056] Receiving standard parameters: Receive the standard parameters for farmland environment data from a relevant database or standard document. These parameters include soil environment parameters (such as soil humidity, soil pH value, etc.), meteorological environment parameters (such as temperature, humidity, wind speed, etc.), and water quality environment parameters (such as dissolved oxygen, pH value, turbidity, etc.).

[0057] Data comparison and screening: From the evaluation results of the preprocessed farmland environment data, compare each piece of data with the received standard parameters for farmland environment data one by one. The comparison process should comprehensively consider all relevant parameters to ensure the accuracy and integrity of the data. Only retain those data entries that fully meet all the standard parameters for farmland environment data, and mark these data entries as valid farmland environment data.

[0058] Feature extraction: For each piece of valid farmland environment data, perform feature extraction according to its associated sensor identity information (such as sensor ID, location, type, etc.). The extracted features should include the average value, maximum value, minimum value, etc. within a period, and these features can reflect the data change situation of the sensor within a specific time period. During the extraction process, ensure the accuracy and consistency of the data, and avoid introducing noise or errors.

[0059] Data combination: Combine the extracted data features related to the sensor identity information into a sensor data group. Each sensor data group should correspond to a specific sensor and its collected valid data features.

[0060] Specifically, for the method for remote collaborative scheduling of robots based on artificial intelligence described in the present invention, step S103 includes: In the sensor warning data, identify the existing incorrect data and locate the position of the incorrect data in the warning data; Retrieve the sensor identity information corresponding to the error data based on the location information of the error data. The sensor identity information includes the identifier, location, and type of the sensor; Collect the time information when the error data occurs. The time information includes the time stamps of year, month, day, hour, and minute; Substitute the time information and the sensor identity information into a preset crop growth data simulation model, run the preset crop growth data simulation model, and calculate the crop growth status at the time when the error data occurs according to the input time information and sensor identity information; The preset crop growth data simulation model outputs the simulation data corresponding to the error data, and verify the generated simulation data.

[0061] Error data identification and location: In the sensor warning data, use data verification algorithms (such as range check, consistency check, etc.) to automatically identify the existing error data. The error data includes outliers, missing values, or data with incorrect formats.

[0062] After identifying the error data, locate the specific location of the error data in the warning data through data indexing or unique identifiers.

[0063] Sensor identity information retrieval: According to the location information of the error data, retrieve the sensor identity information corresponding to the error data from the database or data warehouse. The sensor identity information should include the unique identifier of the sensor (such as sensor ID), installation location (such as the specific coordinates of the farmland), and sensor type (such as temperature sensor, humidity sensor, etc.).

[0064] Time information collection: Collect the time information when the error data occurs. The time information should be accurate to the time stamps of year, month, day, hour, and minute. The time information can be directly extracted from the sensor data record, or calibrated through other time synchronization systems (such as GPS time).

[0065] Model substitution and operation: Substitute the collected time information and sensor identity information into a preset crop growth data simulation model. The crop growth data simulation model should be established based on the crop growth mechanism and environmental factors (such as temperature, humidity, light, etc.). When running the model, calculate the crop growth status at the time when the error data occurs according to the input time information and sensor identity information.

[0066] Simulation data verification: After the preset crop growth data simulation model outputs the simulation data corresponding to the error data, it is necessary to verify the generated simulation data. The verification methods include comparison with the actual situation, cross-verification with other sensor data, etc.

[0067] If there are significant deviations between the simulated data and the actual situation or other reliable data, it is necessary to further analyze the reasons and adjust the model parameters or algorithms.

[0068] Specifically, for the robot remote cooperative scheduling method based on artificial intelligence described in the present invention, the step S104 includes: Substitute the farmland environment data into a preset crop growth image model. The preset crop growth image model calculates the expected growth state of the crops based on the input farmland environment data. Using image processing technology, convert the calculation result into a visual image of the expected crops. Verify the generated image of the expected farmland crops. If there are abnormalities or unreasonable points in the image of the expected farmland crops, return to the steps of checking data replacement and model substitution.

[0069] Data substitution into the model: Select farmland environment data related to crop growth from the database, including key indicators such as soil humidity, temperature, light intensity, CO 2 concentration, etc. Clean the data to remove outliers and missing values to ensure the accuracy and integrity of the data.

[0070] Model substitution: Substitute the preprocessed farmland environment data into a preset crop growth image model. The model should be constructed based on the crop growth mechanism and environmental impact factors and be able to accurately reflect the growth state of crops under different environmental conditions.

[0071] Image processing technology: Visualization of calculation results: Using image processing technology, convert the expected growth state of the crops calculated by the crop growth image model into a visual image of the expected farmland crops. This can be achieved through rendering technology, color mapping, etc., to make the image more intuitive and easy to understand.

[0072] Image optimization: Perform necessary optimization processing on the generated image, such as adjusting contrast, brightness, saturation, etc., to improve the quality and readability of the image.

[0073] Image verification: Verification criteria: Establish clear image verification criteria, including image clarity, color accuracy, and the reasonableness of the crop growth state. These criteria should be formulated based on the actual situation of crop growth and the accuracy of model prediction.

[0074] Verification process: Adopt a combination of automatic verification and manual verification to verify the generated image of the expected farmland crops. Automatic verification can be achieved by setting thresholds, comparing historical data, etc.; manual verification is carried out by professional personnel to check each image one by one.

[0075] Problem handling: If any abnormalities or unreasonableness are found in the expected images of farm crops, the data replacement and model substitution steps should be immediately retraced. First, check the accuracy and integrity of the input data. After confirmation, re-substitute the data into the model for calculation. If the problem persists, the model parameters need to be adjusted or the model algorithm optimized.

[0076] In a second aspect, the present invention provides an artificial intelligence-based robot remote collaborative scheduling system, which is applied to the artificial intelligence-based robot remote collaborative scheduling method described above and includes: A data acquisition module, which is used to collect farmland environment data in real time. The farmland environment data includes farmland soil humidity, farmland soil temperature, and farmland soil nutrient content. The collected farmland environment data is preprocessed to obtain preprocessed farmland environment data. Data feature extraction is performed on the preprocessed farmland environment data, and then weighted evaluation is performed on the extracted farmland environment data features to obtain a farmland environment data evaluation result; A first data processing module: It is used to receive the standard parameters for using farmland environment data, screen out valid farmland environment data from the farmland environment data evaluation result based on the standard parameters for using farmland environment data, perform feature extraction on the valid farmland environment data based on the preset sensor identity information to obtain a sensor data group, classify the data of the sensor data group according to a preset time period to obtain sensor time-segmented data, and substitute the sensor time-segmented data into a preset warning model to obtain sensor warning data; A second data processing module: If there are incorrect data in the sensor warning data, it is used to retrieve the sensor identity information corresponding to the incorrect data in the sensor warning data, collect the time information of the sensor identity information corresponding to the incorrect data in the sensor warning data, and substitute the time information of the sensor identity information corresponding to the incorrect data in the sensor warning data into a preset crop growth data simulation model to obtain the simulation data corresponding to the incorrect data; An image generation module: It is used to replace the real-time collected farmland environment data with the simulation data corresponding to the incorrect data to obtain modified farmland environment data, collect real-time farm crop image data, and substitute the modified farmland environment data into a preset crop growth image model to generate an expected image of farm crops; Data analysis module: It is used to perform hashing processing on the expected images of farmland crops to obtain the hash values of the expected images of farmland crops. Substitute the expected images of farmland crops and the hash values of the expected images of farmland crops into the preset metaverse image model to obtain the metaverse images of farmland crops. Perform hashing processing on the real-time collected farmland environment data to obtain the hash values of the real-time collected farmland environment data. Substitute the real-time collected farmland environment data and the hash values of the real-time collected farmland environment data into the preset metaverse image model to obtain the metaverse environment data of farmland crops. Configure the metaverse environment data of farmland crops in the metaverse images of farmland crops. After the configuration is complete, run the preset metaverse image model. If the preset metaverse image model runs normally after the configuration is complete, retrieve the operation data of the preset metaverse image model after the configuration is complete. Match the operation data of the preset metaverse image model after the configuration is complete with the crop production equipment to obtain the operation data of the metaverse crop production equipment. Substitute the operation data of the metaverse crop production equipment into the preset artificial intelligence robot scheduling model to generate an artificial intelligence robot scheduling command. Collect the image data after the robot executes the artificial intelligence robot scheduling command, and then substitute the image data after the robot executes the artificial intelligence robot scheduling command into the preset metaverse image model after the configuration is complete. If the preset metaverse image model runs abnormally after the configuration is complete, generate a warning message for the error in the execution of the scheduling command.

[0077] Specifically, for the robot remote cooperative scheduling system based on artificial intelligence of the present invention, the first data processing module is further used for: Receiving the standard parameters for the use of farmland environment data, where the standard parameters for the use of farmland environment data include soil environment parameters, meteorological environment parameters, and water quality environment parameters; From the evaluation results of the preprocessed farmland environment data, compare each piece of data in the evaluation results of the preprocessed farmland environment data with the standard parameters for the use of farmland environment data, and only retain the data entries that meet all the standard parameters for the use of farmland environment data as the valid farmland environment data; For each piece of valid farmland environment data, according to the sensor identity information associated with each piece of valid farmland environment data, extract the data features related to the sensor identity information. The data features related to the sensor identity information include the average value, maximum value, and minimum value within a period. Combine the data features related to the sensor identity information into a sensor data group, and each sensor data group corresponds to a sensor and the valid data features collected by the sensor.

[0078] Specifically, for the robot remote cooperative scheduling system based on artificial intelligence of the present invention, the second data processing module is further used for: Identifying the error data existing in the sensor warning data and locating the position of the error data in the warning data; Retrieve the sensor identity information corresponding to the error data according to the positioning information of the error data. The sensor identity information includes the identifier, location, and type of the sensor; Collect the time information when the error data occurs. The time information includes the time stamps of year, month, day, hour, and minute; Substitute the time information and the sensor identity information into a preset crop growth data simulation model, run the preset crop growth data simulation model, and calculate the crop growth state at the time when the error data occurs according to the input time information and sensor identity information; The preset crop growth data simulation model outputs the simulation data corresponding to the error data, and verifies the generated simulation data.

[0079] Specifically, for the robot remote cooperative scheduling system based on artificial intelligence of the present invention, the image generation module is further configured to: Substitute the farmland environment data into a preset crop growth image model. The preset crop growth image model calculates the expected growth state of the crop according to the input farmland environment data; Use image processing technology to convert the calculation result into a visual expected image of the crop; Verify the generated expected image of the farmland crop. If there are abnormalities or unreasonable points in the expected image of the farmland crop, return to check data replacement and model substitution.

[0080] Aiming at the problems that the crop planting environment is complex and changeable, and the data collected by sensors has noise and uncertainty, which affect the analysis and decision-making accuracy of data processing algorithms, the technical solution of the present invention takes the following measures to solve: Perform filtering processing on the original data collected by the sensor to eliminate noise and interference. Common filtering methods include low-pass filtering, high-pass filtering, band-pass filtering, etc. Select a suitable filter according to the characteristics of the sensor signal, filter out the high-frequency components of the clutter, and only leave the low-frequency effective signals.

[0081] Use time series anomaly detection methods, such as distance-based anomaly detection models and prediction model-based anomaly detection methods, to identify and process the outliers in the sensor data. These methods detect the outliers in the data by calculating the distance of the sample points or using the historical data within the window to establish a prediction model, and perform corresponding processing, such as deletion, replacement, or correction.

[0082] Perform comprehensive cleaning and verification on the sensor data to remove the missing values, duplicate values, and error values in the data. At the same time, set reasonable thresholds and data ranges to verify the sensor data to ensure the accuracy and reliability of the data.

[0083] Multi-sensor data fusion: Fuse data from multiple sensors to improve the accuracy and reliability of the data. Common data fusion methods include weighted average, Kalman filtering, etc. By fusing data from multiple sensors, the impact of abnormal data from a single sensor on the overall analysis results can be reduced.

[0084] Feature extraction: Extract features related to crop growth from the original data, such as the average value, maximum value, minimum value, etc. within a period. These features can reflect the environmental conditions and change trends of crop growth, providing a basis for subsequent analysis and decision-making.

[0085] Model training: Select a data processing model suitable for analyzing the crop growth environment, such as a deep learning model, a machine learning model, etc. Use a large amount of historical data and real-time monitoring data to train the model to improve the accuracy and generalization ability of the model.

[0086] Algorithm improvement: Improve the data processing algorithm for the problems of noise and uncertainty in sensor data. For example, adopt a more robust algorithm to reduce the impact of noise on the analysis results; or introduce uncertainty processing methods, such as Bayesian methods, fuzzy logic, etc., to handle the uncertainty in the data.

[0087] Real-time monitoring: Use sensors to collect farmland environment data in real-time, and conduct real-time monitoring and analysis on the data. Once data anomalies or excessive noise are found, take immediate measures to handle them.

[0088] Dynamic adjustment: According to the real-time monitoring results and analysis results, dynamically adjust the parameters and algorithms of the data processing model to adapt to the changes in the crop planting environment. At the same time, calibrate and maintain the sensors regularly to ensure the accuracy and reliability of the data.

[0089] In summary, the technical solution of the present invention effectively solves problems such as the complex and changeable crop planting environment and the noise and uncertainty in the data collected by sensors through measures such as data preprocessing and noise filtering, data fusion and feature extraction, model optimization and algorithm improvement, and real-time monitoring and dynamic adjustment, improving the accuracy of analysis and decision-making of the data processing algorithm.

Claims

1. A robot remote collaborative scheduling method based on artificial intelligence, characterized in that: include: Step S101, real-time collection of farmland environmental data, the farmland environmental data including farmland soil moisture, farmland soil temperature and farmland soil nutrient content, pre-processing the collected farmland environmental data to obtain pre-processed farmland environmental data, extracting data features from the pre-processed farmland environmental data, and then weighted evaluation of the extracted farmland environmental data features to obtain farmland environmental data evaluation results; Step S102: receiving farmland environmental data using standard parameters, screening out effective farmland environmental data from the farmland environmental data evaluation results based on the farmland environmental data using standard parameters, performing feature extraction on the effective farmland environmental data based on preset sensor identity information to obtain a sensor data group, classifying the data of the sensor data group according to a preset time period to obtain sensor time segment data, substituting the sensor time segment data into a preset early warning model to obtain sensor early warning data; Step S103: if there is error data in the sensor warning data, retrieve the sensor identity information corresponding to the error data in the sensor warning data, collect the time information of the sensor identity information corresponding to the error data in the sensor warning data, substitute the time information of the sensor identity information corresponding to the error data in the sensor warning data into the preset crop growth data simulation model, and obtain simulation data corresponding to the error data; Step S104: replacing the farmland environment data collected in real time with the simulation data corresponding to the erroneous data to obtain modified farmland environment data, collecting real-time farmland crop image data, substituting the modified farmland environment data into a preset growth image model to generate an expected image of the farmland crops; Step S105: Hash the expected image of farmland crops to obtain the hash value of the expected image of farmland crops, substitute the expected image of farmland crops and the hash value of the expected image of farmland crops into the preset Metaverse image model to obtain the Metaverse image of farmland crops, hash the farmland environment data collected in real time to obtain the hash value of the farmland environment data collected in real time, substitute the farmland environment data collected in real time and the hash value of the farmland environment data collected in real time into the preset Metaverse image model to obtain the Metaverse environment data of farmland crops, configure the Metaverse environment data of farmland crops in the Metaverse image of farmland crops, run the preset Metaverse image model after the configuration is complete, if the preset Metaverse image model is run after the configuration is complete If the model runs normally, the running data of the preset Metaverse image model after the configuration is complete is retrieved, and the running data of the preset Metaverse image model after the configuration is complete is matched with the crop production equipment to obtain the running data of the Metaverse crop production equipment, and the running data of the Metaverse crop production equipment is substituted into the preset artificial intelligence robot scheduling model to generate the artificial intelligence robot scheduling command, and the image data after the robot executes the artificial intelligence robot scheduling command is collected, and then the image data after the robot executes the artificial intelligence robot scheduling command is substituted into the preset Metaverse image model after the configuration is complete. If the running of the preset Metaverse image model after the configuration is complete runs abnormally, a scheduling command execution error warning information is generated.

2. The method for remote collaborative scheduling of robots based on artificial intelligence according to claim 1, characterized in that: The step S102 includes: Receive farmland environmental data using standard parameters, which include soil environmental parameters, meteorological environmental parameters, and water quality environmental parameters; From the pre-processed farmland environment data assessment results, compare each data in the pre-processed farmland environment data assessment results with the farmland environment data use standard parameters, and only retain the data entries that meet all the farmland environment data use standard parameters as valid farmland environment data; For each valid farmland environmental data, data features related to the sensor identity information are extracted according to the sensor identity information associated with each valid farmland environmental data. The data features related to the sensor identity information include the average value, maximum value, and minimum value within a time period. The data features related to the sensor identity information are combined into a sensor data group. Each sensor data group corresponds to a sensor and the valid data features collected by the sensor.

3. The method for remote collaborative scheduling of robots based on artificial intelligence according to claim 1, characterized in that: The step S103 includes: In the sensor warning data, identify the existing erroneous data and locate the position of the erroneous data in the warning data; According to the location information of the error data, the sensor identity information corresponding to the error data is retrieved, where the sensor identity information includes the identification, location and type of the sensor; Collect the time information when the error data occurs, including the timestamp of year, month, day, hour and minute; Substituting the time information and the sensor identity information into a preset crop growth data simulation model, running the preset crop growth data simulation model, and calculating the crop growth state when the error data occurs according to the input time information and the sensor identity information; The preset crop growth data simulation model outputs simulation data corresponding to the erroneous data, and verifies the generated simulation data.

4. The method for remote collaborative scheduling of robots based on artificial intelligence according to claim 1, characterized in that: The step S104 includes: Substituting the farmland environment data into a preset crop growth image model, the preset crop growth image model calculates the expected growth state of the crop based on the input farmland environment data; Using image processing technology, the calculation results are converted into visual images of expected crops; The generated expected image of farmland crops is verified. If there are any abnormalities or unreasonableness in the expected image of farmland crops, return to check the data replacement and model substitution steps.

5. A robot remote collaborative scheduling system based on artificial intelligence, applied to the robot remote collaborative scheduling method based on artificial intelligence as claimed in any one of claims 1 to 4, characterized in that: include: The data acquisition module is used to collect farmland environmental data in real time. The farmland environmental data includes farmland soil moisture, farmland soil temperature and farmland soil nutrient content. The collected farmland environmental data is preprocessed to obtain preprocessed farmland environmental data, data features are extracted from the preprocessed farmland environmental data, and then weighted evaluation is performed on the extracted farmland environmental data features to obtain farmland environmental data evaluation results. The first data processing module is used to receive the farmland environment data usage standard parameters, screen out effective farmland environment data from the farmland environment data evaluation results based on the farmland environment data usage standard parameters, extract features of the effective farmland environment data based on preset sensor identity information to obtain a sensor data group, classify the data of the sensor data group according to a preset time period to obtain sensor time segment data, substitute the sensor time segment data into a preset early warning model to obtain sensor early warning data; The second data processing module is used for retrieving the sensor identity information corresponding to the error data in the sensor warning data if there is error data in the sensor warning data, collecting the time information of the sensor identity information corresponding to the error data in the sensor warning data, substituting the time information of the sensor identity information corresponding to the error data in the sensor warning data into a preset crop growth data simulation model, and obtaining simulation data corresponding to the error data; Image generation module: used to replace the real-time collected farmland environment data with the simulated data corresponding to the erroneous data, obtain the modified farmland environment data, collect real-time farmland crop image data, substitute the modified farmland environment data into the preset growth image model, and generate the expected image of the farmland crop; Data analysis module: used to hash the expected image of farmland crops to obtain the hash value of the expected image of farmland crops, substitute the expected image of farmland crops and the hash value of the expected image of farmland crops into the preset metaverse image model to obtain the metaverse image of farmland crops, hash the farmland environment data collected in real time to obtain the hash value of the farmland environment data collected in real time, substitute the farmland environment data collected in real time and the hash value of the farmland environment data collected in real time into the preset metaverse image model to obtain the metaverse environment data of farmland crops, configure the metaverse environment data of farmland crops in the metaverse image of farmland crops, run the preset metaverse image model after the configuration is complete, and run the preset metaverse image after the configuration is complete. If the model runs normally, the running data of the preset Metaverse image model after the configuration is complete is retrieved, and the running data of the preset Metaverse image model after the configuration is complete is matched with the crop production equipment to obtain the running data of the Metaverse crop production equipment, and the running data of the Metaverse crop production equipment is substituted into the preset artificial intelligence robot scheduling model to generate the artificial intelligence robot scheduling command, and the image data after the robot executes the artificial intelligence robot scheduling command is collected, and then the image data after the robot executes the artificial intelligence robot scheduling command is substituted into the preset Metaverse image model after the configuration is complete. If the running of the preset Metaverse image model after the configuration is complete runs abnormally, a scheduling command execution error warning information is generated.

6. The artificial intelligence-based robot remote collaborative scheduling system according to claim 5 is characterized in that: The first data processing module is further used for: Receive farmland environmental data using standard parameters, which include soil environmental parameters, meteorological environmental parameters, and water quality environmental parameters; From the pre-processed farmland environment data assessment results, compare each data in the pre-processed farmland environment data assessment results with the farmland environment data use standard parameters, and only retain the data entries that meet all the farmland environment data use standard parameters as valid farmland environment data; For each valid farmland environmental data, data features related to the sensor identity information are extracted according to the sensor identity information associated with each valid farmland environmental data. The data features related to the sensor identity information include the average value, maximum value, and minimum value within a time period. The data features related to the sensor identity information are combined into a sensor data group. Each sensor data group corresponds to a sensor and the valid data features collected by the sensor.

7. The robot remote collaborative scheduling system based on artificial intelligence according to claim 5 is characterized in that: The second data processing module is further used for: In the sensor warning data, identify the existing erroneous data and locate the position of the erroneous data in the warning data; According to the location information of the error data, the sensor identity information corresponding to the error data is retrieved, where the sensor identity information includes the identification, location and type of the sensor; Collect the time information when the error data occurs, including the timestamp of year, month, day, hour and minute; Substituting the time information and the sensor identity information into a preset crop growth data simulation model, running the preset crop growth data simulation model, and calculating the crop growth state when the error data occurs according to the input time information and the sensor identity information; The preset crop growth data simulation model outputs simulation data corresponding to the erroneous data, and verifies the generated simulation data.

8. The artificial intelligence-based robot remote collaborative scheduling system according to claim 5 is characterized in that: The image generation module is further used for: Substituting the farmland environment data into a preset crop growth image model, the preset crop growth image model calculates the expected growth state of the crop based on the input farmland environment data; Using image processing technology, the calculation results are converted into visual images of expected crops; The generated expected image of farmland crops is verified. If there are any abnormalities or unreasonableness in the expected image of farmland crops, return to check data replacement and model substitution.

Images