A method and system for establishing a location based on data transmitted by an unmanned aerial vehicle

Abstract

The present application provides a method and system for establishing a location based on drone transmission data. Among them, the target transmission data of the actual system is collected by drone; based on the collected transmission data corresponding to all areas in the actual system, a digital twin model of the virtual system is established, and the target transmission data is used as a model input parameter and input into the digital twin model to predict the location information of the target area; the digital twin model is verified, and after the digital twin model is verified, the location information of the target area is output based on the digital twin model; the target area is established in the actual system according to the location information of the target area. The technical solution provided by the present application can improve the accuracy of the established digital twin model and the accuracy of the target area location establishment.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of data processing technology, and in particular to a method and system for establishing a location based on data transmitted by an unmanned aerial vehicle. Background Art

[0002] Digital twin is a technology that maps physical systems or processes in the real world into a virtual environment through digital means. It builds a virtual "twin" corresponding to the actual system based on the data, models and algorithms of the physical system, and realizes the simulation, monitoring and prediction of the actual system. Digital twin technology can help realize functions such as real-time monitoring, predictive analysis, and intelligent decision-making, and is widely used in industries such as industry, aerospace, and medical care.

[0003] At present, the common scheme for establishing the location of the target area for drone transmission data includes using GPS (Global Positioning System) or other positioning technologies to obtain the location information of the drone and collect data of the target area through sensors. Then, this data can be combined with map data or previously established map models (i.e. digital twin models) and further analyzed and processed by algorithms to determine the location information of the target area.

[0004] However, this method may be affected by factors such as environmental interference, inaccurate data or sensor accuracy limitations, resulting in poor storage accuracy of the established map model, low accuracy or errors in the location of the target area. Summary of the invention

[0005] The embodiments of the present application provide a method and system for establishing a location based on data transmitted by a drone, so as to solve the problems of poor precision of the digital twin model established in the prior art and low accuracy in establishing the location of the target area.

[0006] In a first aspect, an embodiment of the present application provides a method for establishing a location based on data transmitted by a drone, comprising:

[0007] Collecting target transmission data of the actual system by using a drone, the target transmission data includes but is not limited to the size, shape, material, and structure of the target area;

[0008] Based on the collected transmission data corresponding to all areas in the actual system, a digital twin model of the virtual system is established, and the target transmission data is used as a model input parameter and input into the digital twin model to predict the location information of the target area;

[0009] Verifying the digital twin model, and after the digital twin model is verified, outputting the location information of the target area based on the digital twin model;

[0010] A target area is established in the actual system according to the location information of the target area.

[0011] Optionally, based on the collected transmission data corresponding to all areas in the actual system, a digital twin model of the virtual system is established, including:

[0012] Obtaining transmission data corresponding to all regions in the actual system, the transmission data including but not limited to size, shape, material properties, and structural information of all regions in the actual system;

[0013] A physical model is established based on the transmission data corresponding to all areas in the actual system, and the physical model is converted into a digital model using numerical methods or computer-aided engineering (CAE) software. Then, a digital twin model of the virtual system is established based on the physical model and the digital model of the actual system.

[0014] Optionally, before verifying the digital twin model, the method further includes:

[0015] The transmission data corresponding to all areas in the actual system collected are calibrated with time nodes and organized into original time series data sets in chronological order;

[0016] Each sequence in the original time series dataset is subjected to data enhancement processing such as random rotation, translation and scaling to obtain a series of new sequences to construct the target time series dataset;

[0017] For each new sequence, divide it into multiple subsequences according to the time step;

[0018] Each subsequence is taken as a sample and added to the target time series dataset until new sequences and subsequences that meet the conditions are obtained to complete the construction of the target time series dataset.

[0019] Optionally, verifying the digital twin model includes:

[0020] The target time series dataset is validated to evaluate the comprehensive score of the digital twin model.

[0021] Optionally, the validating the target time series data set to evaluate the comprehensive score of the digital twin model includes:

[0022] Divide the target time series data set into several subsequences, each subsequence is used as a validation set, and the remaining subsequences are used as training sets;

[0023] By formula: , calculate the mean absolute scaled error based on the historical errors of the training set;

[0024] Among them, h represents the prediction step length, Indicates the actual value, represents the predicted value, T represents the index of the last training sample, represents the lag period;

[0025] By formula: , calculating the symmetric mean absolute percentage error based on the historical errors of the training set;

[0026] Among them, h represents the prediction step length, Indicates the actual value, represents the predicted value, and T represents the index of the last training sample;

[0027] By formula: , dividing the training set into different categories and determining the balanced accuracy of the number of samples in different time series classifications;

[0028] in, represents the number of categories, Indicates category The number of correctly classified samples in Indicates category The number of misclassified samples in Indicates non-category The number of correctly classified samples in Indicates non-category The number of misclassified samples in ;

[0029] By formula: , dividing the training set into different categories and determining the influence of the number of samples in random classification;

[0030] in, represents the classification accuracy of the observation, Represents the classification accuracy of random classification;

[0031] By formula group: , evaluating the comprehensive score of the digital twin model according to the mean absolute scaled error, the symmetric mean absolute percentage error, the balanced accuracy of the number of samples in the different time series classifications, and the influence of the number of samples in the random classification;

[0032] in, is the regression coefficient, is the error term, is the evaluation score, x1 is the mean absolute scaled error, x2 is the symmetric mean absolute percentage error, x3 is the balanced accuracy of the number of samples in the different time series classifications, and x4 is the influence of the number of samples in the random classification.

[0033] Optionally, each sequence in the original time series data set is subjected to data enhancement processing such as random rotation, translation and scaling to obtain a series of new sequences, including:

[0034] For each sample, a rotation angle is randomly generated, and each sequence in the original time series data set is rotated along the center point by this angle to obtain a new sequence; the rotation angle can be randomly generated within a certain range, and the rotated sample can be expressed as:

[0035] ;

[0036] in, represents the coordinates of the sample before rotation, represents the coordinates of the sample after rotation, Represents a randomly generated rotation angle;

[0037] For each sample, a translation vector is randomly generated, and each sequence in the original time series data set is translated along the vector to obtain a new sequence; the translation vector can be randomly generated within a certain range, and the translated sample can be expressed as:

[0038] ;

[0039] in, represents the coordinates of the sample before translation, represents the coordinates of the sample after translation, represents a randomly generated translation vector;

[0040] For each sample, a scaling factor is randomly generated, and each sequence in the original time series data set is scaled along the center point to obtain a new sequence; the scaling factor can be randomly generated within a preset range, and the scaled sample can be expressed as:

[0041] ;

[0042] in, represents the coordinates of the sample before scaling, represents the coordinates of the sample after scaling, Represents a randomly generated scaling factor.

[0043] Optionally, it also includes:

[0044] A real-time monitoring device is set up to obtain the actual location information of the UAV in a timely manner, and compare it with the prediction results of the digital twin model to adjust the parameters and algorithms of the digital twin model. The process of adjusting the parameters and algorithms of the digital twin model at least includes: adjusting model hyperparameters, increasing the sample size, and processing sample outliers.

[0045] In a second aspect, an embodiment of the present application provides a location establishment system based on drone transmission data, including:

[0046] A collection module, used to collect target transmission data of the actual system through the drone, wherein the target transmission data includes but is not limited to the size, shape, material, and structure of the target area;

[0047] An input module is used to establish a digital twin model of the virtual system based on the collected transmission data corresponding to all areas in the actual system, and input the target transmission data as a model input parameter into the digital twin model to predict the location information of the target area;

[0048] A verification output module, used to verify the digital twin model, and after the digital twin model is verified, output the location information of the target area based on the digital twin model;

[0049] The establishing module is used to establish the target area in the actual system according to the location information of the target area.

[0050] In a third aspect, an embodiment of the present application provides a computing device, comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement the location establishment method based on drone transmission data as described in the first aspect above.

[0051] In a fourth aspect, an embodiment of the present application provides a computer storage medium, characterized in that a computer program is stored therein, and when the computer program is executed by a computer, the location establishment method based on drone transmission data as described in the first aspect above is implemented.

[0052] In an embodiment of the present application, target transmission data of an actual system is collected by a drone; a digital twin model of a virtual system is established based on the collected transmission data corresponding to all areas in the actual system, and the target transmission data is used as a model input parameter and input into the digital twin model to predict the location information of the target area; the digital twin model is verified, and after the digital twin model is verified, the location information of the target area is output based on the digital twin model; and the target area is established in the actual system according to the location information of the target area.

[0053] This method and system for establishing location intelligence based on drone-transmitted data has many beneficial effects:

[0054] Real-time monitoring and feedback: By collecting data from the actual system in real time through drones, the system can monitor the status and changes of the target area in real time. This real-time feedback can help to detect problems or changes in time and take corresponding measures.

[0055] Digital twin model prediction: Using the established digital twin model, combined with real-time transmission data, the location information of the target area can be predicted. This prediction performance helps optimize system operation and resource allocation, improving overall efficiency and safety.

[0056] Reduce labor costs and risks: Compared with traditional manual inspection or monitoring methods, using drones for data collection and system modeling can greatly reduce manpower input and operational risks. Drones can work in complex or dangerous environments, avoiding the safety risks faced by personnel.

[0057] Accurate positioning and deployment: The location information output by the digital twin model can more accurately locate and deploy the target area in the system. This helps to optimize the system layout and resource allocation, and improve the overall performance and maintainability of the system.

[0058] Data-driven decision-making: Based on real-time data and the prediction results of the digital twin model, decision makers can make more data-driven decisions. This data-based decision-making can improve the system's response speed and intelligence level, making the system more competitive and sustainable.

[0059] In general, the location intelligence establishment method and system based on drone transmitted data combines data collection, model prediction and practical application, bringing beneficial effects such as real-time monitoring, cost savings, efficiency improvement, etc., and promotes the application and development of intelligence and digitalization in actual systems.

[0060] These and other aspects of the present application will become more clearly understood in the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0062] Figure 1 A flowchart of an embodiment of a method for establishing a location based on data transmitted by a drone provided in an embodiment of the present application;

[0063] Figure 2 A schematic diagram of a system for establishing a location based on data transmitted by a drone provided in an embodiment of the present application;

[0064] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0065] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0066] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0067] The technical solutions of the embodiments of the present application have broad application prospects in various practical scenarios, especially in the following aspects:

[0068] Environmental monitoring and disaster warning: Drones carry sensors and fly over forests, mountains, rivers and other areas to collect a large amount of environmental data in real time, such as weather, topography, water quality, etc. By combining digital twin models with real-time data, natural disasters such as geological disasters and floods can be predicted, and early warnings and responses can be given to protect people’s lives and property.

[0069] Agricultural management and precision agriculture: The data transmitted by drones combined with the digital twin model can achieve accurate monitoring and management of farmland, including crop growth, soil moisture, pests and diseases, etc. Farmers can accurately apply fertilizers and irrigation based on the predicted information to improve crop yields and quality.

[0070] Urban planning and smart city construction: Drones can fly over every corner of the city to collect data on urban infrastructure, population density, traffic flow, etc., and combine digital twin models for urban planning and smart city construction. By predicting urban development trends and needs, optimizing urban layout and traffic organization, etc., the quality of life and operational efficiency of the city can be improved.

[0071] Logistics and traffic management: Use drones for real-time monitoring and data collection, and combine digital twin models to predict traffic congestion, cargo transportation routes, etc., optimize logistics distribution plans, improve transportation efficiency, and reduce energy consumption and environmental pollution.

[0072] Resource exploration and mining: Through the transmission of data and digital twin models by drones, mining areas, oil fields and other resources can be efficiently explored and managed. Resource distribution and mining difficulty can be predicted, and reasonable development plans can be formulated to improve resource utilization efficiency and economic benefits.

[0073] In summary, the location establishment method and system based on drone transmission data have broad application prospects in environmental monitoring, agricultural management, urban planning, logistics management, resource exploration and other fields. It can provide data support and intelligent decision-making for various industries and promote the sustainable development of social economy.

[0074] The inventors have found that, at present, the common scheme for establishing the location of the target area for the data transmitted by the drone includes using GPS (Global Positioning System) or other positioning technologies to obtain the location information of the drone and collecting the data of the target area through sensors. Then, these data can be combined with map data or a previously established map model (i.e., a digital twin model) and further analyzed and processed by algorithms to determine the location information of the target area.

[0075] However, this method may be affected by factors such as environmental interference, inaccurate data or sensor accuracy limitations, resulting in poor storage accuracy of the established map model, low accuracy or errors in the location of the target area.

[0076] In view of this, an embodiment of the present application provides a method for establishing a position based on drone transmission data, which collects target transmission data of an actual system through a drone; establishes a digital twin model of a virtual system based on the collected transmission data corresponding to all areas in the actual system, and inputs the target transmission data into the digital twin model as a model input parameter to predict the position information of the target area; verifies the digital twin model, and after the digital twin model is verified, outputs the position information of the target area based on the digital twin model; and establishes the target area in the actual system according to the position information of the target area.

[0077] The technical solution provided in this application can improve the accuracy of the established digital twin model and the accuracy of the target area location establishment

[0078] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0079] Figure 1 The present application provides a flowchart of an embodiment of a method for establishing a location based on data transmitted by a drone, such as Figure 1 As shown, the method includes:

[0080] 101. Collect target transmission data of the actual system through a drone, wherein the target transmission data includes but is not limited to the size, shape, material, and structure of the target area;

[0081] In this step, in actual operation, drones can be equipped with various sensors, such as high-definition cameras, infrared cameras, radars, etc., to collect various data of the target area. For example, in the agricultural field, drones can collect data such as soil moisture, crop growth, and pests and diseases. In the field of urban planning, drones can collect data such as the location of urban infrastructure, population density, and traffic flow.

[0082] 102. Based on the collected transmission data corresponding to all areas in the actual system, a digital twin model of the virtual system is established, and the target transmission data is used as a model input parameter and input into the digital twin model to predict the location information of the target area;

[0083] In this step, mathematical models and algorithms can be used to build a digital twin model based on the actual system data collected. The digital twin model can contain information about the structure, movement laws, physical properties, etc. of the actual system. For example, in the agricultural field, the digital twin model can contain information about the physiological characteristics of crop growth and the impact of environmental factors on crop growth. In the field of urban planning, the digital twin model can contain various parameters of urban planning, such as population density, traffic flow, infrastructure, etc.

[0084] 103. Verify the digital twin model, and after the digital twin model is verified, output the location information of the target area based on the digital twin model;

[0085] In this step, in order to ensure the accuracy of the digital twin model, it needs to be verified. The verification process can use the comparison between experimental data and the prediction results of the digital twin model to evaluate the accuracy of the digital twin model. For example, in the agricultural field, the accuracy of the model can be evaluated by collecting data from actual farmland and comparing it with the prediction results of the digital twin model. After verification, the digital twin model can output the location information of the target area.

[0086] 104. Establish a target area in the actual system according to the location information of the target area.

[0087] In this step, after the digital twin model outputs the location information of the target area, the target area can be established in the actual system based on this information. For example, in the agricultural field, the target area can be marked in the farmland and accurately managed and fertilized. In the field of urban planning, urban planning and traffic organization can be carried out based on the prediction results of the digital twin model to optimize the urban layout and traffic operation efficiency.

[0088] For example, for the agricultural field, the embodiments may include: collecting data such as soil moisture, crop growth, and pest and disease conditions of farmland through drones; establishing a digital twin model based on the collected farmland data, and inputting the target transmission data as a model input parameter into the digital twin model to predict the location information of the target area; verifying the digital twin model, for example, collecting data from actual farmland, comparing it with the prediction results of the digital twin model, and evaluating the accuracy of the digital twin model; after the digital twin model is verified, outputting the location information of the target area; establishing a target area in the farmland based on the location information of the target area, for example, marking the target area in the farmland, and performing precise management and fertilization.

[0089] Optionally, in the embodiment of the present application, step 102 may specifically include:

[0090] 1021. Acquire transmission data corresponding to all regions in the actual system, where the transmission data includes but is not limited to size, shape, material properties, and structural information of all regions in the actual system;

[0091] In this step, the transmission data of all areas in the actual system are collected by means of drones, etc. For example, in the field of urban planning, drones can be used to photograph various buildings, roads, parks, etc. in the city to obtain transmission data such as the size, shape, material properties, and structural information of these areas.

[0092] 1022. Establish a physical model based on the transmission data corresponding to all areas in the actual system, and use numerical methods or computer-aided engineering (CAE) software to convert the physical model into a digital model, and then establish a digital twin model of the virtual system based on the physical model and digital model of the actual system.

[0093] In this step, a physical model can be established using the collected transmission data of the actual system. The physical model can contain information on the structure, motion laws, physical properties, etc. of the actual system. The physical model is converted into a digital model using numerical methods or CAE software, and the digital model can be calculated and simulated more conveniently and accurately. Then, based on the physical model and digital model of the actual system, a digital twin model of the virtual system is established. The digital twin model can contain various parameters and properties of the actual system, such as information on structure, material, motion laws, physical properties, etc.

[0094] For example, in the field of urban planning, embodiments may include: obtaining transmission data corresponding to all areas in the city, including the size, shape, material properties, structural information, etc. of the buildings; establishing a physical model based on the collected urban data, and using CAE software to convert the physical model into a digital model, and then establishing a digital twin model of the virtual system based on the physical model and digital model of the actual system; using the target transmission data as a model input parameter and inputting it into the digital twin model to predict the location information of the target area.

[0095] Optionally, in an embodiment of the present application, before step 103, it also includes: calibrating the transmission data corresponding to all areas in the collected actual system for time nodes, and organizing them into an original time series data set in chronological order; performing data enhancement processing such as random rotation, translation and scaling on each sequence in the original time series data set to obtain a series of new sequences to construct a target time series data set; for each new sequence, dividing it into multiple subsequences according to the time step; taking each subsequence as a sample and adding it to the target time series data set, repeating the above steps until enough new sequences and subsequences are obtained to complete the construction of the target time series data set.

[0096] Specifically, the process of "for each new sequence, dividing it into multiple subsequences according to the time step" may include: randomly selecting a starting time point in each new sequence, and dividing the sequence into multiple adjacent subsequences according to the time step. For example, if the time step is 1 hour and each subsequence contains 10 time points, then each new sequence can be divided into multiple subsequences with a length of 10 hours.

[0097] This embodiment mainly establishes a digital twin model for time series data, and improves the generalization ability and robustness of the model through technologies such as data enhancement and subsequence division.

[0098] The specific steps are as follows:

[0099] The transmission data corresponding to all areas in the actual system are collected and calibrated with time nodes, and organized into original time series data sets in chronological order.

[0100] The collected data is organized into original time series data sets in chronological order for subsequent processing. For example, in the field of urban planning, data such as the city’s population density and traffic flow can be collected and calibrated according to hourly or daily time nodes to organize them into original time series data sets.

[0101] Each sequence in the original time series dataset is subjected to data enhancement processing such as random rotation, translation and scaling to obtain a series of new sequences to construct the target time series dataset.

[0102] Perform data augmentation processing such as random rotation, translation, and scaling on each sequence in the original time series dataset to expand the dataset. For example, in the field of urban planning, data augmentation processing can be performed on the time series data of urban traffic flow, and random rotation, translation, and scaling can be performed at each time point to obtain a series of new sequences.

[0103] For each new sequence, divide it into multiple subsequences according to the time step.

[0104] For each new sequence, divide it into multiple subsequences according to the time step. For example, if the time step is 1 hour and each subsequence contains 10 time points, then each new sequence can be divided into multiple subsequences of length 10 hours.

[0105] Take each subsequence as a sample and add it to the target time series dataset, repeating the above steps until enough new sequences and subsequences are obtained to complete the construction of the target time series dataset.

[0106] Take each subsequence as a sample and add it to the target time series dataset. Repeat the above steps until enough new sequences and subsequences are obtained to complete the construction of the target time series dataset. This process can be automated to improve efficiency and accuracy.

[0107] For example, in the field of urban planning, embodiments may include: for the time series data of urban traffic flow, dividing it into multiple subsequences according to the time step, and taking each subsequence as a sample, repeating the above process to construct a target time series data set; using the target time series data set to train a deep neural network model to achieve the prediction of the location of the target area; combining the trained deep neural network model with the digital twin model to construct a location intelligence establishment system based on drone transmission data.

[0108] The specific steps are as follows:

[0109] The target time series dataset is used to train a deep neural network model to predict the location of the target area.

[0110] The deep neural network model is trained using the target time series data set. The deep neural network model can be various types of neural networks, such as convolutional neural networks, recurrent neural networks, etc. Through training, the deep neural network model can learn the characteristics of the time series data and associate it with the target area location to achieve the prediction of the target area location.

[0111] For example, in the field of urban planning, the target time series dataset can be used to train a recurrent neural network model to predict the urban traffic flow at a certain point in the future, and combine it with the geographic information of the urban road network to predict urban traffic congestion.

[0112] The trained deep neural network model is combined with the digital twin model to build a location intelligence system based on drone transmitted data.

[0113] The trained deep neural network model is combined with the digital twin model to build a location intelligence system based on drone transmission data. The digital twin model can provide empirical data and verification data for the deep neural network model to optimize the model's prediction effect. The drone transmission data can provide real-time, high-quality data support for the model to ensure the timeliness and accuracy of the model.

[0114] For example, in the field of urban planning, a city traffic congestion prediction system based on drone transmission data can be built. The system combines the time series data of urban traffic flow with the geographic information of the urban road network, uses a recurrent neural network model for prediction, and obtains real-time, high-quality traffic flow data through drone transmission data to continuously update and optimize the model's prediction effect.

[0115] Optionally, in the embodiment of the present application, the process of “performing data enhancement processing such as random rotation, translation and scaling on each sequence in the original time series data set to obtain a series of new sequences” may include:

[0116] For each sample, a rotation angle is randomly generated, and each sequence in the original time series data set is rotated along the center point by the angle to obtain a new sequence; the rotation angle can be randomly generated within a certain range, for example, a random angle is generated within the range of [-30,30], and the rotated sample can be expressed as:

[0117] ;

[0118] in, represents the coordinates of the sample before rotation, represents the coordinates of the sample after rotation, Represents a randomly generated rotation angle;

[0119] For each sample, a translation vector is randomly generated, and each sequence in the original time series data set is translated along the vector to obtain a new sequence; the translation vector can be randomly generated within a certain range, for example, a random vector is generated within the range of [-10,10], and the translated sample can be expressed as:

[0120] ;

[0121] in, represents the coordinates of the sample before translation, represents the coordinates of the sample after translation, represents a randomly generated translation vector;

[0122] For each sample, a scaling factor is randomly generated, and each sequence in the original time series data set is scaled along the center point to obtain a new sequence; the scaling factor can be randomly generated within a certain range, for example, a random factor is generated within the range of [0.8, 1.2], and the scaled sample can be expressed as:

[0123] ;

[0124] in, represents the coordinates of the sample before scaling, represents the coordinates of the sample after scaling, Represents a randomly generated scaling factor.

[0125] Optionally, in an embodiment of the present application, step 103 may specifically include: verifying the target time series data set to evaluate the comprehensive score of the digital twin model.

[0126] The specific steps include:

[0127] 1031. Divide the target time series data set into a number of subsequences, each subsequence as a validation set, and the remaining subsequences as training sets;

[0128] By formula: , calculate the mean absolute scaled error based on the historical errors of the training set;

[0129] Among them, h represents the prediction step length, Indicates the actual value, represents the predicted value, T represents the index of the last training sample, represents the lag period;

[0130] In this step, the target time series data set is divided into several subsequences, each of which is used as a validation set, and the remaining subsequences are used as training sets. Using the historical error of the training set, the mean absolute scaled error is calculated by the formula to evaluate the prediction ability of the digital twin model. The mean absolute scaled error reflects the error between the predicted value and the actual value of the digital twin model.

[0131] For example, in the field of urban planning, the time series data of urban traffic flow can be divided into several subsequences, each of which is used as a validation set, and the remaining subsequences are used as training sets. Using the historical error of the training set, the mean absolute scaled error is calculated through the formula to evaluate the digital twin model's ability to predict future traffic flow.

[0132] 1032. Through the formula: , calculating the symmetric mean absolute percentage error based on the historical errors of the training set;

[0133] Among them, h represents the prediction step length, Indicates the actual value, represents the predicted value, and T represents the index of the last training sample;

[0134] In this step, the symmetric mean absolute percentage error is calculated by the formula to evaluate the prediction ability of the digital twin model. The symmetric mean absolute percentage error reflects whether the prediction ability of the digital twin model is robust.

[0135] For example, in the field of urban planning, the symmetric mean absolute percentage error can be used to evaluate the digital twin model's ability to predict future traffic flow.

[0136] 1033. Through the formula: , dividing the training set into different categories and determining the balanced accuracy of the number of samples in different time series classifications;

[0137] in, represents the number of categories, Indicates category The number of correctly classified samples in Indicates category The number of misclassified samples in Indicates non-category The number of correctly classified samples in Indicates non-category The number of misclassified samples in ;

[0138] In this step, the training set is divided into different categories, and the balanced accuracy of the sample quantity is determined according to different time series classifications to evaluate the prediction ability of the digital twin model. The balanced accuracy of the sample quantity reflects whether the prediction ability of the digital twin model is stable.

[0139] For example, in the field of urban planning, the time series data of urban traffic flow can be divided into different categories, and the balanced accuracy of the sample quantity can be determined according to different time series classifications to evaluate the digital twin model's ability to predict future traffic flow.

[0140] 1034. Through the formula: , dividing the training set into different categories and determining the influence of the number of samples in random classification;

[0141] in, represents the classification accuracy of the observation, Represents the classification accuracy of random classification;

[0142] In this step, the training set is randomly classified, and the prediction ability of the digital twin model is evaluated based on the classification accuracy of the random classification. The classification accuracy of the random classification reflects whether the prediction ability of the digital twin model is robust.

[0143] For example, in the field of urban planning, the time series data of urban traffic flow can be randomly classified, and the digital twin model's ability to predict future traffic flow can be evaluated based on the classification accuracy of the random classification.

[0144] 1035. Through the formula group: , evaluating the comprehensive score of the digital twin model according to the mean absolute scaled error, the symmetric mean absolute percentage error, the balanced accuracy of the number of samples in the different time series classifications, and the influence of the number of samples in the random classification;

[0145] in, is the regression coefficient, is the error term, is the evaluation score, x1 is the mean absolute scaled error, x2 is the symmetric mean absolute percentage error, x3 is the balanced accuracy of the number of samples in the different time series classifications, and x4 is the influence of the number of samples in the random classification.

[0146] In this step, the prediction ability of the digital twin model is comprehensively evaluated through the formula group. The comprehensive evaluation method can be weighted average, hierarchical analysis, etc., to comprehensively consider the relationship between various evaluation indicators.

[0147] For example, in the field of urban planning, the weighted average method can be used to comprehensively evaluate evaluation indicators such as the mean absolute scaled error, the symmetric mean absolute percentage error, the balanced accuracy of the sample size, and the influence of random classification to evaluate the digital twin model's ability to predict future traffic flow.

[0148] Optionally, in an embodiment of the present application, the method also includes: setting a real-time monitoring device to promptly obtain the actual location information of the drone, and compare it with the prediction results of the digital twin model to adjust the parameters and algorithms of the digital twin model, including but not limited to adjusting model hyperparameters, increasing the sample size, and processing sample outliers.

[0149] In this embodiment, the method also includes setting up a real-time monitoring device to timely obtain the actual location information of the UAV and compare it with the prediction results of the digital twin model to adjust the parameters and algorithms of the digital twin model.

[0150] The specific steps are as follows: set up a real-time monitoring device to obtain the actual location information of the drone in a timely manner, and compare it with the prediction results of the digital twin model; according to the comparison results, adjust the parameters and algorithms of the digital twin model, including but not limited to adjusting the model hyperparameters, increasing the sample size, and processing sample outliers.

[0151] The specific steps are as follows: Set up a real-time monitoring device to obtain the actual location information of the drone in a timely manner and compare it with the prediction results of the digital twin model. Set up a real-time monitoring device to obtain the actual location information of the drone in a timely manner and compare it with the prediction results of the digital twin model. The real-time monitoring device can use a variety of sensors, such as GPS, inertial navigation system, etc. to obtain the actual location information of the drone. By comparing the actual location information of the drone with the prediction results of the digital twin model, the errors and deficiencies of the digital twin model can be discovered in time for adjustment and optimization.

[0152] For example, in the field of urban planning, real-time monitoring devices can be set up to use sensors such as GPS to obtain the real-time location information of the drone, and compare it with the prediction results of the digital twin model to discover errors and deficiencies in the digital twin model.

[0153] Based on the comparison results, adjust the parameters and algorithms of the digital twin model.

[0154] According to the comparison results, the parameters and algorithms of the digital twin model are adjusted. The adjustment method can be to modify the model's hyperparameters, increase the sample size, process sample outliers, etc., to improve the prediction ability and stability of the digital twin model.

[0155] For example, in the field of urban planning, based on the drone location information obtained by the real-time monitoring device, it can be found that the prediction error of the digital twin model for certain areas is large. At this time, the prediction ability of the digital twin model can be optimized by increasing the number of training samples in the area or processing abnormal samples.

[0156] Figure 2 A schematic diagram of a location establishment system based on drone transmission data provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the system includes:

[0157] A collection module 21 is used to collect target transmission data of the actual system through a drone, wherein the target transmission data includes but is not limited to the size, shape, material, and structure of the target area;

[0158] An input module 22 is used to establish a digital twin model of the virtual system based on the collected transmission data corresponding to all areas in the actual system, and input the target transmission data as a model input parameter into the digital twin model to predict the location information of the target area;

[0159] A verification output module 23, used to verify the digital twin model, and after the digital twin model is verified, output the location information of the target area based on the digital twin model;

[0160] The establishing module 24 is used to establish the target area in the actual system according to the location information of the target area.

[0161] Figure 2 The location establishment system based on drone transmission data can be performed Figure 1 The implementation principle and technical effect of the location establishment method based on drone transmission data described in the illustrated embodiment will not be described in detail. The specific manner in which each module and unit performs operations in the location establishment system based on drone transmission data in the above embodiment has been described in detail in the embodiment of the method, and will not be elaborated here.

[0162] In one possible design, Figure 2 The location establishment system based on drone transmission data of the illustrated embodiment can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0163] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0164] The processing component 32 is used to: collect target transmission data of the actual system through a drone; establish a digital twin model of the virtual system based on the collected transmission data corresponding to all areas in the actual system, and input the target transmission data as a model input parameter into the digital twin model to predict the location information of the target area; verify the digital twin model, and after the digital twin model is verified, output the location information of the target area based on the digital twin model; establish the target area in the actual system according to the location information of the target area.

[0165] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0166] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0167] Of course, a computing device may also include other components, such as input / output interfaces, communication components, etc.

[0168] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0169] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0170] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0171] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The illustrated embodiment is a method for establishing a location based on data transmitted by a drone.

[0172] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0173] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0174] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0175] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for establishing a location based on data transmitted by an unmanned aerial vehicle, characterized in that: include: Collecting target transmission data of the actual system by using a drone, wherein the target transmission data includes the size, shape, material, and structure of the target area; Based on the collected transmission data corresponding to all areas in the actual system, a digital twin model of the virtual system is established, and the target transmission data is used as a model input parameter and input into the digital twin model to predict the location information of the target area; Verifying the digital twin model, and after the digital twin model is verified, outputting the location information of the target area based on the digital twin model; Establish a target area in the actual system according to the location information of the target area Before verifying the digital twin model, the method further includes: The transmission data corresponding to all areas in the actual system collected are calibrated with time nodes and organized into original time series data sets in chronological order; Each sequence in the original time series dataset is subjected to data augmentation processing such as random rotation, translation and scaling to obtain a series of new sequences to construct the target time series dataset; For each new sequence, divide it into multiple subsequences according to the time step; Take each subsequence as a sample and add it to the target time series dataset until new sequences and subsequences that meet the conditions are obtained to complete the construction of the target time series dataset; The data enhancement process of randomly rotating, translating and scaling each sequence in the original time series data set is performed to obtain a series of new sequences, including: For each sample, a rotation angle is randomly generated, and each sequence in the original time series data set is rotated along the center point by this angle to obtain a new sequence; the rotation angle can be randomly generated within a certain range, and the rotated sample can be expressed as: ； in, represents the coordinates of the sample before rotation, represents the coordinates of the sample after rotation, Represents a randomly generated rotation angle; For each sample, a translation vector is randomly generated, and each sequence in the original time series data set is translated along the vector to obtain a new sequence; the translation vector can be randomly generated within a certain range, and the translated sample can be expressed as: ； in, represents the coordinates of the sample before translation, represents the coordinates of the sample after translation, represents a randomly generated translation vector; For each sample, a scaling factor is randomly generated, and each sequence in the original time series data set is scaled along the center point to obtain a new sequence; the scaling factor can be randomly generated within a preset range, and the scaled sample can be expressed as: ； in, represents the coordinates of the sample before scaling, represents the coordinates of the sample after scaling, Represents a randomly generated scaling factor.

2. The method according to claim 1, characterized in that Based on the collected transmission data corresponding to all areas in the actual system, a digital twin model of the virtual system is established, including: Obtaining transmission data corresponding to all regions in the actual system, the transmission data including size, shape, material properties, and structural information of all regions in the actual system; A physical model is established based on the transmission data corresponding to all areas in the actual system, and the physical model is converted into a digital model using numerical methods or computer-aided engineering software. Then, based on the physical model and digital model of the actual system, a digital twin model of the virtual system is established.

3. The method according to claim 1, characterized in that The digital twin model is verified, including: The target time series dataset is validated to evaluate the comprehensive score of the digital twin model.

4. The method according to claim 3, characterized in that The validating the target time series data set to evaluate the comprehensive score of the digital twin model includes: Divide the target time series data set into several subsequences, each subsequence is used as a validation set, and the remaining subsequences are used as training sets; By formula: , calculate the mean absolute scaled error based on the historical errors of the training set; Among them, h represents the prediction step length, Indicates the actual value, represents the predicted value, T represents the index of the last training sample, represents the lag period; By formula: , calculating the symmetric mean absolute percentage error based on the historical errors of the training set; Among them, h represents the prediction step length, Indicates the actual value, represents the predicted value, and T represents the index of the last training sample; By formula: , dividing the training set into different categories and determining the balanced accuracy of the number of samples in different time series classifications; in, represents the number of categories, Indicates category The number of correctly classified samples in Indicates category The number of misclassified samples in Indicates non-category The number of correctly classified samples in Indicates non-category The number of misclassified samples in ; By formula: , dividing the training set into different categories and determining the influence of the number of samples in random classification; Among them, P o represents the classification accuracy of the observation, Represents the classification accuracy of random classification; By formula group: , evaluating the comprehensive score of the digital twin model according to the mean absolute scaled error, the symmetric mean absolute percentage error, the balanced accuracy of the number of samples in the different time series classifications, and the influence of the number of samples in the random classification; in, is the regression coefficient, is the error term, is the evaluation score, x1 is the mean absolute scaled error, x2 is the symmetric mean absolute percentage error, x3 is the balanced accuracy of the number of samples in the different time series classifications, and x4 is the influence of the number of samples in the random classification.

5. The method according to claim 1, characterized in that Also includes: A real-time monitoring device is set up to obtain the actual location information of the UAV in a timely manner, and compare it with the prediction results of the digital twin model to adjust the parameters and algorithms of the digital twin model. The process of adjusting the parameters and algorithms of the digital twin model at least includes: adjusting model hyperparameters, increasing the sample size, and processing sample outliers.

6. A location establishment system based on drone transmission data, characterized in that: include: A collection module, used to collect target transmission data of the actual system through a drone, wherein the target transmission data includes the size, shape, material, and structure of the target area; An input module is used to establish a digital twin model of the virtual system based on the collected transmission data corresponding to all areas in the actual system, and input the target transmission data as a model input parameter into the digital twin model to predict the location information of the target area; A verification output module, used to verify the digital twin model, and after the digital twin model is verified, output the location information of the target area based on the digital twin model; An establishing module, used for establishing a target area in the actual system according to the location information of the target area; Before verifying the digital twin model, the method further includes: The transmission data corresponding to all areas in the actual system collected are calibrated with time nodes and organized into original time series data sets in chronological order; Each sequence in the original time series dataset is subjected to data augmentation processing such as random rotation, translation and scaling to obtain a series of new sequences to construct the target time series dataset; For each new sequence, divide it into multiple subsequences according to the time step; Take each subsequence as a sample and add it to the target time series dataset until new sequences and subsequences that meet the conditions are obtained to complete the construction of the target time series dataset; The data enhancement process of randomly rotating, translating and scaling each sequence in the original time series data set is performed to obtain a series of new sequences, including: For each sample, a rotation angle is randomly generated, and each sequence in the original time series data set is rotated along the center point by this angle to obtain a new sequence; the rotation angle can be randomly generated within a certain range, and the rotated sample can be expressed as: ； in, represents the coordinates of the sample before rotation, represents the coordinates of the sample after rotation, Represents a randomly generated rotation angle; For each sample, a translation vector is randomly generated, and each sequence in the original time series data set is translated along the vector to obtain a new sequence; the translation vector can be randomly generated within a certain range, and the translated sample can be expressed as: ； in, represents the coordinates of the sample before translation, represents the coordinates of the sample after translation, represents a randomly generated translation vector; For each sample, a scaling factor is randomly generated, and each sequence in the original time series data set is scaled along the center point to obtain a new sequence; the scaling factor can be randomly generated within a preset range, and the scaled sample can be expressed as: ； in, represents the coordinates of the sample before scaling, represents the coordinates of the sample after scaling, Represents a randomly generated scaling factor.

7. A computing device, characterized in that It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement the location establishment method based on drone transmission data as described in any one of claims 1 to 5.

8. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, the location establishment method based on drone transmission data as described in any one of claims 1 to 5 is implemented.

Images