Unmanned aerial vehicle edge network efficient data collection method and device based on data importance

Abstract

The invention relates to the technical field of data processing, in particular to an unmanned aerial vehicle edge network efficient data collection method and device based on data importance. Comprises: determining a target area in which an unmanned aerial vehicle is to fly in a to-be-collected area; generating a target flight path for the unmanned aerial vehicle according to the target area; under the condition that a preset common area exists between any two adjacent target areas, the target flight path is updated according to the common area; controlling the unmanned aerial vehicle to fly according to the updated target flight path, and collecting data in all areas in the updated target flight path; data completion is carried out according to the data to obtain the overall data of the to-be-collected area, the length of the flight path of the unmanned aerial vehicle is reduced, the flight cost is reduced, and the data integrity is improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and device for efficiently collecting data in a drone edge network based on data importance. Background Art

[0002] With the development of the information industry, the Internet of Things will make the world networked and informationized, realize the interconnection and integration of the traditionally separated physical world and information space, and represent the development trend of the future network. With the explosive popularity of wireless mobile terminal devices, the Internet of Things will enter a new era of development by using these universal mobile devices to provide larger-scale, more complex, thorough and comprehensive perception services.

[0003] With the explosive growth of portable devices, mobile crowd sensing has become a promising model that recruits crowd workers to perform various sensing tasks at target times and locations. However, in some large-scale tasks, such as environmental monitoring, urban sensing, and spatial crowdsourcing, the large number of sensing areas and long-term continuous sensing requirements require traditional MCS to always cover the entire map, which consumes a lot of energy. At the same time, with the further increase in network coverage and the further improvement in data accuracy requirements, during the collection of measurement data, the data collection time also increases due to the long flight path of the drone, resulting in increased consumption, higher flight costs, and missing and incomplete data. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a method and device for efficient data collection in a drone edge network based on data importance, so as to solve the problems of high flight costs and incomplete data in the prior art.

[0005] In order to achieve the above-mentioned purpose, the first aspect of the present application provides a method for efficiently collecting data in a drone edge network based on data importance, the method comprising: Determine the target area for the drone to fly in the area to be collected; Generate a target flight path for the UAV based on the target area; In the case where there is a preset common area between any two adjacent target areas, updating the target flight path according to the common area; Control the UAV to fly according to the updated target flight path, and collect data in all areas of the updated target flight path; The data is completed according to the data to obtain the overall data of the area to be collected.

[0006] In an embodiment of the present application, when there is a preset common area between any two adjacent target areas, updating the target flight path according to the common area includes: when there is a common area between any two adjacent target areas, obtaining historical data of the common area, and taking the first target area passed by the target flight path as the first target area; obtaining historical data of the first target area to determine the first error of the first target area based on the historical data of the first target area; determining the second error of the common area based on the historical data of the common area; respectively determining the flight distance cost of the UAV from the starting position to the first target area and the common area; and updating the target flight path based on the first error, the second error, the flight distance cost of the first target area, and the flight distance cost of the common area.

[0007] In an embodiment of the present application, updating the target flight path according to the first error, the second error, the flight distance cost of the first target area, and the flight distance cost of the ordinary area includes: when the second error is less than or equal to the first error and the flight distance cost of the ordinary area is less than or equal to the flight distance cost of the first target area, replacing the first target area with the ordinary area, and updating the target flight path based on the updated target area.

[0008] In an embodiment of the present application, when there is a preset common area between any two adjacent target areas, updating the target flight path according to the common area includes: acquiring the flight distance from the drone to the target area in real time; when the flight distance is less than a preset distance threshold, determining the common area as a new target area, and adding the new target area between the two adjacent target areas, and updating the target flight path based on the updated target area.

[0009] In an embodiment of the present application, the data in the updated target flight path includes the longitude, latitude and collection time of each area in the target flight path, and the data completion is performed according to the data to obtain the overall data of the area to be collected, including: generating a sparse matrix according to the data; for each missing data in the sparse matrix, determining the distance between the data point of the collected data and the data point of the missing data; for each missing data in the sparse matrix, determining the predicted value of the missing data according to the distance and value of the data point of the collected data whose distance is less than a preset distance; and adding the predicted values ​​of all missing data to the completion matrix to obtain the overall data.

[0010] In the embodiment of the present application, for each missing data in the sparse matrix, determining the distance between the data point of the collected data and the data point of the missing data includes determining the distance according to formula (1): (1) in, , , Indicates distance, is the weight to measure the degree of spatial correlation and its value is , The data point representing the i-th missing data The longitude, The data point representing the i-th missing data The latitude, Indicates the data point where the drone collects the i-th missing data The collection time, Represents the jth data point of collected data The longitude, Represents the jth data point of collected data The latitude, Indicates the data point of the jth collected data collected by the drone The collection time.

[0011] In an embodiment of the present application, determining the target area for the drone to fly in the area to be collected includes: inputting all sub-areas in the area to be collected into a site selection model; and outputting the target areas in all sub-areas through the site selection model.

[0012] In the embodiment of the present application, the loss function of the site selection model is shown in formula (2): (2) in, represents the loss value of the location selection model, Represents the expectation of the site selection model under one selection, which refers to the selection action performed under state s a Entering a new state And get reward r, represents the reward decay coefficient and takes the value , Perform the selection action in state s a Rewards, New Status Next, perform the selected action The maximum value of the reward.

[0013] The second aspect of the present application provides an efficient data collection device for a drone edge network based on data importance, comprising: a memory configured to store instructions; The processor is configured to call instructions from the memory and implement the above-mentioned efficient data collection method for the drone edge network based on data importance when executing the instructions.

[0014] A third aspect of the present application provides a machine-readable storage medium having instructions stored thereon, the instructions being used to enable a machine to execute the above-mentioned method for efficiently collecting data in a drone edge network based on data importance.

[0015] Through the above technical scheme, the target area where the UAV is to fly in the area to be collected is determined; a target flight path for the UAV is generated according to the target area; when there is a preset common area between any two adjacent target areas, the target flight path is updated according to the common area; the UAV is controlled to fly according to the updated target flight path, and data in all areas in the updated target flight path are collected; data completion is performed based on the data to obtain the overall data of the area to be collected, thereby reducing the length of the UAV's flight trajectory, reducing flight costs, and improving data integrity. Other features and advantages of the embodiments of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present application, but do not constitute a limitation on the embodiments of the present application. In the accompanying drawings: Figure 1 A schematic diagram of a process flow of an efficient data collection method for a drone edge network based on data importance according to an embodiment of the present application is schematically shown; Figure 2 A first schematic diagram of a site selection model according to an embodiment of the present application is schematically shown; Figure 3 A second schematic diagram of a site selection model according to an embodiment of the present application is schematically shown; Figure 4 A first result curve diagram of a method for efficiently collecting data in a drone edge network based on data importance according to an embodiment of the present application is schematically shown; Figure 5 A second result curve diagram schematically illustrates a method for efficiently collecting data in a drone edge network based on data importance according to an embodiment of the present application; Figure 6 The internal structure diagram of a computer device according to an embodiment of the present application is schematically shown. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme 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. It should be understood that the specific implementation methods described herein are only used to illustrate and explain the embodiments of the present application, and are not used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0018] It should be noted that if the embodiments of the present application involve directional indications (such as up, down, left, right, front, back...), such directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0019] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0020] Figure 1 The flowchart of a method for efficiently collecting data in a drone edge network based on data importance according to an embodiment of the present application is schematically shown. Figure 1 As shown, an embodiment of the present application provides an efficient data collection method for a drone edge network based on data importance, which may include the following steps.

[0021] Step 101: Determine the target area for the drone to fly in the area to be collected.

[0022] The processor can determine a target area for the drone to fly in the area to be collected.

[0023] In an embodiment of the present application, determining the target area for the drone to fly in the area to be collected includes: inputting all sub-areas in the area to be collected into a site selection model; and outputting the target areas in all sub-areas through the site selection model.

[0024] The processor may determine a target area for the drone to fly in the area to be collected. Specifically, the processor may input all sub-areas in the area to be collected into the site selection model. The site selection model may output the target areas in all sub-areas.

[0025] In the embodiment of the present application, the loss function of the site selection model is shown in formula (2): (2) in, represents the loss value of the location selection model, Represents the expectation of the site selection model under one selection, which refers to the selection action performed under state s a Entering a new state And get reward r, represents the reward decay coefficient and takes the value , Perform the selection action in state s a Rewards, New Status Next, perform the selected action The maximum value of the reward.

[0026] In an optional embodiment, the processor may construct a site selection model based on reinforcement learning, for example, a Markov decision process may be modeled, such as Figure 2 As shown in Figure 1, the relationship between the state, action, and reward of Markov decision processes (MDP) is used to select actions by selecting β (ϵ-greedy). The action represents the next area where data needs to be collected. After executing the selected action, the state is changed to a new state and the corresponding reward is generated. The processor can generate a unit selection matrix based on the area selection records of the most recent j cycles, which is recorded as state ,in, Indicates that area m is not selected as the target area of ​​the drone in cycle j. Indicates that the mth region is selected as the target region of the drone in the jth cycle. Action represents the selection in the current spatiotemporal perception region. There are m regions in total, and there are m selections in each step. The action set can be expressed as A = {1, 2, ..., m}, where A = m represents the selection of the mth region as the target region. Reward is used to measure the completion accuracy of action a (a∈A) in state s. Regions are selected one by one in each mission cycle. When the number of selected regions N reaches , The upper limit of the region is set according to the actual situation. Otherwise reward , ,in, It is a cycle Time Zone Historical data in After data completion Time Zone Historical data in is the variance.

[0027] In an optional embodiment, the processor can collect two neural network combinations with exactly the same structure, Q-network (reinforcement learning), to represent the Q function in reinforcement learning, and use the historical deep learning algorithm to update the parameters of reinforcement learning to build a site selection model. Reinforcement learning calculates the reward for each possible action in site selection, that is, the reward for selecting each area as the target area, and then uses the greedy algorithm to select the action and then make a new selection, and calculate the corresponding reward. Figure 3 As shown in the figure, the site selection model includes two neural networks with the same structure, namely the evaluate network and the target network. The evaluate network and the target network have the same initialization parameters. The site selection model is used to select the action a with the highest reward in state s. Then, it executes the selected action a and uses the target network to calculate the new state Make every possible move Rewards Finally, the target network outputs To train the site selection model. Use the gradient descent algorithm to update the parameters of the evaluation network , ,in, Representation parameters gradient.

[0028] During the training process of the site selection model, the processor can use the latest overall data obtained each time to update the training data of the site selection model to keep the site selection model up to date. For example, the processor can add the data of the new mobile phone of the drone to the training data set, and then based on the value of the training data of the previous cycle and the average value of the newly collected data in the current cycle Zoom in on the training data, where , represents the newly collected data of the i-th target area in the j-th cycle, m is the m-th area, represents the region selection of the i-th region in the j-th period, Indicates that the i-th region is selected, Indicates that the i-th region is not selected. , Represents training data The historical data of the first n periods of the i-th target area, m is the m-th area. The enlarged training data , The value is 1. Indicates the region selection in the jth cycle.

[0029] Step 102: Generate a target flight path for the UAV according to the target area.

[0030] Step 103: When there is a preset common area between any two adjacent target areas, the target flight path is updated according to the common area.

[0031] After determining the target area, the processor can generate a target flight path for the drone based on the target area. After obtaining the target flight path of the drone, the processor can determine whether there is a preset common area between any two adjacent target areas, wherein the preset common area is the other area in the area to be collected except the target area. In the case that there is a preset common area between any two adjacent target areas, the processor can update the target flight path based on the common area.

[0032] In an embodiment of the present application, when there is a preset common area between any two adjacent target areas, updating the target flight path according to the common area includes: when there is a common area between any two adjacent target areas, obtaining historical data of the common area, and taking the first target area passed by the target flight path as the first target area; obtaining historical data of the first target area to determine the first error of the first target area based on the historical data of the first target area; determining the second error of the common area based on the historical data of the common area; respectively determining the flight distance cost of the UAV from the starting position to the first target area and the common area; and updating the target flight path based on the first error, the second error, the flight distance cost of the first target area, and the flight distance cost of the common area.

[0033] In the case where there is a preset common area between any two adjacent target areas, the processor can update the target flight path based on the common area. Specifically, the processor can determine whether there is a preset common area between any two adjacent target areas. In the case where there is a common area between any two adjacent target areas, the processor can obtain the historical data of the common area, and take the first target area passed by the target flight path as the first target area, that is, the processor can take the first target area passed by the target flight path in the two adjacent target areas as the first target area. The processor can obtain the historical data of the first target area to determine the first error of the first target area based on the historical data of the first target area. The processor can determine the second error of the common area based on the historical data of the common area. The processor can determine the flight distance cost of the drone from the starting position to the first target area and the common area respectively. In a specific embodiment, the processor can determine the flight distance cost of the drone from the starting position to the first target area and the common area based on the first target area. The first error is determined by historical data , where the first error can be a mean absolute error. The processor can be based on the general area The processor can determine the second error of the general area from the historical data of the drone from the starting position Fly to the first target area Flight distance cost , ,in, Indicates that the drone is moving from the starting position Fly to the first target area The path length, Indicates that the drone is from the first target area Fly to the second target area The path length, Indicates that the drone is from the i-1th target area Fly to the i-th target area The processor can determine the path length of the drone from the starting position Fly to normal area Flight distance cost , ,in, Indicates that the drone is moving from the starting position Fly to the first target area The path length, Indicates that the drone is from the first target area Fly to the second target area The path length, Indicates that the drone is moving from the j-1th target area Fly to General Area The path length.

[0034] After obtaining the first error, the second error, the flight distance cost of the first target area, and the flight distance cost of the general area, the processor may update the target flight path according to the first error, the second error, the flight distance cost of the first target area, and the flight distance cost of the general area.

[0035] In an embodiment of the present application, updating the target flight path according to the first error, the second error, the flight distance cost of the first target area, and the flight distance cost of the ordinary area includes: when the second error is less than or equal to the first error and the flight distance cost of the ordinary area is less than or equal to the flight distance cost of the first target area, replacing the first target area with the ordinary area, and updating the target flight path based on the updated target area.

[0036] The processor may update the target flight path according to the first error, the second error, the flight distance cost of the first target area, and the flight distance cost of the common area. Specifically, the processor may determine whether the second error is less than or equal to the first error, and whether the flight cost of the common area is less than or equal to the flight cost of the first target area. In the case where the second error is less than or equal to the first error and the flight distance cost of the common area is less than or equal to the flight distance cost of the first target area, the processor may replace the first target area with the common area, and update the target flight path based on the updated target area.

[0037] In an embodiment of the present application, when there is a preset common area between any two adjacent target areas, updating the target flight path according to the common area includes: acquiring the flight distance from the drone to the target area in real time; when the flight distance is less than a preset distance threshold, determining the common area as a new target area, and adding the new target area between the two adjacent target areas, and updating the target flight path based on the updated target area.

[0038] In the case where there is a preset common area between any two adjacent target areas, the processor can update the target flight path based on the common area. Specifically, the processor can obtain the flight distance from the drone to the target area in real time. After obtaining the flight distance, the processor can determine whether the flight distance is less than a preset distance threshold, where the preset distance threshold can be determined based on actual conditions. In the case where the flight distance is less than the preset distance threshold, the processor can determine that the common area is a new target area, and add the new target area between the two adjacent target areas, and update the target flight path based on the updated target area.

[0039] Step 104: Control the UAV to fly according to the updated target flight path, and collect data in all areas of the updated target flight path.

[0040] Step 105: completing the data according to the data to obtain the overall data of the area to be collected.

[0041] After updating the target flight path, the processor can control the UAV to fly according to the updated target flight path and collect data in all areas of the updated target flight path. After obtaining the data in all areas of the updated target flight path, the processor can complete the data according to the data to obtain the overall data of the area to be collected.

[0042] In an embodiment of the present application, the data in the updated target flight path includes the longitude, latitude and collection time of each area in the target flight path, and the data completion is performed according to the data to obtain the overall data of the area to be collected, including: generating a sparse matrix according to the data; for each missing data in the sparse matrix, determining the distance between the data point of the collected data and the data point of the missing data; for each missing data in the sparse matrix, determining the predicted value of the missing data according to the distance and value of the data point of the collected data whose distance is less than a preset distance; and adding the predicted values ​​of all missing data to the completion matrix to obtain the overall data.

[0043] The data in the updated target flight path includes the longitude, latitude and collection time of each area in the target flight path, and the processor can complete the data according to the data to obtain the overall data of the area to be collected. Specifically, the processor can generate a sparse matrix according to the data. After obtaining the sparse matrix, for each missing data in the sparse matrix, the processor can determine the distance between the data point of the collected data and the data point of the missing data. In an embodiment of the present application, for each missing data in the sparse matrix, determining the distance between the data point of the collected data and the data point of the missing data includes determining the distance according to formula (1): (1) in, , , Indicates distance, is the weight to measure the degree of spatial correlation and its value is , The data point representing the i-th missing data Longitude, The data point representing the i-th missing data The latitude, Indicates the data point where the drone collects the i-th missing data The collection time, Represents the jth data point of collected data The longitude, Represents the jth data point of collected data The latitude, Indicates the data point of the jth collected data collected by the drone The collection time.

[0044] For each missing data in the sparse matrix, the processor can arrange all the data points of the collected data in ascending order according to the distance to obtain at least one data point of the collected data that is closest to the missing data. The processor can determine the predicted value of the missing data based on the distance and value of the data point of the collected data whose distance is less than the preset distance. In a specific embodiment, the processor can obtain the k data points of the collected data that are closest to the missing data, and put the distances of the k data points of the collected data into the set D, that is, ,in Indicates the distance to the first data point that has been collected. represents the distance of the kth collected data point. The processor can put the values ​​of the k collected data points into the set V, that is, ,in Indicates the value of the first data point that has been collected. The value of the kth data point of the collected data is represented, and the weighted average of the set D and the set V is taken as the predicted value of the missing data. After obtaining the predicted value of each missing data, the processor can add the predicted values ​​of all the missing data to the completion matrix to obtain the overall data.

[0045] For example, the overall data can be the temperature, PM2.5, PM10, and CO in the area to be collected. Figure 4 As shown, Figure 4 The graphs are comparisons of the average absolute error of the overall data collected by the drone in the method of the present invention and the average absolute error of the overall data collected by the existing QBC (query based on criteria) method, wherein (a) is a comparison graph of the overall data being temperature, (b) is a comparison graph of the overall data being PM2.5, (c) is a comparison graph of the overall data being PM10, and (d) is a comparison graph of the overall data being CO. The average absolute error of the temperature, PM2.5, PM10, and CO collected by the drone in the method of the present invention strictly decreases with the increase of the sampling rate, and when the sampling rate is 40%, the average absolute errors of the four collection tasks are reduced by 5.38%, 9.04%, 10.92%, and 7.74%, respectively. Figure 5 As shown, Figure 5The figure shows a comparison curve of the path length of the overall data of the area to be collected by the drone before optimization and the path length of the overall data of the area to be collected by the drone after optimization, where (a) is a comparison curve of the path length of the drone collecting temperature, (b) is a comparison curve of the path length of the drone collecting PM10, (c) is a comparison curve of the path length of the drone collecting PM2.5, and (d) is a comparison curve of the path length of the drone collecting CO. It can be seen that the path length is reduced when data is collected based on the updated target flight path.

[0046] Figure 1 FIG. 1 is a flow chart of an efficient data collection method for a drone edge network based on data importance in one embodiment. It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0047] The embodiment of the present application also provides a drone edge network efficient data collection device based on data importance, including: a memory configured to store instructions; The processor is configured to call instructions from the memory and implement the above-mentioned efficient data collection method for the drone edge network based on data importance when executing the instructions.

[0048] An embodiment of the present application also provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute the above-mentioned method for efficient data collection in a drone edge network based on data importance.

[0049] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 6As shown. The computer device includes a processor A01, a network interface A02, a memory (not shown in the figure) and a database (not shown in the figure) connected through a system bus. Among them, the processor A01 of the computer device is used to provide computing and control capabilities. The memory of the computer device includes an internal memory A03 and a non-volatile storage medium A04. The non-volatile storage medium A04 stores an operating system B01, a computer program B02 and a database (not shown in the figure). The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 in the non-volatile storage medium A04. The database of the computer device is used to store data such as the target flight path and overall data. The network interface A02 of the computer device is used to communicate with an external terminal through a network connection. When the computer program B02 is executed by the processor A01, an efficient data collection method for a drone edge network based on data importance is implemented.

[0050] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0051] An embodiment of the present application provides a device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, the following steps are implemented: determining a target area for a drone to fly in an area to be collected; generating a target flight path for the drone based on the target area; updating the target flight path based on the common area when there is a preset common area between any two adjacent target areas; controlling the drone to fly according to the updated target flight path, and collecting data in all areas in the updated target flight path; and completing data based on the data to obtain overall data of the area to be collected.

[0052] In one embodiment, when there is a preset common area between any two adjacent target areas, updating the target flight path according to the common area includes: when there is a common area between any two adjacent target areas, obtaining historical data of the common area, and taking the first target area passed by the target flight path as the first target area; obtaining historical data of the first target area to determine the first error of the first target area based on the historical data of the first target area; determining the second error of the common area based on the historical data of the common area; respectively determining the flight distance cost of the UAV from the starting position to the first target area and the common area; and updating the target flight path based on the first error, the second error, the flight distance cost of the first target area, and the flight distance cost of the common area.

[0053] In one embodiment, updating the target flight path according to the first error, the second error, the flight distance cost of the first target area, and the flight distance cost of the ordinary area includes: when the second error is less than or equal to the first error and the flight distance cost of the ordinary area is less than or equal to the flight distance cost of the first target area, replacing the first target area with the ordinary area, and updating the target flight path based on the updated target area.

[0054] In one embodiment, when there is a preset common area between any two adjacent target areas, updating the target flight path according to the common area includes: obtaining the flight distance from the UAV to the target area in real time; when the flight distance is less than a preset distance threshold, determining the common area as a new target area, adding the new target area between the two adjacent target areas, and updating the target flight path based on the updated target area.

[0055] In one embodiment, the data in the updated target flight path include the longitude, latitude and collection time of each area in the target flight path, and the data completion is performed according to the data to obtain the overall data of the area to be collected, including: generating a sparse matrix according to the data; for each missing data in the sparse matrix, determining the distance between the data point of the collected data and the data point of the missing data; for each missing data in the sparse matrix, determining the predicted value of the missing data according to the distance and value of the data point of the collected data whose distance is less than a preset distance; and adding the predicted values ​​of all missing data to the completion matrix to obtain the overall data.

[0056] In one embodiment, for each missing data in the sparse matrix, determining the distance between a data point of the acquired data and a data point of the missing data includes determining the distance according to formula (1): (1) in, , , Indicates distance, is the weight to measure the degree of spatial correlation and its value is , The data point representing the i-th missing data The longitude, The data point representing the i-th missing data The latitude, Indicates the data point where the drone collects the i-th missing data The collection time, Represents the jth data point of collected data The longitude, Represents the jth data point of collected data The latitude, Indicates the data point of the jth collected data collected by the drone The collection time.

[0057] In one embodiment, determining the target area in the area to be collected where the drone is to fly includes: inputting all sub-areas in the area to be collected into a site selection model; and outputting the target areas in all sub-areas through the site selection model.

[0058] In one embodiment, the loss function of the site selection model is shown in formula (2): (2) in, represents the loss value of the location selection model, Represents the expectation of the site selection model under one selection, which refers to the selection action performed under state s a Entering a new state And get reward r, represents the reward decay coefficient and takes the value , Perform the selection action in state s a Rewards, New Status Next, perform the selected action The maximum value of the reward.

[0059] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing the program steps of initializing a method for efficient data collection of a drone edge network based on data importance.

[0060] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0061] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0062] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0063] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0064] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0065] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0066] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0067] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0068] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. An efficient data collection method for drone edge network based on data importance, characterized in that: The method comprises: Determine a target area for the drone to fly in the area to be collected; generating a target flight path for the UAV according to the target area; In the case where there is a preset common area between any two adjacent target areas, updating the target flight path according to the common area; Controlling the UAV to fly according to the updated target flight path, and collecting data in all areas of the updated target flight path; Data completion is performed based on the data to obtain overall data of the area to be collected.

2. The method for efficient data collection in a drone edge network based on data importance according to claim 1 is characterized in that: In the case where there is a preset common area between any two adjacent target areas, updating the target flight path according to the common area includes: When there is a common area between any two adjacent target areas, historical data of the common area is obtained, and the first target area passed by the target flight path is taken as the first target area; Acquire historical data of the first target area to determine a first error of the first target area; determining a second error of the common area according to the historical data of the common area; Determine the flight distance cost of the UAV from the starting position to the first target area and the general area respectively; The target flight path is updated according to the first error, the second error, the flight distance cost of the first target area, and the flight distance cost of the general area.

3. The method for efficient data collection in a drone edge network based on data importance according to claim 2 is characterized in that: The updating of the target flight path according to the first error, the second error, the flight distance cost of the first target area, and the flight distance cost of the common area comprises: When the second error is less than or equal to the first error and the flight distance cost of the common area is less than or equal to the flight distance cost of the first target area, the common area is used to replace the first target area, and the target flight path is updated based on the updated target area.

4. The method for efficient data collection in a drone edge network based on data importance according to claim 1 is characterized in that: In the case where there is a preset common area between any two adjacent target areas, updating the target flight path according to the common area includes: Acquire the flight distance from the UAV to the target area in real time; When the flight distance is less than a preset distance threshold, the common area is determined to be a new target area, and the new target area is added between the two adjacent target areas, and the target flight path is updated based on the updated target area.

5. The method for efficient data collection in a drone edge network based on data importance according to claim 1 is characterized in that: The data in all areas of the updated target flight path include the longitude and latitude of each area in the target flight path and the collection time of the drone, and the data completion based on the data to obtain the overall data of the area to be collected includes: generating a sparse matrix according to the data; For each missing data in the sparse matrix, determining the distance between a data point of the collected data and a data point of the missing data; For each missing data in the sparse matrix, determining a predicted value of the missing data according to a distance and a value of a data point of the collected data whose distance is less than a preset distance; The predicted values ​​of all missing data are added to the completion matrix to obtain the overall data.

6. The method for efficient data collection in a drone edge network based on data importance according to claim 5 is characterized in that: For each missing data in the sparse matrix, determining the distance between the data point of the collected data and the data point of the missing data includes determining the distance according to formula (1): （1） in, , , represents the distance, is the weight to measure the degree of spatial correlation and its value is , The data point representing the i-th missing data The longitude, The data point representing the i-th missing data The latitude, Indicates the data point where the drone collects the i-th missing data The collection time, Represents the jth data point of collected data The longitude, Represents the jth data point of collected data The latitude, Indicates the data point of the jth collected data collected by the drone The collection time.

7. The method for efficient data collection in a drone edge network based on data importance according to claim 1, characterized in that: Determining the target area to be flown by the drone in the area to be collected includes: Inputting all sub-areas in the area to be collected into the site selection model; The target area in all the sub-areas is outputted through the site selection model.

8. The method for efficient data collection in a drone edge network based on data importance according to claim 7 is characterized in that: The loss function of the site selection model is shown in formula (2): （2） in, represents the loss value of the location selection model, represents the expectation of the site selection model under one selection, which refers to the selection action performed under state s a Entering a new state And get reward r, represents the reward decay coefficient and takes the value , Perform the selection action in state s a Rewards, New Status Next, perform the selected action The maximum value of the reward.

9. An efficient data collection device for drone edge network based on data importance, characterized in that: include: a memory configured to store instructions; A processor is configured to call the instructions from the memory and to implement the efficient data collection method for a drone edge network based on data importance according to any one of claims 1 to 8 when executing the instructions.

10. A machine-readable storage medium, characterized in that: The machine-readable storage medium stores instructions for enabling a machine to execute an efficient data collection method for a drone edge network based on data importance according to any one of claims 1 to 8.

Images